Sixth Edition, last update July 25, 2007
2
Lessons In Electric Circuits, Volume II – AC
By Tony R. Kuphaldt
Sixth Edition, last update July 25, 2007
i
c©20002011, Tony R. Kuphaldt
This book is published under the terms and conditions of the Design Science License. These
terms and conditions allow for free copying, distribution, and/or modification of this document
by the general public. The full Design Science License text is included in the last chapter.
As an open and collaboratively developed text, this book is distributed in the hope that
it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the Design Science
License for more details.
Available in its entirety as part of the Open Book Project collection at:
openbookproject.net/electricCircuits
PRINTING HISTORY
• First Edition: Printed in June of 2000. PlainASCII illustrations for universal computer
readability.
• Second Edition: Printed in September of 2000. Illustrations reworked in standard graphic
(eps and jpeg) format. Source files translated to Texinfo format for easy online and printed
publication.
• Third Edition: Equations and tables reworked as graphic images rather than plainASCII
text.
• Fourth Edition: Printed in November 2001. Source files translated to SubML format.
SubML is a simple markup language designed to easily convert to other markups like
LATEX, HTML, or DocBook using nothing but searchandreplace substitutions.
• Fifth Edition: Printed in November 2002. New sections added, and error corrections
made, since the fourth edition.
• Sixth Edition: Printed in June 2006. Added CH 13, sections added, and error corrections
made, figure numbering and captions added, since the fifth edition.
ii
Contents
1 BASIC AC THEORY 1
1.1 What is alternating current (AC)? . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 AC waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3 Measurements of AC magnitude . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.4 Simple AC circuit calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.5 AC phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
1.6 Principles of radio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.7 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2 COMPLEX NUMBERS 27
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.2 Vectors and AC waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.3 Simple vector addition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.4 Complex vector addition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.5 Polar and rectangular notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.6 Complex number arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
2.7 More on AC ”polarity” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
2.8 Some examples with AC circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
2.9 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
3 REACTANCE AND IMPEDANCE – INDUCTIVE 57
3.1 AC resistor circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
3.2 AC inductor circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.3 Series resistorinductor circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
3.4 Parallel resistorinductor circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
3.5 Inductor quirks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.6 More on the “skin effect” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
3.7 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
4 REACTANCE AND IMPEDANCE – CAPACITIVE 81
4.1 AC resistor circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
4.2 AC capacitor circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.3 Series resistorcapacitor circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
4.4 Parallel resistorcapacitor circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
iii
iv CONTENTS
4.5 Capacitor quirks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
4.6 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
5 REACTANCE AND IMPEDANCE – R, L, AND C 99
5.1 Review of R, X, and Z . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
5.2 Series R, L, and C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
5.3 Parallel R, L, and C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
5.4 Seriesparallel R, L, and C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
5.5 Susceptance and Admittance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
5.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
5.7 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
6 RESONANCE 121
6.1 An electric pendulum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
6.2 Simple parallel (tank circuit) resonance . . . . . . . . . . . . . . . . . . . . . . . . 126
6.3 Simple series resonance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
6.4 Applications of resonance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
6.5 Resonance in seriesparallel circuits . . . . . . . . . . . . . . . . . . . . . . . . . . 136
6.6 Q and bandwidth of a resonant circuit . . . . . . . . . . . . . . . . . . . . . . . . 145
6.7 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
7 MIXEDFREQUENCY AC SIGNALS 153
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
7.2 Square wave signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
7.3 Other waveshapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
7.4 More on spectrum analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
7.5 Circuit effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
7.6 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
8 FILTERS 189
8.1 What is a filter? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
8.2 Lowpass filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
8.3 Highpass filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
8.4 Bandpass filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
8.5 Bandstop filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
8.6 Resonant filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
8.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
8.8 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
9 TRANSFORMERS 217
9.1 Mutual inductance and basic operation . . . . . . . . . . . . . . . . . . . . . . . . 218
9.2 Stepup and stepdown transformers . . . . . . . . . . . . . . . . . . . . . . . . . 232
9.3 Electrical isolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
9.4 Phasing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
9.5 Winding configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
9.6 Voltage regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248
CONTENTS v
9.7 Special transformers and applications . . . . . . . . . . . . . . . . . . . . . . . . . 251
9.8 Practical considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
9.9 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
10 POLYPHASE AC CIRCUITS 283
10.1 Singlephase power systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
10.2 Threephase power systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
10.3 Phase rotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
10.4 Polyphase motor design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
10.5 Threephase Y and Delta configurations . . . . . . . . . . . . . . . . . . . . . . . . 306
10.6 Threephase transformer circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
10.7 Harmonics in polyphase power systems . . . . . . . . . . . . . . . . . . . . . . . . 318
10.8 Harmonic phase sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
10.9 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345
11 POWER FACTOR 347
11.1 Power in resistive and reactive AC circuits . . . . . . . . . . . . . . . . . . . . . . 347
11.2 True, Reactive, and Apparent power . . . . . . . . . . . . . . . . . . . . . . . . . . 352
11.3 Calculating power factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
11.4 Practical power factor correction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360
11.5 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365
12 AC METERING CIRCUITS 367
12.1 AC voltmeters and ammeters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367
12.2 Frequency and phase measurement . . . . . . . . . . . . . . . . . . . . . . . . . . 374
12.3 Power measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382
12.4 Power quality measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
12.5 AC bridge circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
12.6 AC instrumentation transducers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396
12.7 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 406
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 406
13 AC MOTORS 407
13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 408
13.2 Synchronous Motors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412
13.3 Synchronous condenser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 420
13.4 Reluctance motor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
13.5 Stepper motors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 426
13.6 Brushless DC motor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438
13.7 Tesla polyphase induction motors . . . . . . . . . . . . . . . . . . . . . . . . . . . 442
13.8 Wound rotor induction motors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460
13.9 Singlephase induction motors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464
13.10 Other specialized motors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469
13.11 Selsyn (synchro) motors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471
13.12 AC commutator motors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479
vi CONTENTS
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 482
14 TRANSMISSION LINES 483
14.1 A 50ohm cable? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483
14.2 Circuits and the speed of light . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 484
14.3 Characteristic impedance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 486
14.4 Finitelength transmission lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493
14.5 “Long” and “short” transmission lines . . . . . . . . . . . . . . . . . . . . . . . . . 499
14.6 Standing waves and resonance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 502
14.7 Impedance transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 522
14.8 Waveguides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 529
A1 ABOUT THIS BOOK 537
A2 CONTRIBUTOR LIST 541
A3 DESIGN SCIENCE LICENSE 549
INDEX 552
Chapter 1
BASIC AC THEORY
Contents
1.1 What is alternating current (AC)? . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 AC waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3 Measurements of AC magnitude . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.4 Simple AC circuit calculations . . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.5 AC phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
1.6 Principles of radio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.7 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
1.1 What is alternating current (AC)?
Most students of electricity begin their study with what is known as direct current (DC), which
is electricity flowing in a constant direction, and/or possessing a voltage with constant polarity.
DC is the kind of electricity made by a battery (with definite positive and negative terminals),
or the kind of charge generated by rubbing certain types of materials against each other.
As useful and as easy to understand as DC is, it is not the only “kind” of electricity in use.
Certain sources of electricity (most notably, rotary electromechanical generators) naturally
produce voltages alternating in polarity, reversing positive and negative over time. Either as
a voltage switching polarity or as a current switching direction back and forth, this “kind” of
electricity is known as Alternating Current (AC): Figure 1.1
Whereas the familiar battery symbol is used as a generic symbol for any DC voltage source,
the circle with the wavy line inside is the generic symbol for any AC voltage source.
One might wonder why anyone would bother with such a thing as AC. It is true that in
some cases AC holds no practical advantage over DC. In applications where electricity is used
to dissipate energy in the form of heat, the polarity or direction of current is irrelevant, so
long as there is enough voltage and current to the load to produce the desired heat (power
dissipation). However, with AC it is possible to build electric generators, motors and power
1
2 CHAPTER 1. BASIC AC THEORY
I
I
DIRECT CURRENT
(DC)
ALTERNATING CURRENT
(AC)
I
I
Figure 1.1: Direct vs alternating current
distribution systems that are far more efficient than DC, and so we find AC used predominately
across the world in high power applications. To explain the details of why this is so, a bit of
background knowledge about AC is necessary.
If a machine is constructed to rotate a magnetic field around a set of stationary wire coils
with the turning of a shaft, AC voltage will be produced across the wire coils as that shaft
is rotated, in accordance with Faraday’s Law of electromagnetic induction. This is the basic
operating principle of an AC generator, also known as an alternator: Figure 1.2
N S
+ 
Load
II
N
S
Load
no current!
no current!
Load
N
S
N
Load
S
 +
I I
Step #1 Step #2
Step #3 Step #4
Figure 1.2: Alternator operation
1.1. WHAT IS ALTERNATING CURRENT (AC)? 3
Notice how the polarity of the voltage across the wire coils reverses as the opposite poles of
the rotating magnet pass by. Connected to a load, this reversing voltage polarity will create a
reversing current direction in the circuit. The faster the alternator’s shaft is turned, the faster
the magnet will spin, resulting in an alternating voltage and current that switches directions
more often in a given amount of time.
While DC generators work on the same general principle of electromagnetic induction, their
construction is not as simple as their AC counterparts. With a DC generator, the coil of wire
is mounted in the shaft where the magnet is on the AC alternator, and electrical connections
are made to this spinning coil via stationary carbon “brushes” contacting copper strips on the
rotating shaft. All this is necessary to switch the coil’s changing output polarity to the external
circuit so the external circuit sees a constant polarity: Figure 1.3
Load
N S N S
 +
+
I
N S SN
Load
Step #1 Step #2
N S SN
Load
N S
Load
SN


I
+
+
Step #3 Step #4
Figure 1.3: DC generator operation
The generator shown above will produce two pulses of voltage per revolution of the shaft,
both pulses in the same direction (polarity). In order for a DC generator to produce constant
voltage, rather than brief pulses of voltage once every 1/2 revolution, there are multiple sets
of coils making intermittent contact with the brushes. The diagram shown above is a bit more
simplified than what you would see in real life.
The problems involved with making and breaking electrical contact with a moving coil
should be obvious (sparking and heat), especially if the shaft of the generator is revolving
at high speed. If the atmosphere surrounding the machine contains flammable or explosive
4 CHAPTER 1. BASIC AC THEORY
vapors, the practical problems of sparkproducing brush contacts are even greater. An AC gen
erator (alternator) does not require brushes and commutators to work, and so is immune to
these problems experienced by DC generators.
The benefits of AC over DC with regard to generator design is also reflected in electric
motors. While DC motors require the use of brushes to make electrical contact with moving
coils of wire, AC motors do not. In fact, AC and DC motor designs are very similar to their
generator counterparts (identical for the sake of this tutorial), the AC motor being dependent
upon the reversing magnetic field produced by alternating current through its stationary coils
of wire to rotate the rotating magnet around on its shaft, and the DC motor being dependent on
the brush contacts making and breaking connections to reverse current through the rotating
coil every 1/2 rotation (180 degrees).
So we know that AC generators and AC motors tend to be simpler than DC generators
and DC motors. This relative simplicity translates into greater reliability and lower cost of
manufacture. But what else is AC good for? Surely there must be more to it than design details
of generators and motors! Indeed there is. There is an effect of electromagnetism known as
mutual induction, whereby two or more coils of wire placed so that the changing magnetic field
created by one induces a voltage in the other. If we have two mutually inductive coils and we
energize one coil with AC, we will create an AC voltage in the other coil. When used as such,
this device is known as a transformer: Figure 1.4
Transformer
AC
voltage
source
Induced AC
voltage
Figure 1.4: Transformer “transforms” AC voltage and current.
The fundamental significance of a transformer is its ability to step voltage up or down from
the powered coil to the unpowered coil. The AC voltage induced in the unpowered (“secondary”)
coil is equal to the AC voltage across the powered (“primary”) coil multiplied by the ratio of
secondary coil turns to primary coil turns. If the secondary coil is powering a load, the current
through the secondary coil is just the opposite: primary coil current multiplied by the ratio
of primary to secondary turns. This relationship has a very close mechanical analogy, using
torque and speed to represent voltage and current, respectively: Figure 1.5
If the winding ratio is reversed so that the primary coil has less turns than the secondary
coil, the transformer “steps up” the voltage from the source level to a higher level at the load:
Figure 1.6
The transformer’s ability to step AC voltage up or down with ease gives AC an advantage
unmatched by DC in the realm of power distribution in figure 1.7. When transmitting electrical
power over long distances, it is far more efficient to do so with steppedup voltages and stepped
down currents (smallerdiameter wire with less resistive power losses), then step the voltage
back down and the current back up for industry, business, or consumer use.
Transformer technology has made longrange electric power distribution practical. Without
1.1. WHAT IS ALTERNATING CURRENT (AC)? 5
+ +
Large gear
Small gear
(many teeth)
(few teeth)
AC
voltage
source Load
high voltage
low current
low voltage
high current
many
turns few turns
Speed multiplication geartrain
"Stepdown" transformer
high torque
low speed
low torque
high speed
Figure 1.5: Speed multiplication gear train steps torque down and speed up. Stepdown trans
former steps voltage down and current up.
++
Large gear
Small gear
(many teeth)
(few teeth)
AC
voltage
source
Load
low voltage
high current
high voltage
low current
few turns many turns
Speed reduction geartrain "Stepup" transformer
low torque
high speed
high torque
low speed
Figure 1.6: Speed reduction gear train steps torque up and speed down. Stepup transformer
steps voltage up and current down.
Stepup
Stepdown
Power Plant
low voltage
high voltage
low voltage
. . . to other customers
Home or
Business
Figure 1.7: Transformers enable efficient long distance high voltage transmission of electric
energy.
6 CHAPTER 1. BASIC AC THEORY
the ability to efficiently step voltage up and down, it would be costprohibitive to construct
power systems for anything but closerange (within a few miles at most) use.
As useful as transformers are, they only work with AC, not DC. Because the phenomenon of
mutual inductance relies on changingmagnetic fields, and direct current (DC) can only produce
steady magnetic fields, transformers simply will not work with direct current. Of course, direct
current may be interrupted (pulsed) through the primary winding of a transformer to create
a changing magnetic field (as is done in automotive ignition systems to produce highvoltage
spark plug power from a lowvoltage DC battery), but pulsed DC is not that different from
AC. Perhaps more than any other reason, this is why AC finds such widespread application in
power systems.
• REVIEW:
• DC stands for “Direct Current,” meaning voltage or current that maintains constant po
larity or direction, respectively, over time.
• AC stands for “Alternating Current,” meaning voltage or current that changes polarity or
direction, respectively, over time.
• AC electromechanical generators, known as alternators, are of simpler construction than
DC electromechanical generators.
• AC and DC motor design follows respective generator design principles very closely.
• A transformer is a pair of mutuallyinductive coils used to convey AC power from one coil
to the other. Often, the number of turns in each coil is set to create a voltage increase or
decrease from the powered (primary) coil to the unpowered (secondary) coil.
• Secondary voltage = Primary voltage (secondary turns / primary turns)
• Secondary current = Primary current (primary turns / secondary turns)
1.2 AC waveforms
When an alternator produces AC voltage, the voltage switches polarity over time, but does
so in a very particular manner. When graphed over time, the “wave” traced by this voltage
of alternating polarity from an alternator takes on a distinct shape, known as a sine wave:
Figure 1.8
In the voltage plot from an electromechanical alternator, the change from one polarity to
the other is a smooth one, the voltage level changing most rapidly at the zero (“crossover”)
point and most slowly at its peak. If we were to graph the trigonometric function of “sine” over
a horizontal range of 0 to 360 degrees, we would find the exact same pattern as in Table 1.1.
The reason why an electromechanical alternator outputs sinewave AC is due to the physics
of its operation. The voltage produced by the stationary coils by the motion of the rotating
magnet is proportional to the rate at which the magnetic flux is changing perpendicular to the
coils (Faraday’s Law of Electromagnetic Induction). That rate is greatest when the magnet
poles are closest to the coils, and least when the magnet poles are furthest away from the coils.
1.2. AC WAVEFORMS 7
+

Time
(the sine wave)
Figure 1.8: Graph of AC voltage over time (the sine wave).
Table 1.1: Trigonometric “sine” function.
Angle (o) sin(angle) wave Angle (o) sin(angle) wave
0 0.0000 zero 180 0.0000 zero
15 0.2588 + 195 0.2588 
30 0.5000 + 210 0.5000 
45 0.7071 + 225 0.7071 
60 0.8660 + 240 0.8660 
75 0.9659 + 255 0.9659 
90 1.0000 +peak 270 1.0000 peak
105 0.9659 + 285 0.9659 
120 0.8660 + 300 0.8660 
135 0.7071 + 315 0.7071 
150 0.5000 + 330 0.5000 
165 0.2588 + 345 0.2588 
180 0.0000 zero 360 0.0000 zero
8 CHAPTER 1. BASIC AC THEORY
Mathematically, the rate of magnetic flux change due to a rotating magnet follows that of a
sine function, so the voltage produced by the coils follows that same function.
If we were to follow the changing voltage produced by a coil in an alternator from any
point on the sine wave graph to that point when the wave shape begins to repeat itself, we
would have marked exactly one cycle of that wave. This is most easily shown by spanning the
distance between identical peaks, but may be measured between any corresponding points on
the graph. The degree marks on the horizontal axis of the graph represent the domain of the
trigonometric sine function, and also the angular position of our simple twopole alternator
shaft as it rotates: Figure 1.9
one wave cycle
Alternator shaft
position (degrees)
0 90 180 270 360(0) 90 180 270 360(0)
one wave cycle
Figure 1.9: Alternator voltage as function of shaft position (time).
Since the horizontal axis of this graph can mark the passage of time as well as shaft position
in degrees, the dimension marked for one cycle is often measured in a unit of time, most often
seconds or fractions of a second. When expressed as a measurement, this is often called the
period of a wave. The period of a wave in degrees is always 360, but the amount of time one
period occupies depends on the rate voltage oscillates back and forth.
A more popular measure for describing the alternating rate of an AC voltage or current
wave than period is the rate of that backandforth oscillation. This is called frequency. The
modern unit for frequency is the Hertz (abbreviated Hz), which represents the number of wave
cycles completed during one second of time. In the United States of America, the standard
powerline frequency is 60 Hz, meaning that the AC voltage oscillates at a rate of 60 complete
backandforth cycles every second. In Europe, where the power system frequency is 50 Hz,
the AC voltage only completes 50 cycles every second. A radio station transmitter broadcasting
at a frequency of 100 MHz generates an AC voltage oscillating at a rate of 100 million cycles
every second.
Prior to the canonization of the Hertz unit, frequency was simply expressed as “cycles per
second.” Older meters and electronic equipment often bore frequency units of “CPS” (Cycles
Per Second) instead of Hz. Many people believe the change from selfexplanatory units like
CPS to Hertz constitutes a step backward in clarity. A similar change occurred when the unit
of “Celsius” replaced that of “Centigrade” for metric temperature measurement. The name
Centigrade was based on a 100count (“Centi”) scale (“grade”) representing the melting and
boiling points of H2O, respectively. The name Celsius, on the other hand, gives no hint as to
the unit’s origin or meaning.
1.2. AC WAVEFORMS 9
Period and frequency are mathematical reciprocals of one another. That is to say, if a wave
has a period of 10 seconds, its frequency will be 0.1 Hz, or 1/10 of a cycle per second:
Frequency in Hertz = 1
Period in seconds
An instrument called an oscilloscope, Figure 1.10, is used to display a changing voltage over
time on a graphical screen. You may be familiar with the appearance of an ECG or EKG (elec
trocardiograph) machine, used by physicians to graph the oscillations of a patient’s heart over
time. The ECG is a specialpurpose oscilloscope expressly designed for medical use. General
purpose oscilloscopes have the ability to display voltage from virtually any voltage source,
plotted as a graph with time as the independent variable. The relationship between period
and frequency is very useful to know when displaying an AC voltage or current waveform on
an oscilloscope screen. By measuring the period of the wave on the horizontal axis of the oscil
loscope screen and reciprocating that time value (in seconds), you can determine the frequency
in Hertz.
trigger
timebase
s/div DC GND AC
X
GNDDC
V/div
vertical
OSCILLOSCOPE
Y
AC
1m
16 divisions
@ 1ms/div =
a period of 16 ms
Frequency =
period
1 1
16 ms
= = 62.5 Hz
Figure 1.10: Time period of sinewave is shown on oscilloscope.
Voltage and current are by no means the only physical variables subject to variation over
time. Much more common to our everyday experience is sound, which is nothing more than the
alternating compression and decompression (pressure waves) of air molecules, interpreted by
our ears as a physical sensation. Because alternating current is a wave phenomenon, it shares
many of the properties of other wave phenomena, like sound. For this reason, sound (especially
structured music) provides an excellent analogy for relating AC concepts.
In musical terms, frequency is equivalent to pitch. Lowpitch notes such as those produced
by a tuba or bassoon consist of air molecule vibrations that are relatively slow (low frequency).
10 CHAPTER 1. BASIC AC THEORY
Highpitch notes such as those produced by a flute or whistle consist of the same type of vibra
tions in the air, only vibrating at a much faster rate (higher frequency). Figure 1.11 is a table
showing the actual frequencies for a range of common musical notes.
C (middle)
Note Musical designation
C
C sharp (or D flat) C# or Db
D D
D sharp (or E flat) D# or Eb
E E
F F
F sharp (or G flat) F# or Gb
G G
G sharp (or A flat) G# or Ab
A A
A sharp (or B flat) A# or Bb
B B
C
B
A sharp (or B flat) A# or Bb
A A1 220.00
440.00
261.63
Frequency (in hertz)
B1
C1
293.66
233.08
246.94
277.18
311.13
329.63
349.23
369.99
392.00
415.30
466.16
493.88
523.25
Figure 1.11: The frequency in Hertz (Hz) is shown for various musical notes.
Astute observers will notice that all notes on the table bearing the same letter designation
are related by a frequency ratio of 2:1. For example, the first frequency shown (designated with
the letter “A”) is 220 Hz. The next highest “A” note has a frequency of 440 Hz – exactly twice as
many sound wave cycles per second. The same 2:1 ratio holds true for the first A sharp (233.08
Hz) and the next A sharp (466.16 Hz), and for all note pairs found in the table.
Audibly, two notes whose frequencies are exactly double each other sound remarkably sim
ilar. This similarity in sound is musically recognized, the shortest span on a musical scale
separating such note pairs being called an octave. Following this rule, the next highest “A”
note (one octave above 440 Hz) will be 880 Hz, the next lowest “A” (one octave below 220 Hz)
will be 110 Hz. A view of a piano keyboard helps to put this scale into perspective: Figure 1.12
As you can see, one octave is equal to seven white keys’ worth of distance on a piano key
board. The familiar musical mnemonic (doeraymeefahsolahtee) – yes, the same pattern
immortalized in the whimsical Rodgers and Hammerstein song sung in The Sound of Music –
covers one octave from C to C.
While electromechanical alternators and many other physical phenomena naturally pro
duce sine waves, this is not the only kind of alternating wave in existence. Other “waveforms”
of AC are commonly produced within electronic circuitry. Here are but a few sample waveforms
and their common designations in figure 1.13
1.2. AC WAVEFORMS 11
C D E F G A B C D E F G A BC D E F G A B
C#
Db
D#
Eb
F#
Gb
G#
Ab
A#
Bb
C#
Db
D#
Eb
F#
Gb
G#
Ab
A#
Bb
C#
Db
D#
Eb
F#
Gb
G#
Ab
A#
Bb
one octave
Figure 1.12: An octave is shown on a musical keyboard.
Square wave Triangle wave
Sawtooth wave
one wave cycle one wave cycle
Figure 1.13: Some common waveshapes (waveforms).
12 CHAPTER 1. BASIC AC THEORY
These waveforms are by no means the only kinds of waveforms in existence. They’re simply
a few that are common enough to have been given distinct names. Even in circuits that are
supposed to manifest “pure” sine, square, triangle, or sawtooth voltage/current waveforms, the
reallife result is often a distorted version of the intended waveshape. Some waveforms are
so complex that they defy classification as a particular “type” (including waveforms associated
with many kinds of musical instruments). Generally speaking, any waveshape bearing close
resemblance to a perfect sine wave is termed sinusoidal, anything different being labeled as
nonsinusoidal. Being that the waveform of an AC voltage or current is crucial to its impact in
a circuit, we need to be aware of the fact that AC waves come in a variety of shapes.
• REVIEW:
• AC produced by an electromechanical alternator follows the graphical shape of a sine
wave.
• One cycle of a wave is one complete evolution of its shape until the point that it is ready
to repeat itself.
• The period of a wave is the amount of time it takes to complete one cycle.
• Frequency is the number of complete cycles that a wave completes in a given amount of
time. Usually measured in Hertz (Hz), 1 Hz being equal to one complete wave cycle per
second.
• Frequency = 1/(period in seconds)
1.3 Measurements of AC magnitude
So far we know that AC voltage alternates in polarity and AC current alternates in direction.
We also know that AC can alternate in a variety of different ways, and by tracing the alter
nation over time we can plot it as a “waveform.” We can measure the rate of alternation by
measuring the time it takes for a wave to evolve before it repeats itself (the “period”), and
express this as cycles per unit time, or “frequency.” In music, frequency is the same as pitch,
which is the essential property distinguishing one note from another.
However, we encounter a measurement problem if we try to express how large or small an
AC quantity is. With DC, where quantities of voltage and current are generally stable, we have
little trouble expressing how much voltage or current we have in any part of a circuit. But how
do you grant a single measurement of magnitude to something that is constantly changing?
One way to express the intensity, or magnitude (also called the amplitude), of an AC quan
tity is to measure its peak height on a waveform graph. This is known as the peak or crest
value of an AC waveform: Figure 1.14
Another way is to measure the total height between opposite peaks. This is known as the
peaktopeak (PP) value of an AC waveform: Figure 1.15
Unfortunately, either one of these expressions of waveform amplitude can be misleading
when comparing two different types of waves. For example, a square wave peaking at 10 volts
is obviously a greater amount of voltage for a greater amount of time than a triangle wave
1.3. MEASUREMENTS OF AC MAGNITUDE 13
Time
Peak
Figure 1.14: Peak voltage of a waveform.
Time
PeaktoPeak
Figure 1.15: Peaktopeak voltage of a waveform.
Time
10 V
10 V(peak)
10 V(peak)
more heat energy
dissipated dissipated
less heat energy
(same load resistance)
Figure 1.16: A square wave produces a greater heating effect than the same peak voltage
triangle wave.
14 CHAPTER 1. BASIC AC THEORY
peaking at 10 volts. The effects of these two AC voltages powering a load would be quite
different: Figure 1.16
One way of expressing the amplitude of different waveshapes in a more equivalent fashion
is to mathematically average the values of all the points on a waveform’s graph to a single,
aggregate number. This amplitude measure is known simply as the average value of the wave
form. If we average all the points on the waveform algebraically (that is, to consider their sign,
either positive or negative), the average value for most waveforms is technically zero, because
all the positive points cancel out all the negative points over a full cycle: Figure 1.17
+
+
+
+ + +
+
+
+









True average value of all points
(considering their signs) is zero!
Figure 1.17: The average value of a sinewave is zero.
This, of course, will be true for any waveform having equalarea portions above and below
the “zero” line of a plot. However, as a practical measure of a waveform’s aggregate value,
“average” is usually defined as the mathematical mean of all the points’ absolute values over a
cycle. In other words, we calculate the practical average value of the waveform by considering
all points on the wave as positive quantities, as if the waveform looked like this: Figure 1.18
+
+
+
+ + +
+
+
++
+
+
+ + +
+
+
+
Practical average of points, all
values assumed to be positive.
Figure 1.18: Waveform seen by AC “average responding” meter.
Polarityinsensitive mechanical meter movements (meters designed to respond equally to
the positive and negative halfcycles of an alternating voltage or current) register in proportion
to the waveform’s (practical) average value, because the inertia of the pointer against the ten
sion of the spring naturally averages the force produced by the varying voltage/current values
over time. Conversely, polaritysensitive meter movements vibrate uselessly if exposed to AC
voltage or current, their needles oscillating rapidly about the zero mark, indicating the true
(algebraic) average value of zero for a symmetrical waveform. When the “average” value of a
waveform is referenced in this text, it will be assumed that the “practical” definition of average
1.3. MEASUREMENTS OF AC MAGNITUDE 15
is intended unless otherwise specified.
Another method of deriving an aggregate value for waveform amplitude is based on the
waveform’s ability to do useful work when applied to a load resistance. Unfortunately, an AC
measurement based on work performed by a waveform is not the same as that waveform’s
“average” value, because the power dissipated by a given load (work performed per unit time)
is not directly proportional to the magnitude of either the voltage or current impressed upon
it. Rather, power is proportional to the square of the voltage or current applied to a resistance
(P = E2/R, and P = I2R). Although the mathematics of such an amplitude measurement might
not be straightforward, the utility of it is.
Consider a bandsaw and a jigsaw, two pieces of modern woodworking equipment. Both
types of saws cut with a thin, toothed, motorpowered metal blade to cut wood. But while
the bandsaw uses a continuous motion of the blade to cut, the jigsaw uses a backandforth
motion. The comparison of alternating current (AC) to direct current (DC) may be likened to
the comparison of these two saw types: Figure 1.19
blade
motion
(analogous to DC)
blade
motion
(analogous to AC)
Bandsaw
Jigsaw
wood
wood
Figure 1.19: Bandsawjigsaw analogy of DC vs AC.
The problem of trying to describe the changing quantities of AC voltage or current in a
single, aggregate measurement is also present in this saw analogy: how might we express the
speed of a jigsaw blade? A bandsaw blade moves with a constant speed, similar to the way DC
voltage pushes or DC current moves with a constant magnitude. A jigsaw blade, on the other
hand, moves back and forth, its blade speed constantly changing. What is more, the backand
forth motion of any two jigsaws may not be of the same type, depending on the mechanical
design of the saws. One jigsaw might move its blade with a sinewave motion, while another
with a trianglewave motion. To rate a jigsaw based on its peak blade speed would be quite
misleading when comparing one jigsaw to another (or a jigsaw with a bandsaw!). Despite the
fact that these different saws move their blades in different manners, they are equal in one
respect: they all cut wood, and a quantitative comparison of this common function can serve
as a common basis for which to rate blade speed.
Picture a jigsaw and bandsaw sidebyside, equipped with identical blades (same tooth
pitch, angle, etc.), equally capable of cutting the same thickness of the same type of wood at the
same rate. We might say that the two saws were equivalent or equal in their cutting capacity.
16 CHAPTER 1. BASIC AC THEORY
Might this comparison be used to assign a “bandsaw equivalent” blade speed to the jigsaw’s
backandforth blade motion; to relate the woodcutting effectiveness of one to the other? This
is the general idea used to assign a “DC equivalent” measurement to any AC voltage or cur
rent: whatever magnitude of DC voltage or current would produce the same amount of heat
energy dissipation through an equal resistance:Figure 1.20
RMS
power
dissipated
power
dissipated
10 V 10 V2 Ω 2 Ω
50 W 50 W5A RMS
5 A
5 A
Equal power dissipated through
equal resistance loads
5A RMS
Figure 1.20: An RMS voltage produces the same heating effect as a the same DC voltage
In the two circuits above, we have the same amount of load resistance (2 Ω) dissipating the
same amount of power in the form of heat (50 watts), one powered by AC and the other by
DC. Because the AC voltage source pictured above is equivalent (in terms of power delivered
to a load) to a 10 volt DC battery, we would call this a “10 volt” AC source. More specifically,
we would denote its voltage value as being 10 volts RMS. The qualifier “RMS” stands for
Root Mean Square, the algorithm used to obtain the DC equivalent value from points on a
graph (essentially, the procedure consists of squaring all the positive and negative points on a
waveform graph, averaging those squared values, then taking the square root of that average
to obtain the final answer). Sometimes the alternative terms equivalent or DC equivalent are
used instead of “RMS,” but the quantity and principle are both the same.
RMS amplitude measurement is the best way to relate AC quantities to DC quantities, or
other AC quantities of differing waveform shapes, when dealing with measurements of elec
tric power. For other considerations, peak or peaktopeak measurements may be the best to
employ. For instance, when determining the proper size of wire (ampacity) to conduct electric
power from a source to a load, RMS current measurement is the best to use, because the prin
cipal concern with current is overheating of the wire, which is a function of power dissipation
caused by current through the resistance of the wire. However, when rating insulators for
service in highvoltage AC applications, peak voltage measurements are the most appropriate,
because the principal concern here is insulator “flashover” caused by brief spikes of voltage,
irrespective of time.
Peak and peaktopeak measurements are best performed with an oscilloscope, which can
capture the crests of the waveform with a high degree of accuracy due to the fast action of
the cathoderaytube in response to changes in voltage. For RMS measurements, analog meter
movements (D’Arsonval, Weston, iron vane, electrodynamometer) will work so long as they
have been calibrated in RMS figures. Because the mechanical inertia and dampening effects
of an electromechanical meter movement makes the deflection of the needle naturally pro
portional to the average value of the AC, not the true RMS value, analog meters must be
specifically calibrated (or miscalibrated, depending on how you look at it) to indicate voltage
1.3. MEASUREMENTS OF AC MAGNITUDE 17
or current in RMS units. The accuracy of this calibration depends on an assumed waveshape,
usually a sine wave.
Electronic meters specifically designed for RMS measurement are best for the task. Some
instrument manufacturers have designed ingenious methods for determining the RMS value
of any waveform. One such manufacturer produces “TrueRMS” meters with a tiny resistive
heating element powered by a voltage proportional to that being measured. The heating effect
of that resistance element is measured thermally to give a true RMS value with no mathemat
ical calculations whatsoever, just the laws of physics in action in fulfillment of the definition of
RMS. The accuracy of this type of RMS measurement is independent of waveshape.
For “pure” waveforms, simple conversion coefficients exist for equating Peak, PeaktoPeak,
Average (practical, not algebraic), and RMS measurements to one another: Figure 1.21
RMS = 0.707 (Peak)
AVG = 0.637 (Peak)
PP = 2 (Peak)
RMS = Peak
AVG = Peak
PP = 2 (Peak)
RMS = 0.577 (Peak)
AVG = 0.5 (Peak)
PP = 2 (Peak)
Figure 1.21: Conversion factors for common waveforms.
In addition to RMS, average, peak (crest), and peaktopeak measures of an AC waveform,
there are ratios expressing the proportionality between some of these fundamental measure
ments. The crest factor of an AC waveform, for instance, is the ratio of its peak (crest) value
divided by its RMS value. The form factor of an AC waveform is the ratio of its RMS value
divided by its average value. Squareshaped waveforms always have crest and form factors
equal to 1, since the peak is the same as the RMS and average values. Sinusoidal waveforms
have an RMS value of 0.707 (the reciprocal of the square root of 2) and a form factor of 1.11
(0.707/0.636). Triangle and sawtoothshaped waveforms have RMS values of 0.577 (the recip
rocal of square root of 3) and form factors of 1.15 (0.577/0.5).
Bear in mind that the conversion constants shown here for peak, RMS, and average ampli
tudes of sine waves, square waves, and triangle waves hold true only for pure forms of these
waveshapes. The RMS and average values of distorted waveshapes are not related by the same
ratios: Figure 1.22
RMS = ???
AVG = ???
PP = 2 (Peak)
Figure 1.22: Arbitrary waveforms have no simple conversions.
This is a very important concept to understand when using an analog D’Arsonval meter
18 CHAPTER 1. BASIC AC THEORY
movement to measure AC voltage or current. An analog D’Arsonval movement, calibrated to
indicate sinewave RMS amplitude, will only be accurate when measuring pure sine waves.
If the waveform of the voltage or current being measured is anything but a pure sine wave,
the indication given by the meter will not be the true RMS value of the waveform, because
the degree of needle deflection in an analog D’Arsonval meter movement is proportional to the
average value of the waveform, not the RMS. RMS meter calibration is obtained by “skewing”
the span of the meter so that it displays a small multiple of the average value, which will be
equal to be the RMS value for a particular waveshape and a particular waveshape only.
Since the sinewave shape is most common in electrical measurements, it is the waveshape
assumed for analog meter calibration, and the small multiple used in the calibration of the me
ter is 1.1107 (the form factor: 0.707/0.636: the ratio of RMS divided by average for a sinusoidal
waveform). Any waveshape other than a pure sine wave will have a different ratio of RMS and
average values, and thus a meter calibrated for sinewave voltage or current will not indicate
true RMS when reading a nonsinusoidal wave. Bear in mind that this limitation applies only
to simple, analog AC meters not employing “TrueRMS” technology.
• REVIEW:
• The amplitude of an AC waveform is its height as depicted on a graph over time. An am
plitude measurement can take the form of peak, peaktopeak, average, or RMS quantity.
• Peak amplitude is the height of an AC waveform as measured from the zero mark to the
highest positive or lowest negative point on a graph. Also known as the crest amplitude
of a wave.
• Peaktopeak amplitude is the total height of an AC waveform as measured from maxi
mum positive to maximum negative peaks on a graph. Often abbreviated as “PP”.
• Average amplitude is the mathematical “mean” of all a waveform’s points over the period
of one cycle. Technically, the average amplitude of any waveform with equalarea portions
above and below the “zero” line on a graph is zero. However, as a practical measure of
amplitude, a waveform’s average value is often calculated as the mathematical mean of
all the points’ absolute values (taking all the negative values and considering them as
positive). For a sine wave, the average value so calculated is approximately 0.637 of its
peak value.
• “RMS” stands for Root Mean Square, and is a way of expressing an AC quantity of volt
age or current in terms functionally equivalent to DC. For example, 10 volts AC RMS is
the amount of voltage that would produce the same amount of heat dissipation across a
resistor of given value as a 10 volt DC power supply. Also known as the “equivalent” or
“DC equivalent” value of an AC voltage or current. For a sine wave, the RMS value is
approximately 0.707 of its peak value.
• The crest factor of an AC waveform is the ratio of its peak (crest) to its RMS value.
• The form factor of an AC waveform is the ratio of its RMS value to its average value.
• Analog, electromechanical meter movements respond proportionally to the average value
of an AC voltage or current. When RMS indication is desired, the meter’s calibration
1.4. SIMPLE AC CIRCUIT CALCULATIONS 19
must be “skewed” accordingly. This means that the accuracy of an electromechanical
meter’s RMS indication is dependent on the purity of the waveform: whether it is the
exact same waveshape as the waveform used in calibrating.
1.4 Simple AC circuit calculations
Over the course of the next few chapters, you will learn that AC circuit measurements and cal
culations can get very complicated due to the complex nature of alternating current in circuits
with inductance and capacitance. However, with simple circuits (figure 1.23) involving nothing
more than an AC power source and resistance, the same laws and rules of DC apply simply
and directly.
10 V
R1
R2
R3
100 Ω
500 Ω
400 Ω
Figure 1.23: AC circuit calculations for resistive circuits are the same as for DC.
Rtotal = R1 + R2 + R3
Rtotal = 1 kΩ
Itotal =
Etotal
Rtotal
Itotal =
10 V
1 kΩ
Itotal = 10 mA
ER1 = ItotalR1 ER2 = ItotalR2 ER3 = ItotalR3
ER1 = 1 V ER2 = 5 V ER3 = 4 V
Series resistances still add, parallel resistances still diminish, and the Laws of Kirchhoff
and Ohm still hold true. Actually, as we will discover later on, these rules and laws always
hold true, its just that we have to express the quantities of voltage, current, and opposition to
current in more advanced mathematical forms. With purely resistive circuits, however, these
complexities of AC are of no practical consequence, and so we can treat the numbers as though
we were dealing with simple DC quantities.
20 CHAPTER 1. BASIC AC THEORY
Because all these mathematical relationships still hold true, we can make use of our famil
iar “table” method of organizing circuit values just as with DC:
E
I
R
Volts
Amps
Ohms
1
100
R1 R2 R3 Total
10
400500
10m 10m 10m 10m
1k
5 4
One major caveat needs to be given here: all measurements of AC voltage and current
must be expressed in the same terms (peak, peaktopeak, average, or RMS). If the source
voltage is given in peak AC volts, then all currents and voltages subsequently calculated are
cast in terms of peak units. If the source voltage is given in AC RMS volts, then all calculated
currents and voltages are cast in AC RMS units as well. This holds true for any calculation
based on Ohm’s Laws, Kirchhoff ’s Laws, etc. Unless otherwise stated, all values of voltage and
current in AC circuits are generally assumed to be RMS rather than peak, average, or peakto
peak. In some areas of electronics, peak measurements are assumed, but in most applications
(especially industrial electronics) the assumption is RMS.
• REVIEW:
• All the old rules and laws of DC (Kirchhoff ’s Voltage and Current Laws, Ohm’s Law) still
hold true for AC. However, with more complex circuits, we may need to represent the AC
quantities in more complex form. More on this later, I promise!
• The “table” method of organizing circuit values is still a valid analysis tool for AC circuits.
1.5 AC phase
Things start to get complicated when we need to relate two or more AC voltages or currents
that are out of step with each other. By “out of step,” I mean that the two waveforms are not
synchronized: that their peaks and zero points do not match up at the same points in time.
The graph in figure 1.24 illustrates an example of this.
A B
A B
A B
A B
A B
A B
Figure 1.24: Out of phase waveforms
The two waves shown above (A versus B) are of the same amplitude and frequency, but
they are out of step with each other. In technical terms, this is called a phase shift. Earlier
1.5. AC PHASE 21
we saw how we could plot a “sine wave” by calculating the trigonometric sine function for
angles ranging from 0 to 360 degrees, a full circle. The starting point of a sine wave was zero
amplitude at zero degrees, progressing to full positive amplitude at 90 degrees, zero at 180
degrees, full negative at 270 degrees, and back to the starting point of zero at 360 degrees. We
can use this angle scale along the horizontal axis of our waveform plot to express just how far
out of step one wave is with another: Figure 1.25
A B
0 90 180 270 360
(0)
90 180 270 360
(0)
0 90 180 270 360
(0)
90 180 270 360
(0)
A
B
degrees
degrees
Figure 1.25: Wave A leads wave B by 45o
The shift between these two waveforms is about 45 degrees, the “A” wave being ahead of
the “B” wave. A sampling of different phase shifts is given in the following graphs to better
illustrate this concept: Figure 1.26
Because the waveforms in the above examples are at the same frequency, they will be out of
step by the same angular amount at every point in time. For this reason, we can express phase
shift for two or more waveforms of the same frequency as a constant quantity for the entire
wave, and not just an expression of shift between any two particular points along the waves.
That is, it is safe to say something like, “voltage ’A’ is 45 degrees out of phase with voltage ’B’.”
Whichever waveform is ahead in its evolution is said to be leading and the one behind is said
to be lagging.
Phase shift, like voltage, is always a measurement relative between two things. There’s
really no such thing as a waveform with an absolute phase measurement because there’s no
known universal reference for phase. Typically in the analysis of AC circuits, the voltage
waveform of the power supply is used as a reference for phase, that voltage stated as “xxx
volts at 0 degrees.” Any other AC voltage or current in that circuit will have its phase shift
expressed in terms relative to that source voltage.
This is what makes AC circuit calculations more complicated than DC. When applying
Ohm’s Law and Kirchhoff ’s Laws, quantities of AC voltage and current must reflect phase
shift as well as amplitude. Mathematical operations of addition, subtraction, multiplication,
and division must operate on these quantities of phase shift as well as amplitude. Fortunately,
22 CHAPTER 1. BASIC AC THEORY
A B
Phase shift = 90 degrees
A is ahead of B
(A "leads" B)
B A
Phase shift = 90 degrees
B is ahead of A
(B "leads" A)
A
B
Phase shift = 180 degrees
A and B waveforms are
mirrorimages of each other
A B
Phase shift = 0 degrees
A and B waveforms are
in perfect step with each other
Figure 1.26: Examples of phase shifts.
1.6. PRINCIPLES OF RADIO 23
there is a mathematical system of quantities called complex numbers ideally suited for this
task of representing amplitude and phase.
Because the subject of complex numbers is so essential to the understanding of AC circuits,
the next chapter will be devoted to that subject alone.
• REVIEW:
• Phase shift is where two or more waveforms are out of step with each other.
• The amount of phase shift between two waves can be expressed in terms of degrees, as
defined by the degree units on the horizontal axis of the waveform graph used in plotting
the trigonometric sine function.
• A leading waveform is defined as one waveform that is ahead of another in its evolution.
A lagging waveform is one that is behind another. Example:
A B
Phase shift = 90 degrees
A leads B; B lags A
•
• Calculations for AC circuit analysis must take into consideration both amplitude and
phase shift of voltage and current waveforms to be completely accurate. This requires
the use of a mathematical system called complex numbers.
1.6 Principles of radio
One of the more fascinating applications of electricity is in the generation of invisible ripples
of energy called radio waves. The limited scope of this lesson on alternating current does not
permit full exploration of the concept, some of the basic principles will be covered.
With Oersted’s accidental discovery of electromagnetism, it was realized that electricity and
magnetism were related to each other. When an electric current was passed through a conduc
tor, a magnetic field was generated perpendicular to the axis of flow. Likewise, if a conductor
was exposed to a change in magnetic flux perpendicular to the conductor, a voltage was pro
duced along the length of that conductor. So far, scientists knew that electricity and magnetism
always seemed to affect each other at right angles. However, a major discovery lay hidden just
beneath this seemingly simple concept of related perpendicularity, and its unveiling was one
of the pivotal moments in modern science.
This breakthrough in physics is hard to overstate. The man responsible for this concep
tual revolution was the Scottish physicist James Clerk Maxwell (18311879), who “unified” the
study of electricity and magnetism in four relatively tidy equations. In essence, what he dis
covered was that electric and magnetic fields were intrinsically related to one another, with or
without the presence of a conductive path for electrons to flow. Stated more formally, Maxwell’s
discovery was this:
24 CHAPTER 1. BASIC AC THEORY
A changing electric field produces a perpendicular magnetic field, and
A changing magnetic field produces a perpendicular electric field.
All of this can take place in open space, the alternating electric and magnetic fields support
ing each other as they travel through space at the speed of light. This dynamic structure of
electric and magnetic fields propagating through space is better known as an electromagnetic
wave.
There are many kinds of natural radiative energy composed of electromagnetic waves. Even
light is electromagnetic in nature. So are Xrays and “gamma” ray radiation. The only dif
ference between these kinds of electromagnetic radiation is the frequency of their oscillation
(alternation of the electric and magnetic fields back and forth in polarity). By using a source of
AC voltage and a special device called an antenna, we can create electromagnetic waves (of a
much lower frequency than that of light) with ease.
An antenna is nothing more than a device built to produce a dispersing electric or magnetic
field. Two fundamental types of antennae are the dipole and the loop: Figure 1.27
DIPOLE LOOP
Basic antenna designs
Figure 1.27: Dipole and loop antennae
While the dipole looks like nothing more than an open circuit, and the loop a short circuit,
these pieces of wire are effective radiators of electromagnetic fields when connected to AC
sources of the proper frequency. The two open wires of the dipole act as a sort of capacitor
(two conductors separated by a dielectric), with the electric field open to dispersal instead of
being concentrated between two closelyspaced plates. The closed wire path of the loop antenna
acts like an inductor with a large air core, again providing ample opportunity for the field to
disperse away from the antenna instead of being concentrated and contained as in a normal
inductor.
As the powered dipole radiates its changing electric field into space, a changing magnetic
field is produced at right angles, thus sustaining the electric field further into space, and so
on as the wave propagates at the speed of light. As the powered loop antenna radiates its
changing magnetic field into space, a changing electric field is produced at right angles, with
the same endresult of a continuous electromagnetic wave sent away from the antenna. Either
antenna achieves the same basic task: the controlled production of an electromagnetic field.
When attached to a source of highfrequency AC power, an antenna acts as a transmitting
device, converting AC voltage and current into electromagnetic wave energy. Antennas also
have the ability to intercept electromagnetic waves and convert their energy into AC voltage
and current. In this mode, an antenna acts as a receiving device: Figure 1.28
1.7. CONTRIBUTORS 25
AC voltage
produced
AC current
produced
electromagnetic radiation electromagnetic radiation
Radio receivers
Radio transmitters
Figure 1.28: Basic radio transmitter and receiver
While there ismuchmore that may be said about antenna technology, this brief introduction
is enough to give you the general idea of what’s going on (and perhaps enough information to
provoke a few experiments).
• REVIEW:
• James Maxwell discovered that changing electric fields produce perpendicular magnetic
fields, and vice versa, even in empty space.
• A twin set of electric and magnetic fields, oscillating at right angles to each other and
traveling at the speed of light, constitutes an electromagnetic wave.
• An antenna is a device made of wire, designed to radiate a changing electric field or
changing magnetic field when powered by a highfrequency AC source, or intercept an
electromagnetic field and convert it to an AC voltage or current.
• The dipole antenna consists of two pieces of wire (not touching), primarily generating an
electric field when energized, and secondarily producing a magnetic field in space.
• The loop antenna consists of a loop of wire, primarily generating a magnetic field when
energized, and secondarily producing an electric field in space.
1.7 Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
Harvey Lew (February 7, 2004): Corrected typographical error: “circuit” should have been
“circle”.
26 CHAPTER 1. BASIC AC THEORY
Duane Damiano (February 25, 2003): Pointed out magnetic polarity error in DC generator
illustration.
Mark D. Zarella (April 28, 2002): Suggestion for improving explanation of “average” wave
form amplitude.
John Symonds (March 28, 2002): Suggestion for improving explanation of the unit “Hertz.”
Jason Starck (June 2000): HTML document formatting, which led to a much better
looking second edition.
Chapter 2
COMPLEX NUMBERS
Contents
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.2 Vectors and AC waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.3 Simple vector addition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.4 Complex vector addition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.5 Polar and rectangular notation . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.6 Complex number arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
2.7 More on AC ”polarity” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
2.8 Some examples with AC circuits . . . . . . . . . . . . . . . . . . . . . . . . . 49
2.9 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
2.1 Introduction
If I needed to describe the distance between two cities, I could provide an answer consisting of
a single number in miles, kilometers, or some other unit of linear measurement. However, if I
were to describe how to travel from one city to another, I would have to provide more informa
tion than just the distance between those two cities; I would also have to provide information
about the direction to travel, as well.
The kind of information that expresses a single dimension, such as linear distance, is called
a scalar quantity in mathematics. Scalar numbers are the kind of numbers you’ve used in most
all of your mathematical applications so far. The voltage produced by a battery, for example,
is a scalar quantity. So is the resistance of a piece of wire (ohms), or the current through it
(amps).
However, when we begin to analyze alternating current circuits, we find that quantities
of voltage, current, and even resistance (called impedance in AC) are not the familiar one
dimensional quantities we’re used to measuring in DC circuits. Rather, these quantities, be
cause they’re dynamic (alternating in direction and amplitude), possess other dimensions that
27
28 CHAPTER 2. COMPLEX NUMBERS
must be taken into account. Frequency and phase shift are two of these dimensions that come
into play. Even with relatively simple AC circuits, where we’re only dealing with a single fre
quency, we still have the dimension of phase shift to contend with in addition to the amplitude.
In order to successfully analyze AC circuits, we need to work with mathematical objects
and techniques capable of representing these multidimensional quantities. Here is where
we need to abandon scalar numbers for something better suited: complex numbers. Just like
the example of giving directions from one city to another, AC quantities in a singlefrequency
circuit have both amplitude (analogy: distance) and phase shift (analogy: direction). A complex
number is a single mathematical quantity able to express these two dimensions of amplitude
and phase shift at once.
Complex numbers are easier to grasp when they’re represented graphically. If I draw a line
with a certain length (magnitude) and angle (direction), I have a graphic representation of a
complex number which is commonly known in physics as a vector: (Figure 2.1)
length = 7
angle = 0 degrees
length = 10
angle = 180 degrees
length = 5
angle = 90 degrees
length = 4
angle = 270 degrees
(90 degrees)
length = 5.66
angle = 45 degrees
length = 9.43
(57.99 degrees)
angle = 302.01 degrees
Figure 2.1: A vector has both magnitude and direction.
Like distances and directions on a map, there must be some common frame of reference for
angle figures to have any meaning. In this case, directly right is considered to be 0o, and angles
are counted in a positive direction going counterclockwise: (Figure 2.2)
The idea of representing a number in graphical form is nothing new. We all learned this in
grade school with the “number line:” (Figure 2.3)
We even learned how addition and subtraction works by seeing how lengths (magnitudes)
stacked up to give a final answer: (Figure 2.4)
Later, we learned that there were ways to designate the values between the whole numbers
marked on the line. These were fractional or decimal quantities: (Figure 2.5)
Later yet we learned that the number line could extend to the left of zero as well: (Fig
ure 2.6)
2.1. INTRODUCTION 29
0o
90o
180o
270o (90o)
The vector "compass"
Figure 2.2: The vector compass
0 1 2 3 4 5 6 7 8 9 10
. . .
Figure 2.3: Number line.
0 1 2 3 4 5 6 7 8 9 10
. . .
5 3
8
5 + 3 = 8
Figure 2.4: Addition on a “number line”.
0 1 2 3 4 5 6 7 8 9 10
. . .
31/2 or 3.5
Figure 2.5: Locating a fraction on the “number line”
30 CHAPTER 2. COMPLEX NUMBERS
0 1 2 3 4 5
. . .. . .
12345
Figure 2.6: “Number line” shows both positive and negative numbers.
These fields of numbers (whole, integer, rational, irrational, real, etc.) learned in grade
school share a common trait: they’re all onedimensional. The straightness of the number
line illustrates this graphically. You can move up or down the number line, but all “motion”
along that line is restricted to a single axis (horizontal). Onedimensional, scalar numbers are
perfectly adequate for counting beads, representing weight, or measuring DC battery voltage,
but they fall short of being able to represent something more complex like the distance and
direction between two cities, or the amplitude and phase of an AC waveform. To represent
these kinds of quantities, we need multidimensional representations. In other words, we need
a number line that can point in different directions, and that’s exactly what a vector is.
• REVIEW:
• A scalar number is the type of mathematical object that people are used to using in
everyday life: a onedimensional quantity like temperature, length, weight, etc.
• A complex number is a mathematical quantity representing two dimensions of magnitude
and direction.
• A vector is a graphical representation of a complex number. It looks like an arrow, with
a starting point, a tip, a definite length, and a definite direction. Sometimes the word
phasor is used in electrical applications where the angle of the vector represents phase
shift between waveforms.
2.2 Vectors and AC waveforms
OK, so how exactly can we represent AC quantities of voltage or current in the form of a vector?
The length of the vector represents the magnitude (or amplitude) of the waveform, like this:
(Figure 2.7)
The greater the amplitude of the waveform, the greater the length of its corresponding
vector. The angle of the vector, however, represents the phase shift in degrees between the
waveform in question and another waveform acting as a “reference” in time. Usually, when the
phase of a waveform in a circuit is expressed, it is referenced to the power supply voltage wave
form (arbitrarily stated to be “at” 0o). Remember that phase is always a relative measurement
between two waveforms rather than an absolute property. (Figure 2.8) (Figure 2.9)
The greater the phase shift in degrees between two waveforms, the greater the angle dif
ference between the corresponding vectors. Being a relative measurement, like voltage, phase
shift (vector angle) only has meaning in reference to some standard waveform. Generally this
“reference” waveform is the main AC power supply voltage in the circuit. If there is more than
2.2. VECTORS AND AC WAVEFORMS 31
Amplitude
Length
Waveform Vector representation
Figure 2.7: Vector length represents AC voltage magnitude.
A B
Phase shift = 90 degrees
A is ahead of B
(A "leads" B)
B A
Phase shift = 90 degrees
B is ahead of A
(B "leads" A)
A
B
Phase shift = 180 degrees
A and B waveforms are
mirrorimages of each other
A B
Phase shift = 0 degrees
A and B waveforms are
in perfect step with each other
(of "A" waveform with
reference to "B" waveform)
B
A
B
A
BA
A B
90 degrees
90 degrees
180 degrees
Waveforms Phase relations Vector representations
Figure 2.8: Vector angle is the phase with respect to another waveform.
32 CHAPTER 2. COMPLEX NUMBERS
BA
B
A
phase shift
angle
Figure 2.9: Phase shift between waves and vector phase angle
one AC voltage source, then one of those sources is arbitrarily chosen to be the phase reference
for all other measurements in the circuit.
This concept of a reference point is not unlike that of the “ground” point in a circuit for
the benefit of voltage reference. With a clearly defined point in the circuit declared to be
“ground,” it becomes possible to talk about voltage “on” or “at” single points in a circuit, being
understood that those voltages (always relative between two points) are referenced to “ground.”
Correspondingly, with a clearly defined point of reference for phase it becomes possible to speak
of voltages and currents in an AC circuit having definite phase angles. For example, if the
current in an AC circuit is described as “24.3 milliamps at 64 degrees,” it means that the
current waveform has an amplitude of 24.3 mA, and it lags 64o behind the reference waveform,
usually assumed to be the main source voltage waveform.
• REVIEW:
• When used to describe an AC quantity, the length of a vector represents the amplitude
of the wave while the angle of a vector represents the phase angle of the wave relative to
some other (reference) waveform.
2.3 Simple vector addition
Remember that vectors are mathematical objects just like numbers on a number line: they
can be added, subtracted, multiplied, and divided. Addition is perhaps the easiest vector op
eration to visualize, so we’ll begin with that. If vectors with common angles are added, their
magnitudes (lengths) add up just like regular scalar quantities: (Figure 2.10)
length = 6
angle = 0 degrees
length = 8
angle = 0 degrees
total length = 6 + 8 = 14
angle = 0 degrees
Figure 2.10: Vector magnitudes add like scalars for a common angle.
Similarly, if AC voltage sources with the same phase angle are connected together in series,
their voltages add just as you might expect with DC batteries: (Figure 2.11)
Please note the (+) and () polarity marks next to the leads of the two AC sources. Even
though we know AC doesn’t have “polarity” in the same sense that DC does, these marks are
2.3. SIMPLE VECTOR ADDITION 33
0 deg 0 deg
0 deg
 +  +
 +
 +  +
 +
6 V 8 V
14 V 14 V
6 V 8 V
Figure 2.11: “In phase” AC voltages add like DC battery voltages.
essential to knowing how to reference the given phase angles of the voltages. This will become
more apparent in the next example.
If vectors directly opposing each other (180o out of phase) are added together, their magni
tudes (lengths) subtract just like positive and negative scalar quantities subtract when added:
(Figure 2.12)
length = 6 angle = 0 degrees
length = 8
total length = 6  8 = 2 at 0 degrees
angle = 180 degrees
or 2 at 180 degrees
Figure 2.12: Directly opposing vector magnitudes subtract.
Similarly, if opposing AC voltage sources are connected in series, their voltages subtract as
you might expect with DC batteries connected in an opposing fashion: (Figure 2.13)
Determining whether or not these voltage sources are opposing each other requires an ex
amination of their polarity markings and their phase angles. Notice how the polarity markings
in the above diagram seem to indicate additive voltages (from left to right, we see  and + on
the 6 volt source,  and + on the 8 volt source). Even though these polarity markings would
normally indicate an additive effect in a DC circuit (the two voltages working together to pro
duce a greater total voltage), in this AC circuit they’re actually pushing in opposite directions
because one of those voltages has a phase angle of 0o and the other a phase angle of 180o. The
result, of course, is a total voltage of 2 volts.
We could have just as well shown the opposing voltages subtracting in series like this:
(Figure 2.14)
Note how the polarities appear to be opposed to each other now, due to the reversal of
wire connections on the 8 volt source. Since both sources are described as having equal phase
34 CHAPTER 2. COMPLEX NUMBERS
0 deg
 +  +
 +
 + +
+
180 deg
180 deg
6 V 8 V
6 V 8 V
2 V 2 V
Figure 2.13: Opposing AC voltages subtract like opposing battery voltages.
0 deg
 +  +
 +
 +
+
180 deg
6 V
8 V
6 V
8 V
2 V 2 V
0 deg
 +
Figure 2.14: Opposing voltages in spite of equal phase angles.
2.4. COMPLEX VECTOR ADDITION 35
angles (0o), they truly are opposed to one another, and the overall effect is the same as the
former scenario with “additive” polarities and differing phase angles: a total voltage of only 2
volts. (Figure 2.15)
0 deg
 + +
 +
180 deg
0 deg
0 deg
+ 
6 V 8 V
2 V
2 V
Figure 2.15: Just as there are two ways to express the phase of the sources, there are two ways
to express the resultant their sum.
The resultant voltage can be expressed in two different ways: 2 volts at 180o with the ()
symbol on the left and the (+) symbol on the right, or 2 volts at 0o with the (+) symbol on the
left and the () symbol on the right. A reversal of wires from an AC voltage source is the same
as phaseshifting that source by 180o. (Figure 2.16)
8 V
180 deg
 +
8 V
+
0 degThese voltage sources
are equivalent!
Figure 2.16: Example of equivalent voltage sources.
2.4 Complex vector addition
If vectors with uncommon angles are added, their magnitudes (lengths) add up quite differ
ently than that of scalar magnitudes: (Figure 2.17)
If two AC voltages – 90o out of phase – are added together by being connected in series, their
voltage magnitudes do not directly add or subtract as with scalar voltages in DC. Instead, these
voltage quantities are complex quantities, and just like the above vectors, which add up in a
trigonometric fashion, a 6 volt source at 0o added to an 8 volt source at 90o results in 10 volts
at a phase angle of 53.13o: (Figure 2.18)
Compared to DC circuit analysis, this is very strange indeed. Note that it is possible to
obtain voltmeter indications of 6 and 8 volts, respectively, across the two AC voltage sources,
36 CHAPTER 2. COMPLEX NUMBERS
length = 6
angle = 0 degrees
length = 8
angle = 90 degrees
length = 10
angle = 53.13
degrees
6 at 0 degrees
8 at 90 degrees+
10 at 53.13 degrees
Vector addition
Figure 2.17: Vector magnitudes do not directly add for unequal angles.
0 deg
 +  +
90 deg
53.13 deg
 +
6 V 8 V
10 V
Figure 2.18: The 6V and 8V sources add to 10V with the help of trigonometry.
2.5. POLAR AND RECTANGULAR NOTATION 37
yet only read 10 volts for a total voltage!
There is no suitable DC analogy for what we’re seeing here with two AC voltages slightly
out of phase. DC voltages can only directly aid or directly oppose, with nothing in between.
With AC, two voltages can be aiding or opposing one another to any degree between fully
aiding and fullyopposing, inclusive. Without the use of vector (complex number) notation to
describe AC quantities, it would be very difficult to perform mathematical calculations for AC
circuit analysis.
In the next section, we’ll learn how to represent vector quantities in symbolic rather than
graphical form. Vector and triangle diagrams suffice to illustrate the general concept, but more
precise methods of symbolism must be used if any serious calculations are to be performed on
these quantities.
• REVIEW:
• DC voltages can only either directly aid or directly oppose each other when connected in
series. AC voltages may aid or oppose to any degree depending on the phase shift between
them.
2.5 Polar and rectangular notation
In order to work with these complex numbers without drawing vectors, we first need some kind
of standard mathematical notation. There are two basic forms of complex number notation:
polar and rectangular.
Polar form is where a complex number is denoted by the length (otherwise known as the
magnitude, absolute value, or modulus) and the angle of its vector (usually denoted by an
angle symbol that looks like this: 6 ). To use the map analogy, polar notation for the vector
from New York City to San Diego would be something like “2400 miles, southwest.” Here are
two examples of vectors and their polar notations: (Figure 2.19)
8.49 ∠ 45o
8.06 ∠ 29.74o
(8.06 ∠ 330.26o)
5.39 ∠ 158.2o 7.81 ∠ 230.19o
(7.81 ∠ 129.81o)
Note: the proper notation for designating a vector’s angle
is this symbol: ∠
Figure 2.19: Vectors with polar notations.
38 CHAPTER 2. COMPLEX NUMBERS
Standard orientation for vector angles in AC circuit calculations defines 0o as being to the
right (horizontal), making 90o straight up, 180o to the left, and 270o straight down. Please note
that vectors angled “down” can have angles represented in polar form as positive numbers in
excess of 180, or negative numbers less than 180. For example, a vector angled 6 270o (straight
down) can also be said to have an angle of 90o. (Figure 2.20) The above vector on the right
(7.81 6 230.19o) can also be denoted as 7.81 6 129.81o.
0o
90o
180o
270o (90o)
The vector "compass"
Figure 2.20: The vector compass
Rectangular form, on the other hand, is where a complex number is denoted by its re
spective horizontal and vertical components. In essence, the angled vector is taken to be the
hypotenuse of a right triangle, described by the lengths of the adjacent and opposite sides.
Rather than describing a vector’s length and direction by denoting magnitude and angle, it is
described in terms of “how far left/right” and “how far up/down.”
These two dimensional figures (horizontal and vertical) are symbolized by two numerical
figures. In order to distinguish the horizontal and vertical dimensions from each other, the
vertical is prefixed with a lowercase “i” (in pure mathematics) or “j” (in electronics). These
lowercase letters do not represent a physical variable (such as instantaneous current, also
symbolized by a lowercase letter “i”), but rather are mathematical operators used to distin
guish the vector’s vertical component from its horizontal component. As a complete complex
number, the horizontal and vertical quantities are written as a sum: (Figure 2.21)
The horizontal component is referred to as the real component, since that dimension is
compatible with normal, scalar (“real”) numbers. The vertical component is referred to as the
imaginary component, since that dimension lies in a different direction, totally alien to the
scale of the real numbers. (Figure 2.22)
The “real” axis of the graph corresponds to the familiar number line we saw earlier: the one
with both positive and negative values on it. The “imaginary” axis of the graph corresponds to
another number line situated at 90o to the “real” one. Vectors being twodimensional things,
2.5. POLAR AND RECTANGULAR NOTATION 39
4 + j4
"4 right and 4 up"
4 + j0
"4 right and 0 up/down"
4  j4
"4 right and 4 down"
4 + j0
"4 left and 0 up/down"
4 + j4
"4 left and 4 up"
4 j4
"4 left and 4 down"
Figure 2.21: In “rectangular” form the vector’s length and direction are denoted in terms of its
horizontal and vertical span, the first number representing the the horizontal (“real”) and the
second number (with the “j” prefix) representing the vertical (“imaginary”) dimensions.
+j
j
+ "imaginary"
 "imaginary"
+ "real" "real"
Figure 2.22: Vector compass showing real and imaginary axes
40 CHAPTER 2. COMPLEX NUMBERS
we must have a twodimensional “map” upon which to express them, thus the two number
lines perpendicular to each other: (Figure 2.23)
0
1 2 3 4 5
. . .. . .
12345
1
2
3
4
5
1
2
3
4
5
"real" number line
"imaginary"
number line
Figure 2.23: Vector compass with real and imaginary (“j”) number lines.
Either method of notation is valid for complex numbers. The primary reason for having
two methods of notation is for ease of longhand calculation, rectangular form lending itself to
addition and subtraction, and polar form lending itself to multiplication and division.
Conversion between the two notational forms involves simple trigonometry. To convert from
polar to rectangular, find the real component by multiplying the polar magnitude by the cosine
of the angle, and the imaginary component by multiplying the polar magnitude by the sine of
the angle. This may be understood more readily by drawing the quantities as sides of a right
triangle, the hypotenuse of the triangle representing the vector itself (its length and angle
with respect to the horizontal constituting the polar form), the horizontal and vertical sides
representing the “real” and “imaginary” rectangular components, respectively: (Figure 2.24)
2.5. POLAR AND RECTANGULAR NOTATION 41
+j3
+4
length = 5
angle =
36.87o
Figure 2.24: Magnitude vector in terms of real (4) and imaginary (j3) components.
(polar form)
(real component)
(imaginary component)
4 + j3 (rectangular form)
(5)(cos 36.87o) = 4
(5)(sin 36.87o) = 3
5 ∠ 36.87o
To convert from rectangular to polar, find the polar magnitude through the use of the
Pythagorean Theorem (the polar magnitude is the hypotenuse of a right triangle, and the real
and imaginary components are the adjacent and opposite sides, respectively), and the angle by
taking the arctangent of the imaginary component divided by the real component:
4 + j3 (rectangular form)
c = a2 + b2 (pythagorean theorem)
polar magnitude = 42 + 32
polar magnitude = 5
polar angle = arctan 3
4
polar angle =
(polar form)
36.87o
5 ∠ 36.87o
• REVIEW:
• Polar notation denotes a complex number in terms of its vector’s length and angular
direction from the starting point. Example: fly 45 miles 6 203o (West by Southwest).
42 CHAPTER 2. COMPLEX NUMBERS
• Rectangular notation denotes a complex number in terms of its horizontal and vertical
dimensions. Example: drive 41 miles West, then turn and drive 18 miles South.
• In rectangular notation, the first quantity is the “real” component (horizontal dimension
of vector) and the second quantity is the “imaginary” component (vertical dimension of
vector). The imaginary component is preceded by a lowercase “j,” sometimes called the j
operator.
• Both polar and rectangular forms of notation for a complex number can be related graph
ically in the form of a right triangle, with the hypotenuse representing the vector itself
(polar form: hypotenuse length = magnitude; angle with respect to horizontal side = an
gle), the horizontal side representing the rectangular “real” component, and the vertical
side representing the rectangular “imaginary” component.
2.6 Complex number arithmetic
Since complex numbers are legitimate mathematical entities, just like scalar numbers, they
can be added, subtracted, multiplied, divided, squared, inverted, and such, just like any other
kind of number. Some scientific calculators are programmed to directly perform these opera
tions on two or more complex numbers, but these operations can also be done “by hand.” This
section will show you how the basic operations are performed. It is highly recommended that
you equip yourself with a scientific calculator capable of performing arithmetic functions easily
on complex numbers. It will make your study of AC circuit much more pleasant than if you’re
forced to do all calculations the longer way.
Addition and subtraction with complex numbers in rectangular form is easy. For addition,
simply add up the real components of the complex numbers to determine the real component
of the sum, and add up the imaginary components of the complex numbers to determine the
imaginary component of the sum:
2 + j5
4  j3+
6 + j2
175  j34
80  j15+
255  j49
36 + j10
20 + j82+
16 + j92
When subtracting complex numbers in rectangular form, simply subtract the real compo
nent of the second complex number from the real component of the first to arrive at the real
component of the difference, and subtract the imaginary component of the second complex
number from the imaginary component of the first to arrive the imaginary component of the
difference:
2 + j5
(4  j3)
175  j34
(80  j15)
36 + j10
(20 + j82)  
2 + j8 95  j19 56  j72
For longhand multiplication and division, polar is the favored notation to work with. When
multiplying complex numbers in polar form, simply multiply the polar magnitudes of the com
plex numbers to determine the polar magnitude of the product, and add the angles of the
complex numbers to determine the angle of the product:
2.6. COMPLEX NUMBER ARITHMETIC 43
(35 ∠ 65o)(10 ∠ 12o) = 350 ∠ 53o
(124 ∠ 250o)(11 ∠ 100o) = 1364 ∠ 10o
or
1364 ∠ 350o
(3 ∠ 30o)(5 ∠ 30o) = 15 ∠ 0o
Division of polarform complex numbers is also easy: simply divide the polar magnitude
of the first complex number by the polar magnitude of the second complex number to arrive
at the polar magnitude of the quotient, and subtract the angle of the second complex number
from the angle of the first complex number to arrive at the angle of the quotient:
35 ∠ 65o
10 ∠ 12o
=
124 ∠ 250o
11 ∠ 100o
=
3 ∠ 30o
5 ∠ 30o
=
3.5 ∠ 77o
11.273 ∠ 150o
0.6 ∠ 60o
To obtain the reciprocal, or “invert” (1/x), a complex number, simply divide the number (in
polar form) into a scalar value of 1, which is nothing more than a complex number with no
imaginary component (angle = 0):
1
= =
1
= =
1
= =
1 ∠ 0o
1 ∠ 0o
1 ∠ 0o
35 ∠ 65o 35 ∠ 65o
10 ∠ 12o 10 ∠ 12o
0.0032 ∠ 10o 0.0032 ∠ 10o
0.02857 ∠ 65o
0.1 ∠ 12o
312.5 ∠ 10o
These are the basic operations you will need to know in order to manipulate complex num
bers in the analysis of AC circuits. Operations with complex numbers are by no means limited
just to addition, subtraction, multiplication, division, and inversion, however. Virtually any
arithmetic operation that can be done with scalar numbers can be done with complex num
bers, including powers, roots, solving simultaneous equations with complex coefficients, and
even trigonometric functions (although this involves a whole new perspective in trigonometry
called hyperbolic functions which is well beyond the scope of this discussion). Be sure that
you’re familiar with the basic arithmetic operations of addition, subtraction, multiplication,
division, and inversion, and you’ll have little trouble with AC circuit analysis.
• REVIEW:
44 CHAPTER 2. COMPLEX NUMBERS
• To add complex numbers in rectangular form, add the real components and add the imag
inary components. Subtraction is similar.
• To multiply complex numbers in polar form, multiply the magnitudes and add the angles.
To divide, divide the magnitudes and subtract one angle from the other.
2.7 More on AC ”polarity”
Complex numbers are useful for AC circuit analysis because they provide a convenient method
of symbolically denoting phase shift between AC quantities like voltage and current. However,
for most people the equivalence between abstract vectors and real circuit quantities is not an
easy one to grasp. Earlier in this chapter we saw how AC voltage sources are given voltage
figures in complex form (magnitude and phase angle), as well as polarity markings. Being that
alternating current has no set “polarity” as direct current does, these polarity markings and
their relationship to phase angle tends to be confusing. This section is written in the attempt
to clarify some of these issues.
Voltage is an inherently relative quantity. When we measure a voltage, we have a choice in
how we connect a voltmeter or other voltagemeasuring instrument to the source of voltage, as
there are two points between which the voltage exists, and two test leads on the instrument
with which to make connection. In DC circuits, we denote the polarity of voltage sources and
voltage drops explicitly, using “+” and “” symbols, and use colorcoded meter test leads (red
and black). If a digital voltmeter indicates a negative DC voltage, we know that its test leads
are connected “backward” to the voltage (red lead connected to the “” and black lead to the
“+”).
Batteries have their polarity designated by way of intrinsic symbology: the shortline side
of a battery is always the negative () side and the longline side always the positive (+): (Fig
ure 2.25)
6 V
+

Figure 2.25: Conventional battery polarity.
Although it would be mathematically correct to represent a battery’s voltage as a negative
figure with reversed polarity markings, it would be decidedly unconventional: (Figure 2.26)
+

6 V
Figure 2.26: Decidedly unconventional polarity marking.
2.7. MORE ON AC ”POLARITY” 45
Interpreting such notation might be easier if the “+” and “” polarity markings were viewed
as reference points for voltmeter test leads, the “+” meaning “red” and the “” meaning “black.”
A voltmeter connected to the above battery with red lead to the bottom terminal and black
lead to the top terminal would indeed indicate a negative voltage (6 volts). Actually, this
form of notation and interpretation is not as unusual as you might think: it is commonly
encountered in problems of DC network analysis where “+” and “” polarity marks are initially
drawn according to educated guess, and later interpreted as correct or “backward” according
to the mathematical sign of the figure calculated.
In AC circuits, though, we don’t deal with “negative” quantities of voltage. Instead, we
describe to what degree one voltage aids or opposes another by phase: the timeshift between
two waveforms. We never describe an AC voltage as being negative in sign, because the facility
of polar notation allows for vectors pointing in an opposite direction. If one AC voltage directly
opposes another AC voltage, we simply say that one is 180o out of phase with the other.
Still, voltage is relative between two points, and we have a choice in how we might connect
a voltagemeasuring instrument between those two points. The mathematical sign of a DC
voltmeter’s reading has meaning only in the context of its test lead connections: which terminal
the red lead is touching, and which terminal the black lead is touching. Likewise, the phase
angle of an AC voltage has meaning only in the context of knowing which of the two points
is considered the “reference” point. Because of this fact, “+” and “” polarity marks are often
placed by the terminals of an AC voltage in schematic diagrams to give the stated phase angle
a frame of reference.
Let’s review these principles with some graphical aids. First, the principle of relating test
lead connections to the mathematical sign of a DC voltmeter indication: (Figure 2.27)
The mathematical sign of a digital DC voltmeter’s display has meaning only in the context
of its test lead connections. Consider the use of a DC voltmeter in determining whether or
not two DC voltage sources are aiding or opposing each other, assuming that both sources
are unlabeled as to their polarities. Using the voltmeter to measure across the first source:
(Figure 2.28)
This first measurement of +24 across the lefthand voltage source tells us that the black
lead of the meter really is touching the negative side of voltage source #1, and the red lead of
the meter really is touching the positive. Thus, we know source #1 is a battery facing in this
orientation: (Figure 2.29)
Measuring the other unknown voltage source: (Figure 2.30)
This second voltmeter reading, however, is a negative () 17 volts, which tells us that the
black test lead is actually touching the positive side of voltage source #2, while the red test
lead is actually touching the negative. Thus, we know that source #2 is a battery facing in the
opposite direction: (Figure 2.31)
It should be obvious to any experienced student of DC electricity that these two batteries
are opposing one another. By definition, opposing voltages subtract from one another, so we
subtract 17 volts from 24 volts to obtain the total voltage across the two: 7 volts.
We could, however, draw the two sources as nondescript boxes, labeled with the exact volt
age figures obtained by the voltmeter, the polarity marks indicating voltmeter test lead place
ment: (Figure 2.32)
According to this diagram, the polarity marks (which indicate meter test lead placement)
indicate the sources aiding each other. By definition, aiding voltage sources add with one an
other to form the total voltage, so we add 24 volts to 17 volts to obtain 7 volts: still the correct
46 CHAPTER 2. COMPLEX NUMBERS
COMA
V
V A
A
OFF
6 V
COMA
V
V A
A
OFF
6 V
Figure 2.27: Test lead colors provide a frame of reference for interpreting the sign (+ or ) of
the meter’s indication.
COMA
V
V A
A
OFF
Source 1 Source 2
Total voltage?
The meter tells us +24 volts
Figure 2.28: (+) Reading indicates black is (), red is (+).
2.7. MORE ON AC ”POLARITY” 47
Source 1 Source 2
Total voltage?
24 V
Figure 2.29: 24V source is polarized () to (+).
COMA
V
V A
A
OFF
The meter tells us 17 volts
Source 1 Source 2
Total voltage?
Figure 2.30: () Reading indicates black is (+), red is ().
Source 1 Source 2
24 V 17 V
Total voltage = 7 V +
Figure 2.31: 17V source is polarized (+) to ()
Source 1 Source 2
24 V 17 V
 +  +
Figure 2.32: Voltmeter readings as read from meters.
48 CHAPTER 2. COMPLEX NUMBERS
answer. If we let the polarity markings guide our decision to either add or subtract voltage fig
ures – whether those polarity markings represent the true polarity or just the meter test lead
orientation – and include the mathematical signs of those voltage figures in our calculations,
the result will always be correct. Again, the polarity markings serve as frames of reference to
place the voltage figures’ mathematical signs in proper context.
The same is true for AC voltages, except that phase angle substitutes for mathematical
sign. In order to relate multiple AC voltages at different phase angles to each other, we need
polarity markings to provide frames of reference for those voltages’ phase angles. (Figure 2.33)
Take for example the following circuit:
10 V ∠ 0o 6 V ∠ 45o
 +  +
14.861 V ∠ 16.59o
Figure 2.33: Phase angle substitutes for ± sign.
The polarity markings show these two voltage sources aiding each other, so to determine
the total voltage across the resistor we must add the voltage figures of 10 V 6 0o and 6 V 6 45o
together to obtain 14.861 V 6 16.59o. However, it would be perfectly acceptable to represent
the 6 volt source as 6 V 6 225o, with a reversed set of polarity markings, and still arrive at the
same total voltage: (Figure 2.34)
10 V ∠ 0o
 + +
6 V ∠ 225o
14.861 V ∠ 16.59o
Figure 2.34: Reversing the voltmeter leads on the 6V source changes the phase angle by 180o.
6 V 6 45o with negative on the left and positive on the right is exactly the same as 6 V
6 225o with positive on the left and negative on the right: the reversal of polarity markings
perfectly complements the addition of 180o to the phase angle designation: (Figure 2.35)
Unlike DC voltage sources, whose symbols intrinsically define polarity by means of short
and long lines, AC voltage symbols have no intrinsic polarity marking. Therefore, any polarity
marks must be included as additional symbols on the diagram, and there is no one “correct”
way in which to place them. They must, however, correlate with the given phase angle to
represent the true phase relationship of that voltage with other voltages in the circuit.
2.8. SOME EXAMPLES WITH AC CIRCUITS 49
6 V ∠ 45o
 +
. . . is equivalent to . . .
+
6 V ∠ 225o
Figure 2.35: Reversing polarity adds 180oto phase angle
• REVIEW:
• Polarity markings are sometimes given to AC voltages in circuit schematics in order to
provide a frame of reference for their phase angles.
2.8 Some examples with AC circuits
Let’s connect three AC voltage sources in series and use complex numbers to determine addi
tive voltages. All the rules and laws learned in the study of DC circuits apply to AC circuits
as well (Ohm’s Law, Kirchhoff ’s Laws, network analysis methods), with the exception of power
calculations (Joule’s Law). The only qualification is that all variables must be expressed in
complex form, taking into account phase as well as magnitude, and all voltages and currents
must be of the same frequency (in order that their phase relationships remain constant). (Fig
ure 2.36)
load
+


+

+
E1
E2
E3
22 V ∠ 64o
12 V ∠ 35o
15 V ∠ 0o
Figure 2.36: KVL allows addition of complex voltages.
The polarity marks for all three voltage sources are oriented in such a way that their stated
voltages should add to make the total voltage across the load resistor. Notice that although
magnitude and phase angle is given for each AC voltage source, no frequency value is specified.
If this is the case, it is assumed that all frequencies are equal, thus meeting our qualifications
for applying DC rules to an AC circuit (all figures given in complex form, all of the same
frequency). The setup of our equation to find total voltage appears as such:
50 CHAPTER 2. COMPLEX NUMBERS
Etotal = E1 + E2 + E3
(22 V ∠ 64o) + (12 V ∠ 35o) + (15 V ∠ 0o)Etotal =
Graphically, the vectors add up as shown in Figure 2.37.
22 ∠ 64o
12 ∠ 35o
15 ∠ 0o
Figure 2.37: Graphic addition of vector voltages.
The sum of these vectors will be a resultant vector originating at the starting point for the
22 volt vector (dot at upperleft of diagram) and terminating at the ending point for the 15 volt
vector (arrow tip at the middleright of the diagram): (Figure 2.38)
resultant vector
22 ∠ 64o
12 ∠ 35o
15 ∠ 0o
Figure 2.38: Resultant is equivalent to the vector sum of the three original voltages.
In order to determine what the resultant vector’s magnitude and angle are without re
sorting to graphic images, we can convert each one of these polarform complex numbers into
rectangular form and add. Remember, we’re adding these figures together because the polarity
marks for the three voltage sources are oriented in an additive manner:
2.8. SOME EXAMPLES WITH AC CIRCUITS 51
15
9.8298
9.6442
+ j6.8829 V
 j19.7735 V
+ j0 V
+
34.4740  j12.8906 V
15 V ∠ 0o = 15 + j0 V
12 V ∠ 35o = 9.8298 + j6.8829 V
22 V ∠ 64o = 9.6442  j19.7735 V
In polar form, this equates to 36.8052 volts 6 20.5018o. What this means in real terms
is that the voltage measured across these three voltage sources will be 36.8052 volts, lagging
the 15 volt (0o phase reference) by 20.5018o. A voltmeter connected across these points in
a real circuit would only indicate the polar magnitude of the voltage (36.8052 volts), not the
angle. An oscilloscope could be used to display two voltage waveforms and thus provide a phase
shift measurement, but not a voltmeter. The same principle holds true for AC ammeters: they
indicate the polar magnitude of the current, not the phase angle.
This is extremely important in relating calculated figures of voltage and current to real
circuits. Although rectangular notation is convenient for addition and subtraction, and was
indeed the final step in our sample problem here, it is not very applicable to practical measure
ments. Rectangular figures must be converted to polar figures (specifically polar magnitude)
before they can be related to actual circuit measurements.
We can use SPICE to verify the accuracy of our results. In this test circuit, the 10 kΩ
resistor value is quite arbitrary. It’s there so that SPICE does not declare an opencircuit
error and abort analysis. Also, the choice of frequencies for the simulation (60 Hz) is quite
arbitrary, because resistors respond uniformly for all frequencies of AC voltage and current.
There are other components (notably capacitors and inductors) which do not respond uniformly
to different frequencies, but that is another subject! (Figure 2.39)
ac voltage addition
v1 1 0 ac 15 0 sin
v2 2 1 ac 12 35 sin
v3 3 2 ac 22 64 sin
r1 3 0 10k
.ac lin 1 60 60 I’m using a frequency of 60 Hz
.print ac v(3,0) vp(3,0) as a default value
.end
freq v(3) vp(3)
6.000E+01 3.681E+01 2.050E+01
Sure enough, we get a total voltage of 36.81 volts 6 20.5o (with reference to the 15 volt
source, whose phase angle was arbitrarily stated at zero degrees so as to be the “reference”
52 CHAPTER 2. COMPLEX NUMBERS
+


+

+
3
2
1
0
3
0
V1
V2
V3
R1 10 kΩ
22 V ∠ 64o
12 V ∠ 35o
15 V ∠ 0o
Figure 2.39: Spice circuit schematic.
waveform).
At first glance, this is counterintuitive. How is it possible to obtain a total voltage of
just over 36 volts with 15 volt, 12 volt, and 22 volt supplies connected in series? With DC,
this would be impossible, as voltage figures will either directly add or subtract, depending on
polarity. But with AC, our “polarity” (phase shift) can vary anywhere in between fullaiding
and fullopposing, and this allows for such paradoxical summing.
What if we took the same circuit and reversed one of the supply’s connections? Its contri
bution to the total voltage would then be the opposite of what it was before: (Figure 2.40)
load
+


+

+
E1
E2
E3
Polarity reversed on
source E2 !
22 V ∠ 64o
12 V ∠ 35o
15 V ∠ 0o
Figure 2.40: Polarity of E2 (12V) is reversed.
Note how the 12 volt supply’s phase angle is still referred to as 35o, even though the leads
have been reversed. Remember that the phase angle of any voltage drop is stated in reference
to its noted polarity. Even though the angle is still written as 35o, the vector will be drawn
180o opposite of what it was before: (Figure 2.41)
The resultant (sum) vector should begin at the upperleft point (origin of the 22 volt vector)
2.8. SOME EXAMPLES WITH AC CIRCUITS 53
22 ∠ 64o
12 ∠ 35o (reversed) = 12 ∠ 215o
or
12 ∠ 35o
15 ∠ 0o
Figure 2.41: Direction of E2 is reversed.
and terminate at the right arrow tip of the 15 volt vector: (Figure 2.42)
resultant vector
22 ∠ 64o
12 ∠ 35o (reversed) = 12 ∠ 215o
or
12 ∠ 35ο
15 ∠ 0o
Figure 2.42: Resultant is vector sum of voltage sources.
The connection reversal on the 12 volt supply can be represented in two different ways in
54 CHAPTER 2. COMPLEX NUMBERS
polar form: by an addition of 180o to its vector angle (making it 12 volts 6 215o), or a reversal
of sign on the magnitude (making it 12 volts 6 35o). Either way, conversion to rectangular
form yields the same result:
(reversed) =
or
=
=
9.8298  j6.8829 V
9.8298  j6.8829 V
12 V ∠ 215o
12 V ∠ 35o
12 V ∠ 35o
The resulting addition of voltages in rectangular form, then:
15
9.6442  j19.7735 V
+ j0 V
+
9.8298  j6.8829 V
14.8143  j26.6564 V
In polar form, this equates to 30.4964 V 6 60.9368o. Once again, we will use SPICE to
verify the results of our calculations:
ac voltage addition
v1 1 0 ac 15 0 sin
v2 1 2 ac 12 35 sin Note the reversal of node numbers 2 and 1
v3 3 2 ac 22 64 sin to simulate the swapping of connections
r1 3 0 10k
.ac lin 1 60 60
.print ac v(3,0) vp(3,0)
.end
freq v(3) vp(3)
6.000E+01 3.050E+01 6.094E+01
• REVIEW:
• All the laws and rules of DC circuits apply to AC circuits, with the exception of power
calculations (Joule’s Law), so long as all values are expressed andmanipulated in complex
form, and all voltages and currents are at the same frequency.
• When reversing the direction of a vector (equivalent to reversing the polarity of an AC
voltage source in relation to other voltage sources), it can be expressed in either of two
different ways: adding 180o to the angle, or reversing the sign of the magnitude.
• Meter measurements in an AC circuit correspond to the polar magnitudes of calculated
values. Rectangular expressions of complex quantities in an AC circuit have no direct,
empirical equivalent, although they are convenient for performing addition and subtrac
tion, as Kirchhoff ’s Voltage and Current Laws require.
2.9. CONTRIBUTORS 55
2.9 Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
Jason Starck (June 2000): HTML document formatting, which led to a much better
looking second edition.
56 CHAPTER 2. COMPLEX NUMBERS
Chapter 3
REACTANCE AND IMPEDANCE
– INDUCTIVE
Contents
3.1 AC resistor circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
3.2 AC inductor circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.3 Series resistorinductor circuits . . . . . . . . . . . . . . . . . . . . . . . . . 64
3.4 Parallel resistorinductor circuits . . . . . . . . . . . . . . . . . . . . . . . . 71
3.5 Inductor quirks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.6 More on the “skin effect” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
3.7 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
3.1 AC resistor circuits
R
ET
ET I
IR
ER
ET = ER I = IR
0° IR
Figure 3.1: Pure resistive AC circuit: resistor voltage and current are in phase.
If we were to plot the current and voltage for a very simple AC circuit consisting of a source
and a resistor (Figure 3.1), it would look something like this: (Figure 3.2)
57
58 CHAPTER 3. REACTANCE AND IMPEDANCE – INDUCTIVE
Time
+

e =
i =
Figure 3.2: Voltage and current “in phase” for resistive circuit.
Because the resistor simply and directly resists the flow of electrons at all periods of time,
the waveform for the voltage drop across the resistor is exactly in phase with the waveform for
the current through it. We can look at any point in time along the horizontal axis of the plot
and compare those values of current and voltage with each other (any “snapshot” look at the
values of a wave are referred to as instantaneous values, meaning the values at that instant in
time). When the instantaneous value for current is zero, the instantaneous voltage across the
resistor is also zero. Likewise, at the moment in time where the current through the resistor
is at its positive peak, the voltage across the resistor is also at its positive peak, and so on. At
any given point in time along the waves, Ohm’s Law holds true for the instantaneous values of
voltage and current.
We can also calculate the power dissipated by this resistor, and plot those values on the
same graph: (Figure 3.3)
Time
+

e =
i =
p =
Figure 3.3: Instantaneous AC power in a pure resistive circuit is always positive.
Note that the power is never a negative value. When the current is positive (above the
line), the voltage is also positive, resulting in a power (p=ie) of a positive value. Conversely,
when the current is negative (below the line), the voltage is also negative, which results in a
positive value for power (a negative number multiplied by a negative number equals a positive
number). This consistent “polarity” of power tells us that the resistor is always dissipating
power, taking it from the source and releasing it in the form of heat energy. Whether the
3.2. AC INDUCTOR CIRCUITS 59
current is positive or negative, a resistor still dissipates energy.
3.2 AC inductor circuits
Inductors do not behave the same as resistors. Whereas resistors simply oppose the flow of
electrons through them (by dropping a voltage directly proportional to the current), inductors
oppose changes in current through them, by dropping a voltage directly proportional to the
rate of change of current. In accordance with Lenz’s Law, this induced voltage is always of such
a polarity as to try to maintain current at its present value. That is, if current is increasing
in magnitude, the induced voltage will “push against” the electron flow; if current is decreas
ing, the polarity will reverse and “push with” the electron flow to oppose the decrease. This
opposition to current change is called reactance, rather than resistance.
Expressed mathematically, the relationship between the voltage dropped across the induc
tor and rate of current change through the inductor is as such:
e = didtL
The expression di/dt is one from calculus, meaning the rate of change of instantaneous cur
rent (i) over time, in amps per second. The inductance (L) is in Henrys, and the instantaneous
voltage (e), of course, is in volts. Sometimes you will find the rate of instantaneous voltage
expressed as “v” instead of “e” (v = L di/dt), but it means the exact same thing. To show what
happens with alternating current, let’s analyze a simple inductor circuit: (Figure 3.4)
L
EL
IL
ET
IL
EL
I
ET = EL I = IL
90°
Figure 3.4: Pure inductive circuit: Inductor current lags inductor voltage by 90o.
If we were to plot the current and voltage for this very simple circuit, it would look some
thing like this: (Figure 3.5)
Remember, the voltage dropped across an inductor is a reaction against the change in cur
rent through it. Therefore, the instantaneous voltage is zero whenever the instantaneous
current is at a peak (zero change, or level slope, on the current sine wave), and the instan
taneous voltage is at a peak wherever the instantaneous current is at maximum change (the
points of steepest slope on the current wave, where it crosses the zero line). This results in a
voltage wave that is 90o out of phase with the current wave. Looking at the graph, the voltage
wave seems to have a “head start” on the current wave; the voltage “leads” the current, and
the current “lags” behind the voltage. (Figure 3.6)
Things get even more interesting when we plot the power for this circuit: (Figure 3.7)
Because instantaneous power is the product of the instantaneous voltage and the instanta
neous current (p=ie), the power equals zero whenever the instantaneous current or voltage is
60 CHAPTER 3. REACTANCE AND IMPEDANCE – INDUCTIVE
Time
+

e =
i =
Figure 3.5: Pure inductive circuit, waveforms.
Time
+

current slope = max. ()
voltage = max. ()
current slope = 0
voltage = 0
voltage = 0
e =
i =
current slope =
current slope = max. (+)
voltage = max. (+)
0
Figure 3.6: Current lags voltage by 90o in a pure inductive circuit.
Time
+

e =
i =
p =
Figure 3.7: In a pure inductive circuit, instantaneous power may be positive or negative
3.2. AC INDUCTOR CIRCUITS 61
zero. Whenever the instantaneous current and voltage are both positive (above the line), the
power is positive. As with the resistor example, the power is also positive when the instanta
neous current and voltage are both negative (below the line). However, because the current
and voltage waves are 90o out of phase, there are times when one is positive while the other is
negative, resulting in equally frequent occurrences of negative instantaneous power.
But what does negative power mean? It means that the inductor is releasing power back to
the circuit, while a positive power means that it is absorbing power from the circuit. Since the
positive and negative power cycles are equal in magnitude and duration over time, the inductor
releases just as much power back to the circuit as it absorbs over the span of a complete cycle.
What this means in a practical sense is that the reactance of an inductor dissipates a net
energy of zero, quite unlike the resistance of a resistor, which dissipates energy in the form of
heat. Mind you, this is for perfect inductors only, which have no wire resistance.
An inductor’s opposition to change in current translates to an opposition to alternating
current in general, which is by definition always changing in instantaneous magnitude and
direction. This opposition to alternating current is similar to resistance, but different in that
it always results in a phase shift between current and voltage, and it dissipates zero power.
Because of the differences, it has a different name: reactance. Reactance to AC is expressed
in ohms, just like resistance is, except that its mathematical symbol is X instead of R. To be
specific, reactance associate with an inductor is usually symbolized by the capital letter X with
a letter L as a subscript, like this: XL.
Since inductors drop voltage in proportion to the rate of current change, they will drop more
voltage for fasterchanging currents, and less voltage for slowerchanging currents. What this
means is that reactance in ohms for any inductor is directly proportional to the frequency of
the alternating current. The exact formula for determining reactance is as follows:
XL = 2pifL
If we expose a 10 mH inductor to frequencies of 60, 120, and 2500 Hz, it will manifest the
reactances in Table Figure 3.1.
Table 3.1: Reactance of a 10 mH inductor:
Frequency (Hertz) Reactance (Ohms)
60 3.7699
120 7.5398
2500 157.0796
In the reactance equation, the term “2pif” (everything on the righthand side except the L)
has a special meaning unto itself. It is the number of radians per second that the alternating
current is “rotating” at, if you imagine one cycle of AC to represent a full circle’s rotation.
A radian is a unit of angular measurement: there are 2pi radians in one full circle, just as
there are 360o in a full circle. If the alternator producing the AC is a doublepole unit, it will
produce one cycle for every full turn of shaft rotation, which is every 2pi radians, or 360o. If
this constant of 2pi is multiplied by frequency in Hertz (cycles per second), the result will be a
figure in radians per second, known as the angular velocity of the AC system.
Angular velocity may be represented by the expression 2pif, or it may be represented by its
own symbol, the lowercase Greek letter Omega, which appears similar to our Roman lower
case “w”: ω. Thus, the reactance formula XL = 2pifL could also be written as XL = ωL.
62 CHAPTER 3. REACTANCE AND IMPEDANCE – INDUCTIVE
It must be understood that this “angular velocity” is an expression of how rapidly the AC
waveforms are cycling, a full cycle being equal to 2pi radians. It is not necessarily representa
tive of the actual shaft speed of the alternator producing the AC. If the alternator has more
than two poles, the angular velocity will be a multiple of the shaft speed. For this reason, ω is
sometimes expressed in units of electrical radians per second rather than (plain) radians per
second, so as to distinguish it from mechanical motion.
Any way we express the angular velocity of the system, it is apparent that it is directly pro
portional to reactance in an inductor. As the frequency (or alternator shaft speed) is increased
in an AC system, an inductor will offer greater opposition to the passage of current, and vice
versa. Alternating current in a simple inductive circuit is equal to the voltage (in volts) divided
by the inductive reactance (in ohms), just as either alternating or direct current in a simple re
sistive circuit is equal to the voltage (in volts) divided by the resistance (in ohms). An example
circuit is shown here: (Figure 3.8)
L 10 mH10 V
60 Hz
Figure 3.8: Inductive reactance
XL = 3.7699 Ω
(inductive reactance of 10 mH inductor at 60 Hz)
I = E
X
I = 10 V
3.7699 Ω
I = 2.6526 A
However, we need to keep in mind that voltage and current are not in phase here. As was
shown earlier, the voltage has a phase shift of +90o with respect to the current. (Figure 3.9) If
we represent these phase angles of voltage and current mathematically in the form of complex
numbers, we find that an inductor’s opposition to current has a phase angle, too:
3.2. AC INDUCTOR CIRCUITS 63
Opposition =
Current
Voltage
Opposition = 10 V ∠ 90
o
2.6526 A ∠ 0ο
Opposition = 3.7699 Ω ∠ 90o
or
0 + j3.7699 Ω
Opposition
For an inductor:
90o
0o
90o
(XL)
E
I
Figure 3.9: Current lags voltage by 90o in an inductor.
Mathematically, we say that the phase angle of an inductor’s opposition to current is 90o,
meaning that an inductor’s opposition to current is a positive imaginary quantity. This phase
angle of reactive opposition to current becomes critically important in circuit analysis, espe
cially for complex AC circuits where reactance and resistance interact. It will prove beneficial
to represent any component’s opposition to current in terms of complex numbers rather than
scalar quantities of resistance and reactance.
• REVIEW:
• Inductive reactance is the opposition that an inductor offers to alternating current due
to its phaseshifted storage and release of energy in its magnetic field. Reactance is
symbolized by the capital letter “X” and is measured in ohms just like resistance (R).
• Inductive reactance can be calculated using this formula: XL = 2pifL
• The angular velocity of an AC circuit is another way of expressing its frequency, in units
of electrical radians per second instead of cycles per second. It is symbolized by the lower
case Greek letter “omega,” or ω.
64 CHAPTER 3. REACTANCE AND IMPEDANCE – INDUCTIVE
• Inductive reactance increases with increasing frequency. In other words, the higher the
frequency, the more it opposes the AC flow of electrons.
3.3 Series resistorinductor circuits
In the previous section, we explored what would happen in simple resistoronly and inductor
only AC circuits. Now we will mix the two components together in series form and investigate
the effects.
Take this circuit as an example to work with: (Figure 3.10)
R
ET
IL
EL L
I
= IR = IL
ET = ER+ EL
ER
37°
R
L
5 Ω
10
mH10 V60 Hz
ET
IR
EL
I
I
Figure 3.10: Series resistor inductor circuit: Current lags applied voltage by 0o to 90o.
The resistor will offer 5 Ω of resistance to AC current regardless of frequency, while the
inductor will offer 3.7699 Ω of reactance to AC current at 60 Hz. Because the resistor’s re
sistance is a real number (5 Ω 6 0o, or 5 + j0 Ω), and the inductor’s reactance is an imaginary
number (3.7699 Ω 6 90o, or 0 + j3.7699 Ω), the combined effect of the two components will be an
opposition to current equal to the complex sum of the two numbers. This combined opposition
will be a vector combination of resistance and reactance. In order to express this opposition
succinctly, we need a more comprehensive term for opposition to current than either resistance
or reactance alone. This term is called impedance, its symbol is Z, and it is also expressed in
the unit of ohms, just like resistance and reactance. In the above example, the total circuit
impedance is:
Ztotal = (5 Ω resistance) + (3.7699 Ω inductive reactance)
Ztotal = 5 Ω (R) + 3.7699 Ω (XL)
Ztotal = (5 Ω ∠ 0o) + (3.7699 Ω ∠ 900)
or
(5 + j0 Ω) + (0 + j3.7699 Ω)
Ztotal = 5 + j3.7699 Ω or 6.262 Ω ∠ 37.016o
3.3. SERIES RESISTORINDUCTOR CIRCUITS 65
Impedance is related to voltage and current just as you might expect, in a manner similar
to resistance in Ohm’s Law:
Ohm’s Law for AC circuits:
E = IZ I = Z =E
Z
E
I
All quantities expressed in
complex, not scalar, form
In fact, this is a far more comprehensive form of Ohm’s Law than what was taught in DC
electronics (E=IR), just as impedance is a far more comprehensive expression of opposition to
the flow of electrons than resistance is. Any resistance and any reactance, separately or in
combination (series/parallel), can be and should be represented as a single impedance in an
AC circuit.
To calculate current in the above circuit, we first need to give a phase angle reference for
the voltage source, which is generally assumed to be zero. (The phase angles of resistive and
inductive impedance are always 0o and +90o, respectively, regardless of the given phase angles
for voltage or current).
I = E
Z
I = 10 V ∠ 0
o
6.262 Ω ∠ 37.016o
I = 1.597 A ∠ 37.016o
As with the purely inductive circuit, the current wave lags behind the voltage wave (of the
source), although this time the lag is not as great: only 37.016o as opposed to a full 90o as was
the case in the purely inductive circuit. (Figure 3.11)
Time
+

e =
i =
phase shift =
37.016o
Figure 3.11: Current lags voltage in a series LR circuit.
For the resistor and the inductor, the phase relationships between voltage and current
haven’t changed. Voltage across the resistor is in phase (0o shift) with the current through
66 CHAPTER 3. REACTANCE AND IMPEDANCE – INDUCTIVE
it; and the voltage across the inductor is +90o out of phase with the current going through it.
We can verify this mathematically:
E = IZ
ER = IRZR
ER = (1.597 A ∠ 37.016o)(5 Ω ∠ 0o)
ER = 7.9847 V ∠ 37.016o
Notice that the phase angle of ER is equal to
the phase angle of the current.
The voltage across the resistor has the exact same phase angle as the current through it,
telling us that E and I are in phase (for the resistor only).
E = IZ
EL = ILZL
EL = (1.597 A ∠ 37.016o)(3.7699 Ω ∠ 90o)
EL = 6.0203 V ∠ 52.984o
Notice that the phase angle of EL is exactly
90o more than the phase angle of the current.
The voltage across the inductor has a phase angle of 52.984o, while the current through the
inductor has a phase angle of 37.016o, a difference of exactly 90o between the two. This tells
us that E and I are still 90o out of phase (for the inductor only).
We can also mathematically prove that these complex values add together to make the total
voltage, just as Kirchhoff ’s Voltage Law would predict:
Etotal = ER + EL
Etotal = (7.9847 V ∠ 37.016o) + (6.0203 V ∠ 52.984o)
Etotal = 10 V ∠ 0o
Let’s check the validity of our calculations with SPICE: (Figure 3.12)
3.3. SERIES RESISTORINDUCTOR CIRCUITS 67
1
0
2
0
R
L
5 Ω
10 mH10 V
60 Hz
Figure 3.12: Spice circuit: RL.
ac rl circuit
v1 1 0 ac 10 sin
r1 1 2 5
l1 2 0 10m
.ac lin 1 60 60
.print ac v(1,2) v(2,0) i(v1)
.print ac vp(1,2) vp(2,0) ip(v1)
.end
freq v(1,2) v(2) i(v1)
6.000E+01 7.985E+00 6.020E+00 1.597E+00
freq vp(1,2) vp(2) ip(v1)
6.000E+01 3.702E+01 5.298E+01 1.430E+02
ER = 7.985 V ∠ 37.02o
EL = 6.020 V ∠ 52.98o
I = 1.597 A ∠ 143.0o
Interpreted SPICE results
Note that just as with DC circuits, SPICE outputs current figures as though they were
negative (180o out of phase) with the supply voltage. Instead of a phase angle of 37.016o,
we get a current phase angle of 143o (37o + 180o). This is merely an idiosyncrasy of SPICE
and does not represent anything significant in the circuit simulation itself. Note how both
the resistor and inductor voltage phase readings match our calculations (37.02o and 52.98o,
respectively), just as we expected them to.
With all these figures to keep track of for even such a simple circuit as this, it would be
beneficial for us to use the “table” method. Applying a table to this simple series resistor
inductor circuit would proceed as such. First, draw up a table for E/I/Z figures and insert all
component values in these terms (in other words, don’t insert actual resistance or inductance
values in Ohms and Henrys, respectively, into the table; rather, convert them into complex
figures of impedance and write those in):
68 CHAPTER 3. REACTANCE AND IMPEDANCE – INDUCTIVE
E
I
Volts
Amps
OhmsZ
R L Total
5 + j0
5 ∠ 0o
0 + j3.7699
3.7699 ∠ 90o
10 + j0
10 ∠ 0o
Although it isn’t necessary, I find it helpful to write both the rectangular and polar forms of
each quantity in the table. If you are using a calculator that has the ability to perform complex
arithmetic without the need for conversion between rectangular and polar forms, then this
extra documentation is completely unnecessary. However, if you are forced to perform complex
arithmetic “longhand” (addition and subtraction in rectangular form, and multiplication and
division in polar form), writing each quantity in both forms will be useful indeed.
Now that our “given” figures are inserted into their respective locations in the table, we can
proceed just as with DC: determine the total impedance from the individual impedances. Since
this is a series circuit, we know that opposition to electron flow (resistance or impedance) adds
to form the total opposition:
E
I
Volts
Amps
OhmsZ
R L Total
5 + j0
5 ∠ 0o
0 + j3.7699
3.7699 ∠ 90o
10 + j0
10 ∠ 0o
5 + j3.7699
6.262 ∠ 37.016o
Rule of series
circuits
Ztotal = ZR + ZL
Now that we know total voltage and total impedance, we can apply Ohm’s Law (I=E/Z) to
determine total current:
3.3. SERIES RESISTORINDUCTOR CIRCUITS 69
E
I
Volts
Amps
OhmsZ
R L Total
5 + j0
5 ∠ 0o
0 + j3.7699
3.7699 ∠ 90o
10 + j0
10 ∠ 0o
5 + j3.7699
6.262 ∠ 37.016o
Ohm’s
Law
1.597 ∠ 37.016o
1.2751  j0.9614
I = E
Z
Just as with DC, the total current in a series AC circuit is shared equally by all components.
This is still true because in a series circuit there is only a single path for electrons to flow,
therefore the rate of their flow must uniform throughout. Consequently, we can transfer the
figures for current into the columns for the resistor and inductor alike:
E
I
Volts
Amps
OhmsZ
R L Total
5 + j0
5 ∠ 0o
0 + j3.7699
3.7699 ∠ 90o
10 + j0
10 ∠ 0o
5 + j3.7699
6.262 ∠ 37.016o
1.597 ∠ 37.016o
Rule of series
circuits:
Itotal = IR = IL
1.597 ∠ 37.016o1.597 ∠ 37.016o
1.2751  j0.96141.2751  j0.96141.2751  j0.9614
Now all that’s left to figure is the voltage drop across the resistor and inductor, respectively.
This is done through the use of Ohm’s Law (E=IZ), applied vertically in each column of the
table:
70 CHAPTER 3. REACTANCE AND IMPEDANCE – INDUCTIVE
E
I
Volts
Amps
OhmsZ
R L Total
5 + j0
5 ∠ 0o
0 + j3.7699
3.7699 ∠ 90o
10 + j0
10 ∠ 0o
5 + j3.7699
6.262 ∠ 37.016o
1.597 ∠ 37.016o1.597 ∠ 37.016o1.597 ∠ 37.016o
Ohm’s
Law
Ohm’s
Law
6.3756  j4.8071
7.9847 ∠ 37.016o
3.6244 + j4.8071
6.0203 ∠ 52.984o
1.2751  j0.96141.2751  j0.96141.2751  j0.9614
E = IZ E = IZ
And with that, our table is complete. The exact same rules we applied in the analysis of DC
circuits apply to AC circuits as well, with the caveat that all quantities must be represented
and calculated in complex rather than scalar form. So long as phase shift is properly repre
sented in our calculations, there is no fundamental difference in how we approach basic AC
circuit analysis versus DC.
Now is a good time to review the relationship between these calculated figures and read
ings given by actual instrument measurements of voltage and current. The figures here that
directly relate to reallife measurements are those in polar notation, not rectangular! In other
words, if you were to connect a voltmeter across the resistor in this circuit, it would indicate
7.9847 volts, not 6.3756 (real rectangular) or 4.8071 (imaginary rectangular) volts. To describe
this in graphical terms, measurement instruments simply tell you how long the vector is for
that particular quantity (voltage or current).
Rectangular notation, while convenient for arithmetical addition and subtraction, is a more
abstract form of notation than polar in relation to realworld measurements. As I stated before,
I will indicate both polar and rectangular forms of each quantity in my AC circuit tables simply
for convenience of mathematical calculation. This is not absolutely necessary, but may be
helpful for those following along without the benefit of an advanced calculator. If we were to
restrict ourselves to the use of only one form of notation, the best choice would be polar, because
it is the only one that can be directly correlated to real measurements.
Impedance (Z) of a series RL circuit may be calculated, given the resistance (R) and the
inductive reactance (XL). Since E=IR, E=IXL, and E=IZ, resistance, reactance, and impedance
are proportional to voltage, respectively. Thus, the voltage phasor diagram can be replaced by
a similar impedance diagram. (Figure 3.13)
Example:
Given: A 40 Ω resistor in series with a 79.58 millihenry inductor. Find the impedance at 60
hertz.
XL = 2pifL
XL = 2pi·60·79.58×10−3
XL = 30 Ω
Z = R + jXL
Z = 40 + j30
3.4. PARALLEL RESISTORINDUCTOR CIRCUITS 71
R
ET
IL
EL
ER
37°
ET
IR
EL
I
I
R
37°
ZXL
XL
Voltage Impedance
Figure 3.13: Series: RL circuit Impedance phasor diagram.
Z = sqrt(402 + 302) = 50 Ω
6 Z = arctangent(30/40) = 36.87o
Z = 40 + j30 = 506 36.87o
• REVIEW:
• Impedance is the total measure of opposition to electric current and is the complex (vec
tor) sum of (“real”) resistance and (“imaginary”) reactance. It is symbolized by the letter
“Z” and measured in ohms, just like resistance (R) and reactance (X).
• Impedances (Z) are managed just like resistances (R) in series circuit analysis: series
impedances add to form the total impedance. Just be sure to perform all calculations in
complex (not scalar) form! ZTotal = Z1 + Z2 + . . . Zn
• A purely resistive impedance will always have a phase angle of exactly 0o (ZR = R Ω 6
0o).
• A purely inductive impedance will always have a phase angle of exactly +90o (ZL = XL Ω
6 90o).
• Ohm’s Law for AC circuits: E = IZ ; I = E/Z ; Z = E/I
• When resistors and inductors are mixed together in circuits, the total impedance will
have a phase angle somewhere between 0o and +90o. The circuit current will have a
phase angle somewhere between 0o and 90o.
• Series AC circuits exhibit the same fundamental properties as series DC circuits: cur
rent is uniform throughout the circuit, voltage drops add to form the total voltage, and
impedances add to form the total impedance.
3.4 Parallel resistorinductor circuits
Let’s take the same components for our series example circuit and connect them in parallel:
(Figure 3.14)
Because the power source has the same frequency as the series example circuit, and the
resistor and inductor both have the same values of resistance and inductance, respectively,
72 CHAPTER 3. REACTANCE AND IMPEDANCE – INDUCTIVE
R L5 Ω 10
mH
10 V
60 Hz
R L
53°E
IIL
IR
IR IL
E
= ER = EL
I
= IR+ IL
E
I
Figure 3.14: Parallel RL circuit.
they must also have the same values of impedance. So, we can begin our analysis table with
the same “given” values:
E
I
Volts
Amps
OhmsZ
R L Total
5 + j0
5 ∠ 0o
0 + j3.7699
3.7699 ∠ 90o
10 + j0
10 ∠ 0o
The only difference in our analysis technique this time is that we will apply the rules of
parallel circuits instead of the rules for series circuits. The approach is fundamentally the
same as for DC. We know that voltage is shared uniformly by all components in a parallel
circuit, so we can transfer the figure of total voltage (10 volts 6 0o) to all components columns:
E
I
Volts
Amps
OhmsZ
R L Total
5 + j0
5 ∠ 0o
0 + j3.7699
3.7699 ∠ 90o
10 + j0
10 ∠ 0o
10 + j0
10 ∠ 0o
10 + j0
10 ∠ 0o
Rule of parallel
circuits:
Etotal = ER = EL
Now we can apply Ohm’s Law (I=E/Z) vertically to two columns of the table, calculating
current through the resistor and current through the inductor:
3.4. PARALLEL RESISTORINDUCTOR CIRCUITS 73
E
I
Volts
Amps
OhmsZ
R L Total
5 + j0
5 ∠ 0o
0 + j3.7699
3.7699 ∠ 90o
10 + j0
10 ∠ 0o
10 + j0
10 ∠ 0o
10 + j0
10 ∠ 0o
0  j2.6526
2.6526 ∠ 90o
2 + j0
2 ∠ 0o
Ohm’s
Law
Ohm’s
Law
I = E
Z
I = E
Z
Just as with DC circuits, branch currents in a parallel AC circuit add to form the total
current (Kirchhoff ’s Current Law still holds true for AC as it did for DC):
E
I
Volts
Amps
OhmsZ
R L Total
5 + j0
5 ∠ 0o
0 + j3.7699
3.7699 ∠ 90o
10 + j0
10 ∠ 0o
10 + j0
10 ∠ 0o
10 + j0
10 ∠ 0o
0  j2.6526
2.6526 ∠ 90o
2 + j0
2 ∠ 0o
2  j2.6526
Rule of parallel
circuits:
Itotal = IR + IL
3.3221 ∠ 52.984o
Finally, total impedance can be calculated by using Ohm’s Law (Z=E/I) vertically in the
“Total” column. Incidentally, parallel impedance can also be calculated by using a reciprocal
formula identical to that used in calculating parallel resistances.
Zparallel =
Z1 Z2 Zn
1
+
1
+ . . .
1
1
The only problem with using this formula is that it typically involves a lot of calculator
keystrokes to carry out. And if you’re determined to run through a formula like this “longhand,”
be prepared for a very large amount of work! But, just as with DC circuits, we often have
multiple options in calculating the quantities in our analysis tables, and this example is no
different. No matter which way you calculate total impedance (Ohm’s Law or the reciprocal
formula), you will arrive at the same figure:
74 CHAPTER 3. REACTANCE AND IMPEDANCE – INDUCTIVE
E
I
Volts
Amps
OhmsZ
R L Total
5 + j0
5 ∠ 0o
0 + j3.7699
3.7699 ∠ 90o
10 + j0
10 ∠ 0o
10 + j0
10 ∠ 0o
10 + j0
10 ∠ 0o
0  j2.6526
2.6526 ∠ 90o
2 + j0
2 ∠ 0o
2  j2.6526
3.322 ∠ 52.984o
1.8122 + j2.4035
3.0102 ∠ 52.984o
Ohm’s
Law or
Rule of parallel
circuits:
Ztotal =
ZR ZL
1
+
1
1Z = E
I
• REVIEW:
• Impedances (Z) are managed just like resistances (R) in parallel circuit analysis: parallel
impedances diminish to form the total impedance, using the reciprocal formula. Just be
sure to perform all calculations in complex (not scalar) form! ZTotal = 1/(1/Z1 + 1/Z2 + . . .
1/Zn)
• Ohm’s Law for AC circuits: E = IZ ; I = E/Z ; Z = E/I
• When resistors and inductors are mixed together in parallel circuits (just as in series
circuits), the total impedance will have a phase angle somewhere between 0o and +90o.
The circuit current will have a phase angle somewhere between 0o and 90o.
• Parallel AC circuits exhibit the same fundamental properties as parallel DC circuits:
voltage is uniform throughout the circuit, branch currents add to form the total current,
and impedances diminish (through the reciprocal formula) to form the total impedance.
3.5 Inductor quirks
In an ideal case, an inductor acts as a purely reactive device. That is, its opposition to AC
current is strictly based on inductive reaction to changes in current, and not electron friction as
is the case with resistive components. However, inductors are not quite so pure in their reactive
behavior. To begin with, they’re made of wire, and we know that all wire possesses some
measurable amount of resistance (unless its superconducting wire). This builtin resistance
acts as though it were connected in series with the perfect inductance of the coil, like this:
(Figure 3.15)
Consequently, the impedance of any real inductor will always be a complex combination of
resistance and inductive reactance.
Compounding this problem is something called the skin effect, which is AC’s tendency to
flow through the outer areas of a conductor’s crosssection rather than through the middle.
3.5. INDUCTOR QUIRKS 75
Wire resistance
Ideal inductor
Equivalent circuit for a real inductor
R
L
Figure 3.15: Inductor Equivalent circuit of a real inductor.
When electrons flow in a single direction (DC), they use the entire crosssectional area of the
conductor to move. Electrons switching directions of flow, on the other hand, tend to avoid
travel through the very middle of a conductor, limiting the effective crosssectional area avail
able. The skin effect becomes more pronounced as frequency increases.
Also, the alternating magnetic field of an inductor energized with AC may radiate off into
space as part of an electromagnetic wave, especially if the AC is of high frequency. This ra
diated energy does not return to the inductor, and so it manifests itself as resistance (power
dissipation) in the circuit.
Added to the resistive losses of wire and radiation, there are other effects at work in iron
core inductors which manifest themselves as additional resistance between the leads. When
an inductor is energized with AC, the alternating magnetic fields produced tend to induce
circulating currents within the iron core known as eddy currents. These electric currents in
the iron core have to overcome the electrical resistance offered by the iron, which is not as
good a conductor as copper. Eddy current losses are primarily counteracted by dividing the
iron core up into many thin sheets (laminations), each one separated from the other by a
thin layer of electrically insulating varnish. With the crosssection of the core divided up into
many electrically isolated sections, current cannot circulate within that crosssectional area
and there will be no (or very little) resistive losses from that effect.
As you might have expected, eddy current losses in metallic inductor cores manifest them
selves in the form of heat. The effect is more pronounced at higher frequencies, and can be so
extreme that it is sometimes exploited in manufacturing processes to heat metal objects! In
fact, this process of “inductive heating” is often used in highpurity metal foundry operations,
where metallic elements and alloys must be heated in a vacuum environment to avoid contam
ination by air, and thus where standard combustion heating technology would be useless. It is
a “noncontact” technology, the heated substance not having to touch the coil(s) producing the
magnetic field.
In highfrequency service, eddy currents can even develop within the crosssection of the
wire itself, contributing to additional resistive effects. To counteract this tendency, special
76 CHAPTER 3. REACTANCE AND IMPEDANCE – INDUCTIVE
wire made of very fine, individually insulated strands called Litz wire (short for Litzendraht)
can be used. The insulation separating strands from each other prevent eddy currents from
circulating through the whole wire’s crosssectional area.
Additionally, any magnetic hysteresis that needs to be overcome with every reversal of the
inductor’s magnetic field constitutes an expenditure of energy that manifests itself as resis
tance in the circuit. Some core materials (such as ferrite) are particularly notorious for their
hysteretic effect. Counteracting this effect is best done by means of proper core material selec
tion and limits on the peak magnetic field intensity generated with each cycle.
Altogether, the stray resistive properties of a real inductor (wire resistance, radiation losses,
eddy currents, and hysteresis losses) are expressed under the single term of “effective resis
tance:” (Figure 3.16)
Ideal inductor
Equivalent circuit for a real inductor
R
L
"Effective" resistance
Figure 3.16: Equivalent circuit of a real inductor with skineffect, radiation, eddy current, and
hysteresis losses.
It is worthy to note that the skin effect and radiation losses apply just as well to straight
lengths of wire in an AC circuit as they do a coiled wire. Usually their combined effect is too
small to notice, but at radio frequencies they can be quite large. A radio transmitter antenna,
for example, is designed with the express purpose of dissipating the greatest amount of energy
in the form of electromagnetic radiation.
Effective resistance in an inductor can be a serious consideration for the AC circuit designer.
To help quantify the relative amount of effective resistance in an inductor, another value exists
called the Q factor, or “quality factor” which is calculated as follows:
Q = XL
R
The symbol “Q” has nothing to do with electric charge (coulombs), which tends to be con
fusing. For some reason, the Powers That Be decided to use the same letter of the alphabet to
denote a totally different quantity.
The higher the value for “Q,” the “purer” the inductor is. Because its so easy to add ad
ditional resistance if needed, a highQ inductor is better than a lowQ inductor for design
3.6. MORE ON THE “SKIN EFFECT” 77
purposes. An ideal inductor would have a Q of infinity, with zero effective resistance.
Because inductive reactance (X) varies with frequency, so will Q. However, since the resis
tive effects of inductors (wire skin effect, radiation losses, eddy current, and hysteresis) also
vary with frequency, Q does not vary proportionally with reactance. In order for a Q value to
have precise meaning, it must be specified at a particular test frequency.
Stray resistance isn’t the only inductor quirk we need to be aware of. Due to the fact that the
multiple turns of wire comprising inductors are separated from each other by an insulating gap
(air, varnish, or some other kind of electrical insulation), we have the potential for capacitance
to develop between turns. AC capacitance will be explored in the next chapter, but it suffices
to say at this point that it behaves very differently from AC inductance, and therefore further
“taints” the reactive purity of real inductors.
3.6 More on the “skin effect”
As previously mentioned, the skin effect is where alternating current tends to avoid travel
through the center of a solid conductor, limiting itself to conduction near the surface. This
effectively limits the crosssectional conductor area available to carry alternating electron flow,
increasing the resistance of that conductor above what it would normally be for direct current:
(Figure 3.17)
Crosssectional area of a round
conductor available for conducting
DC current
"DC resistance"
Crosssectional area of the same
conductor available for conducting
lowfrequency AC
"AC resistance"
Crosssectional area of the same
conductor available for conducting
highfrequency AC
"AC resistance"
Figure 3.17: Skin effect: skin depth decreases with increasing frequency.
78 CHAPTER 3. REACTANCE AND IMPEDANCE – INDUCTIVE
The electrical resistance of the conductor with all its crosssectional area in use is known
as the “DC resistance,” the “AC resistance” of the same conductor referring to a higher figure
resulting from the skin effect. As you can see, at high frequencies the AC current avoids travel
through most of the conductor’s crosssectional area. For the purpose of conducting current,
the wire might as well be hollow!
In some radio applications (antennas, most notably) this effect is exploited. Since radio
frequency (“RF”) AC currents wouldn’t travel through the middle of a conductor anyway, why
not just use hollow metal rods instead of solid metal wires and save both weight and cost?
(Figure 3.18) Most antenna structures and RF power conductors are made of hollow metal
tubes for this reason.
In the following photograph you can see some large inductors used in a 50 kW radio trans
mitting circuit. The inductors are hollow copper tubes coated with silver, for excellent conduc
tivity at the “skin” of the tube:
Figure 3.18: High power inductors formed from hollow tubes.
The degree to which frequency affects the effective resistance of a solid wire conductor is
impacted by the gauge of that wire. As a rule, largegauge wires exhibit a more pronounced
3.7. CONTRIBUTORS 79
skin effect (change in resistance from DC) than smallgauge wires at any given frequency. The
equation for approximating skin effect at high frequencies (greater than 1 MHz) is as follows:
RAC = (RDC)(k) f
Where,
RAC =
RDC =
k =
f =
AC resistance at given frequency "f"
Resistance at zero frequency (DC)
Wire gage factor (see table below)
Frequency of AC in MHz (MegaHertz)
Table 3.2 gives approximate values of “k” factor for various round wire sizes.
Table 3.2: “k” factor for various AWG wire sizes.
gage size k factor gage size k factor
4/0 124.5 8 34.8
2/0 99.0 10 27.6
1/0 88.0 14 17.6
2 69.8 18 10.9
4 55.5 22 6.86
6 47.9  
For example, a length of number 10gauge wire with a DC endtoend resistance of 25 Ω
would have an AC (effective) resistance of 2.182 kΩ at a frequency of 10 MHz:
RAC = (RDC)(k) f
RAC = (25 Ω)(27.6) 10
RAC = 2.182 kΩ
Please remember that this figure is not impedance, and it does not consider any reactive
effects, inductive or capacitive. This is simply an estimated figure of pure resistance for the
conductor (that opposition to the AC flow of electrons which does dissipate power in the form
of heat), corrected for the skin effect. Reactance, and the combined effects of reactance and
resistance (impedance), are entirely different matters.
3.7 Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
80 CHAPTER 3. REACTANCE AND IMPEDANCE – INDUCTIVE
Jim Palmer (June 2001): Identified and offered correction for typographical error in com
plex number calculation.
Jason Starck (June 2000): HTML document formatting, which led to a much better
looking second edition.
Chapter 4
REACTANCE AND IMPEDANCE
– CAPACITIVE
Contents
4.1 AC resistor circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
4.2 AC capacitor circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.3 Series resistorcapacitor circuits . . . . . . . . . . . . . . . . . . . . . . . . . 87
4.4 Parallel resistorcapacitor circuits . . . . . . . . . . . . . . . . . . . . . . . . 92
4.5 Capacitor quirks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
4.6 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4.1 AC resistor circuits
R
ET
ET I
IR
ER
ET = ER I = IR
0° IR
Figure 4.1: Pure resistive AC circuit: voltage and current are in phase.
If we were to plot the current and voltage for a very simple AC circuit consisting of a source
and a resistor, (Figure 4.1) it would look something like this: (Figure 4.2)
Because the resistor allows an amount of current directly proportional to the voltage across
it at all periods of time, the waveform for the current is exactly in phase with the waveform for
81
82 CHAPTER 4. REACTANCE AND IMPEDANCE – CAPACITIVE
Time
+

e =
i =
Figure 4.2: Voltage and current “in phase” for resistive circuit.
the voltage. We can look at any point in time along the horizontal axis of the plot and compare
those values of current and voltage with each other (any “snapshot” look at the values of a wave
are referred to as instantaneous values, meaning the values at that instant in time). When the
instantaneous value for voltage is zero, the instantaneous current through the resistor is also
zero. Likewise, at the moment in time where the voltage across the resistor is at its positive
peak, the current through the resistor is also at its positive peak, and so on. At any given point
in time along the waves, Ohm’s Law holds true for the instantaneous values of voltage and
current.
We can also calculate the power dissipated by this resistor, and plot those values on the
same graph: (Figure 4.3)
Time
+

e =
i =
p =
Figure 4.3: Instantaneous AC power in a resistive circuit is always positive.
Note that the power is never a negative value. When the current is positive (above the
line), the voltage is also positive, resulting in a power (p=ie) of a positive value. Conversely,
when the current is negative (below the line), the voltage is also negative, which results in a
positive value for power (a negative number multiplied by a negative number equals a positive
number). This consistent “polarity” of power tells us that the resistor is always dissipating
power, taking it from the source and releasing it in the form of heat energy. Whether the
current is positive or negative, a resistor still dissipates energy.
4.2. AC CAPACITOR CIRCUITS 83
4.2 AC capacitor circuits
Capacitors do not behave the same as resistors. Whereas resistors allow a flow of electrons
through them directly proportional to the voltage drop, capacitors oppose changes in voltage
by drawing or supplying current as they charge or discharge to the new voltage level. The flow
of electrons “through” a capacitor is directly proportional to the rate of change of voltage across
the capacitor. This opposition to voltage change is another form of reactance, but one that is
precisely opposite to the kind exhibited by inductors.
Expressed mathematically, the relationship between the current “through” the capacitor
and rate of voltage change across the capacitor is as such:
i = C de
dt
The expression de/dt is one from calculus, meaning the rate of change of instantaneous
voltage (e) over time, in volts per second. The capacitance (C) is in Farads, and the instan
taneous current (i), of course, is in amps. Sometimes you will find the rate of instantaneous
voltage change over time expressed as dv/dt instead of de/dt: using the lowercase letter “v”
instead or “e” to represent voltage, but it means the exact same thing. To show what happens
with alternating current, let’s analyze a simple capacitor circuit: (Figure 4.4)
C
ET I
VC
IC
ET = EC I = IC
EC
IC
90°
Figure 4.4: Pure capacitive circuit: capacitor voltage lags capacitor current by 90o
If we were to plot the current and voltage for this very simple circuit, it would look some
thing like this: (Figure 4.5)
Time
+

e =
i =
Figure 4.5: Pure capacitive circuit waveforms.
Remember, the current through a capacitor is a reaction against the change in voltage
across it. Therefore, the instantaneous current is zero whenever the instantaneous voltage is
84 CHAPTER 4. REACTANCE AND IMPEDANCE – CAPACITIVE
at a peak (zero change, or level slope, on the voltage sine wave), and the instantaneous current
is at a peak wherever the instantaneous voltage is at maximum change (the points of steepest
slope on the voltage wave, where it crosses the zero line). This results in a voltage wave that
is 90o out of phase with the current wave. Looking at the graph, the current wave seems to
have a “head start” on the voltage wave; the current “leads” the voltage, and the voltage “lags”
behind the current. (Figure 4.6)
Time
+

e =
i =
voltage slope = 0
current = 0
voltage slope = max. ()
current = max. ()
voltage slope = 0
current = 0
voltage slope = max. (+)
current = max. (+)
Figure 4.6: Voltage lags current by 90o in a pure capacitive circuit.
As you might have guessed, the same unusual power wave that we saw with the simple
inductor circuit is present in the simple capacitor circuit, too: (Figure 4.7)
Time
+

e =
i =
p =
Figure 4.7: In a pure capacitive circuit, the instantaneous power may be positive or negative.
As with the simple inductor circuit, the 90 degree phase shift between voltage and current
results in a power wave that alternates equally between positive and negative. This means
that a capacitor does not dissipate power as it reacts against changes in voltage; it merely
absorbs and releases power, alternately.
4.2. AC CAPACITOR CIRCUITS 85
A capacitor’s opposition to change in voltage translates to an opposition to alternating volt
age in general, which is by definition always changing in instantaneous magnitude and direc
tion. For any given magnitude of AC voltage at a given frequency, a capacitor of given size will
“conduct” a certain magnitude of AC current. Just as the current through a resistor is a func
tion of the voltage across the resistor and the resistance offered by the resistor, the AC current
through a capacitor is a function of the AC voltage across it, and the reactance offered by the
capacitor. As with inductors, the reactance of a capacitor is expressed in ohms and symbolized
by the letter X (or XC to be more specific).
Since capacitors “conduct” current in proportion to the rate of voltage change, they will pass
more current for fasterchanging voltages (as they charge and discharge to the same voltage
peaks in less time), and less current for slowerchanging voltages. What this means is that
reactance in ohms for any capacitor is inversely proportional to the frequency of the alternating
current. (Table 4.1)
XC = 2pifC
1
Table 4.1: Reactance of a 100 uF capacitor:
Frequency (Hertz) Reactance (Ohms)
60 26.5258
120 13.2629
2500 0.6366
Please note that the relationship of capacitive reactance to frequency is exactly opposite
from that of inductive reactance. Capacitive reactance (in ohms) decreases with increasing AC
frequency. Conversely, inductive reactance (in ohms) increases with increasing AC frequency.
Inductors oppose faster changing currents by producing greater voltage drops; capacitors op
pose faster changing voltage drops by allowing greater currents.
As with inductors, the reactance equation’s 2pif term may be replaced by the lowercase
Greek letter Omega (ω), which is referred to as the angular velocity of the AC circuit. Thus,
the equation XC = 1/(2pifC) could also be written as XC = 1/(ωC), with ω cast in units of radians
per second.
Alternating current in a simple capacitive circuit is equal to the voltage (in volts) divided
by the capacitive reactance (in ohms), just as either alternating or direct current in a simple
resistive circuit is equal to the voltage (in volts) divided by the resistance (in ohms). The
following circuit illustrates this mathematical relationship by example: (Figure 4.8)
10 V
60 Hz C 100 µF
Figure 4.8: Capacitive reactance.
86 CHAPTER 4. REACTANCE AND IMPEDANCE – CAPACITIVE
XC = 26.5258 Ω
I = E
X
I = 10 V
26.5258 Ω
I = 0.3770 A
However, we need to keep in mind that voltage and current are not in phase here. As was
shown earlier, the current has a phase shift of +90o with respect to the voltage. If we represent
these phase angles of voltage and current mathematically, we can calculate the phase angle of
the capacitor’s reactive opposition to current.
Opposition = Voltage
Current
Opposition = 10 V ∠ 0
o
0.3770 A ∠ 90o
Opposition = 26.5258 Ω ∠ 90o
I
E Opposition
For a capacitor:
90o
0o
90o
(XC)
Figure 4.9: Voltage lags current by 90o in a capacitor.
Mathematically, we say that the phase angle of a capacitor’s opposition to current is 90o,
meaning that a capacitor’s opposition to current is a negative imaginary quantity. (Figure 4.9)
This phase angle of reactive opposition to current becomes critically important in circuit anal
ysis, especially for complex AC circuits where reactance and resistance interact. It will prove
beneficial to represent any component’s opposition to current in terms of complex numbers,
and not just scalar quantities of resistance and reactance.
4.3. SERIES RESISTORCAPACITOR CIRCUITS 87
• REVIEW:
• Capacitive reactance is the opposition that a capacitor offers to alternating current due
to its phaseshifted storage and release of energy in its electric field. Reactance is sym
bolized by the capital letter “X” and is measured in ohms just like resistance (R).
• Capacitive reactance can be calculated using this formula: XC = 1/(2pifC)
• Capacitive reactance decreases with increasing frequency. In other words, the higher the
frequency, the less it opposes (the more it “conducts”) the AC flow of electrons.
4.3 Series resistorcapacitor circuits
In the last section, we learned what would happen in simple resistoronly and capacitoronly
AC circuits. Now we will combine the two components together in series form and investigate
the effects. (Figure 4.10)
R
C
EC
79.3°
ET
I ER
10 V
60 Hz
5 Ω
R
C
100
µF
ET I IR
VC
IC
I
= IR = IC
ET = ER+ EC
Figure 4.10: Series capacitor circuit: voltage lags current by 0o to 90o.
The resistor will offer 5 Ω of resistance to AC current regardless of frequency, while the
capacitor will offer 26.5258 Ω of reactance to AC current at 60 Hz. Because the resistor’s
resistance is a real number (5 Ω 6 0o, or 5 + j0 Ω), and the capacitor’s reactance is an imaginary
number (26.5258 Ω 6 90o, or 0  j26.5258 Ω), the combined effect of the two components will
be an opposition to current equal to the complex sum of the two numbers. The term for this
complex opposition to current is impedance, its symbol is Z, and it is also expressed in the unit
of ohms, just like resistance and reactance. In the above example, the total circuit impedance
is:
88 CHAPTER 4. REACTANCE AND IMPEDANCE – CAPACITIVE
Ztotal = (5 Ω resistance) + (26.5258 Ω capacitive reactance)
Ztotal = (5 Ω ∠ 0o) + (26.5258 Ω ∠ 90o)
or
(5 + j0 Ω) + (0  j26.5258 Ω)
Ztotal = 5  j26.5258 Ω or 26.993 Ω ∠ 79.325o
5 Ω (R) + 26.5258 Ω (XC)Ztotal =
Impedance is related to voltage and current just as you might expect, in a manner similar
to resistance in Ohm’s Law:
Ohm’s Law for AC circuits:
E = IZ I = Z =E
Z
E
I
All quantities expressed in
complex, not scalar, form
In fact, this is a far more comprehensive form of Ohm’s Law than what was taught in DC
electronics (E=IR), just as impedance is a far more comprehensive expression of opposition to
the flow of electrons than simple resistance is. Any resistance and any reactance, separately
or in combination (series/parallel), can be and should be represented as a single impedance.
To calculate current in the above circuit, we first need to give a phase angle reference for
the voltage source, which is generally assumed to be zero. (The phase angles of resistive and
capacitive impedance are always 0o and 90o, respectively, regardless of the given phase angles
for voltage or current).
I = E
Z
I = 10 V ∠ 0
o
26.933 Ω ∠ 79.325o
I = 370.5 mA ∠ 79.325o
As with the purely capacitive circuit, the current wave is leading the voltage wave (of the
source), although this time the difference is 79.325o instead of a full 90o. (Figure 4.11)
As we learned in the AC inductance chapter, the “table” method of organizing circuit quan
tities is a very useful tool for AC analysis just as it is for DC analysis. Let’s place out known
figures for this series circuit into a table and continue the analysis using this tool:
4.3. SERIES RESISTORCAPACITOR CIRCUITS 89
Time
+

e =
i =
phase shift =
79.325 degrees
Figure 4.11: Voltage lags current (current leads voltage)in a series RC circuit.
E
I
Volts
Amps
OhmsZ
R TotalC
10 + j0
10 ∠ 0o
5 + j0
5 ∠ 0o
0  j26.5258
26.5258 ∠ 90o
5  j26.5258
26.993 ∠ 79.325o
370.5m ∠ 79.325o
68.623m + j364.06m
Current in a series circuit is shared equally by all components, so the figures placed in the
“Total” column for current can be distributed to all other columns as well:
E
I
Volts
Amps
OhmsZ
R TotalC
10 + j0
10 ∠ 0o
5 + j0
5 ∠ 0o
0  j26.5258
26.5258 ∠ 90o
5  j26.5258
26.993 ∠ 79.325o
370.5m ∠ 79.325o
68.623m + j364.06m68.623m + j364.06m
370.5m ∠ 79.325o
68.623m + j364.06m
370.5m ∠ 79.325o
Rule of series
circuits:
Itotal = IR = IC
Continuing with our analysis, we can apply Ohm’s Law (E=IR) vertically to determine volt
age across the resistor and capacitor:
90 CHAPTER 4. REACTANCE AND IMPEDANCE – CAPACITIVE
E
I
Volts
Amps
OhmsZ
R TotalC
10 + j0
10 ∠ 0o
5 + j0
5 ∠ 0o
0  j26.5258
26.5258 ∠ 90o
5  j26.5258
26.993 ∠ 79.325o
370.5m ∠ 79.325o
68.623m + j364.06m68.623m + j364.06m
370.5m ∠ 79.325o
68.623m + j364.06m
370.5m ∠ 79.325o
343.11m + j1.8203
1.8523 ∠ 79.325o
Ohm’s
Law
Ohm’s
Law
9.6569  j1.8203
9.8269 ∠ 10.675o
E = IZ E = IZ
Notice how the voltage across the resistor has the exact same phase angle as the current
through it, telling us that E and I are in phase (for the resistor only). The voltage across the
capacitor has a phase angle of 10.675o, exactly 90o less than the phase angle of the circuit
current. This tells us that the capacitor’s voltage and current are still 90o out of phase with
each other.
Let’s check our calculations with SPICE: (Figure 4.12)
1 2
0 0
10 V
60 Hz
5 Ω
R
C 100 µF
Figure 4.12: Spice circuit: RC.
ac rc circuit
v1 1 0 ac 10 sin
r1 1 2 5
c1 2 0 100u
.ac lin 1 60 60
.print ac v(1,2) v(2,0) i(v1)
.print ac vp(1,2) vp(2,0) ip(v1)
.end
4.3. SERIES RESISTORCAPACITOR CIRCUITS 91
freq v(1,2) v(2) i(v1)
6.000E+01 1.852E+00 9.827E+00 3.705E01
freq vp(1,2) vp(2) ip(v1)
6.000E+01 7.933E+01 1.067E+01 1.007E+02
ER = 1.852 V ∠ 79.33o
EC = 9.827 V ∠ 10.67o
I = 370.5 mA ∠ 100.7o
Interpreted SPICE results
Once again, SPICE confusingly prints the current phase angle at a value equal to the real
phase angle plus 180o (or minus 180o). However, its a simple matter to correct this figure and
check to see if our work is correct. In this case, the 100.7o output by SPICE for current phase
angle equates to a positive 79.3o, which does correspond to our previously calculated figure of
79.325o.
Again, it must be emphasized that the calculated figures corresponding to reallife voltage
and current measurements are those in polar form, not rectangular form! For example, if
we were to actually build this series resistorcapacitor circuit and measure voltage across the
resistor, our voltmeter would indicate 1.8523 volts, not 343.11 millivolts (real rectangular)
or 1.8203 volts (imaginary rectangular). Real instruments connected to real circuits provide
indications corresponding to the vector length (magnitude) of the calculated figures. While the
rectangular form of complex number notation is useful for performing addition and subtraction,
it is a more abstract form of notation than polar, which alone has direct correspondence to true
measurements.
Impedance (Z) of a series RC circuit may be calculated, given the resistance (R) and the
capacitive reactance (XC). Since E=IR, E=IXC , and E=IZ, resistance, reactance, and impedance
are proportional to voltage, respectively. Thus, the voltage phasor diagram can be replaced by
a similar impedance diagram. (Figure 4.13)
R C
Voltage Impedance
ER
Z XC
R
ET I IR
EC
IC
37°
I
ET EC
37°
IER
Figure 4.13: Series: RC circuit Impedance phasor diagram.
Example:
Given: A 40 Ω resistor in series with a 88.42 microfarad capacitor. Find the impedance at
60 hertz.
92 CHAPTER 4. REACTANCE AND IMPEDANCE – CAPACITIVE
XC = 1/2pifC)
XC = 1/(2pi·60·88.42×10−6
XC = 30 Ω
Z = R  jXC
Z = 40  j30
Z = sqrt(402 + (30)2) = 50 Ω
6 Z = arctangent(30/40) = 36.87o
Z = 40  j30 = 506 36.87o
• REVIEW:
• Impedance is the total measure of opposition to electric current and is the complex (vec
tor) sum of (“real”) resistance and (“imaginary”) reactance.
• Impedances (Z) are managed just like resistances (R) in series circuit analysis: series
impedances add to form the total impedance. Just be sure to perform all calculations in
complex (not scalar) form! ZTotal = Z1 + Z2 + . . . Zn
• Please note that impedances always add in series, regardless of what type of components
comprise the impedances. That is, resistive impedance, inductive impedance, and capac
itive impedance are to be treated the same way mathematically.
• A purely resistive impedance will always have a phase angle of exactly 0o (ZR = R Ω 6
0o).
• A purely capacitive impedance will always have a phase angle of exactly 90o (ZC = XC Ω
6 90o).
• Ohm’s Law for AC circuits: E = IZ ; I = E/Z ; Z = E/I
• When resistors and capacitors are mixed together in circuits, the total impedance will
have a phase angle somewhere between 0o and 90o.
• Series AC circuits exhibit the same fundamental properties as series DC circuits: cur
rent is uniform throughout the circuit, voltage drops add to form the total voltage, and
impedances add to form the total impedance.
4.4 Parallel resistorcapacitor circuits
Using the same value components in our series example circuit, we will connect them in par
allel and see what happens: (Figure 4.14)
Because the power source has the same frequency as the series example circuit, and the
resistor and capacitor both have the same values of resistance and capacitance, respectively,
they must also have the same values of impedance. So, we can begin our analysis table with
the same “given” values:
4.4. PARALLEL RESISTORCAPACITOR CIRCUITS 93
R
C 10 V
60 Hz
5 ΩR
C100
µF
E I
IR IC
E
= ER = EC
I
= IR+ IL
IR
10.7°
IIC
E
Figure 4.14: Parallel RC circuit.
E
I
Volts
Amps
OhmsZ
R TotalC
10 + j0
10 ∠ 0o
5 + j0
5 ∠ 0o
0  j26.5258
26.5258 ∠ 90o
This being a parallel circuit now, we know that voltage is shared equally by all components,
so we can place the figure for total voltage (10 volts 6 0o) in all the columns:
E
I
Volts
Amps
OhmsZ
R TotalC
10 + j0
10 ∠ 0o
5 + j0
5 ∠ 0o
0  j26.5258
26.5258 ∠ 90o
10 + j0
10 ∠ 0o
10 + j0
10 ∠ 0o
Rule of parallel
circuits:
Etotal = ER = EC
Now we can apply Ohm’s Law (I=E/Z) vertically to two columns in the table, calculating
current through the resistor and current through the capacitor:
94 CHAPTER 4. REACTANCE AND IMPEDANCE – CAPACITIVE
E
I
Volts
Amps
OhmsZ
R TotalC
10 + j0
10 ∠ 0o
5 + j0
5 ∠ 0o
0  j26.5258
26.5258 ∠ 90o
10 + j0
10 ∠ 0o
10 + j0
10 ∠ 0o
2 + j0
2 ∠ 0o
0 + j376.99m
376.99m ∠ 90o
Ohm’s
Law
Ohm’s
Law
I = E
Z
I = E
Z
Just as with DC circuits, branch currents in a parallel AC circuit add up to form the total
current (Kirchhoff ’s Current Law again):
E
I
Volts
Amps
OhmsZ
R TotalC
10 + j0
10 ∠ 0o
5 + j0
5 ∠ 0o
0  j26.5258
26.5258 ∠ 90o
10 + j0
10 ∠ 0o
10 + j0
10 ∠ 0o
2 + j0
2 ∠ 0o
0 + j376.99m
376.99m ∠ 90o
2 + j376.99m
2.0352 ∠ 10.675o
Rule of parallel
circuits:
Itotal = IR + IC
Finally, total impedance can be calculated by using Ohm’s Law (Z=E/I) vertically in the
“Total” column. As we saw in the AC inductance chapter, parallel impedance can also be cal
culated by using a reciprocal formula identical to that used in calculating parallel resistances.
It is noteworthy to mention that this parallel impedance rule holds true regardless of the kind
of impedances placed in parallel. In other words, it doesn’t matter if we’re calculating a cir
cuit composed of parallel resistors, parallel inductors, parallel capacitors, or some combination
thereof: in the form of impedances (Z), all the terms are common and can be applied uniformly
to the same formula. Once again, the parallel impedance formula looks like this:
Zparallel =
Z1 Z2 Zn
1
+
1
+ . . .
1
1
The only drawback to using this equation is the significant amount of work required to
work it out, especially without the assistance of a calculator capable of manipulating complex
quantities. Regardless of how we calculate total impedance for our parallel circuit (either
Ohm’s Law or the reciprocal formula), we will arrive at the same figure:
4.5. CAPACITOR QUIRKS 95
E
I
Volts
Amps
OhmsZ
R TotalC
10 + j0
10 ∠ 0o
5 + j0
5 ∠ 0o
0  j26.5258
26.5258 ∠ 90o
10 + j0
10 ∠ 0o
10 + j0
10 ∠ 0o
2 + j0
2 ∠ 0o
0 + j376.99m
376.99m ∠ 90o
2 + j376.99m
2.0352 ∠ 10.675o
4.8284  j910.14m
4.9135 ∠ 10.675o
Ohm’s
Law or
Rule of parallel
circuits:
Ztotal =
ZR ZC
1
+
1
1Z = E
I
• REVIEW:
• Impedances (Z) are managed just like resistances (R) in parallel circuit analysis: parallel
impedances diminish to form the total impedance, using the reciprocal formula. Just be
sure to perform all calculations in complex (not scalar) form! ZTotal = 1/(1/Z1 + 1/Z2 + . . .
1/Zn)
• Ohm’s Law for AC circuits: E = IZ ; I = E/Z ; Z = E/I
• When resistors and capacitors are mixed together in parallel circuits (just as in series
circuits), the total impedance will have a phase angle somewhere between 0o and 90o.
The circuit current will have a phase angle somewhere between 0o and +90o.
• Parallel AC circuits exhibit the same fundamental properties as parallel DC circuits:
voltage is uniform throughout the circuit, branch currents add to form the total current,
and impedances diminish (through the reciprocal formula) to form the total impedance.
4.5 Capacitor quirks
As with inductors, the ideal capacitor is a purely reactive device, containing absolutely zero
resistive (power dissipative) effects. In the real world, of course, nothing is so perfect. However,
capacitors have the virtue of generally being purer reactive components than inductors. It is
a lot easier to design and construct a capacitor with low internal series resistance than it is
to do the same with an inductor. The practical result of this is that real capacitors typically
have impedance phase angles more closely approaching 90o (actually, 90o) than inductors.
Consequently, they will tend to dissipate less power than an equivalent inductor.
Capacitors also tend to be smaller and lighter weight than their equivalent inductor coun
terparts, and since their electric fields are almost totally contained between their plates (unlike
inductors, whose magnetic fields naturally tend to extend beyond the dimensions of the core),
96 CHAPTER 4. REACTANCE AND IMPEDANCE – CAPACITIVE
they are less prone to transmitting or receiving electromagnetic “noise” to/from other compo
nents. For these reasons, circuit designers tend to favor capacitors over inductors wherever a
design permits either alternative.
Capacitors with significant resistive effects are said to be lossy, in reference to their ten
dency to dissipate (“lose”) power like a resistor. The source of capacitor loss is usually the
dielectric material rather than any wire resistance, as wire length in a capacitor is very mini
mal.
Dielectric materials tend to react to changing electric fields by producing heat. This heating
effect represents a loss in power, and is equivalent to resistance in the circuit. The effect is more
pronounced at higher frequencies and in fact can be so extreme that it is sometimes exploited
in manufacturing processes to heat insulating materials like plastic! The plastic object to be
heated is placed between two metal plates, connected to a source of highfrequency AC voltage.
Temperature is controlled by varying the voltage or frequency of the source, and the plates
never have to contact the object being heated.
This effect is undesirable for capacitors where we expect the component to behave as a
purely reactive circuit element. One of the ways to mitigate the effect of dielectric “loss” is
to choose a dielectric material less susceptible to the effect. Not all dielectric materials are
equally “lossy.” A relative scale of dielectric loss from least to greatest is given in Table 4.2.
Table 4.2: Dielectric loss
Material Loss
Vacuum Low
Air 
Polystyrene 
Mica 
Glass 
LowK ceramic 
Plastic film (Mylar) 
Paper 
HighK ceramic 
Aluminum oxide 
Tantalum pentoxide high
Dielectric resistivity manifests itself both as a series and a parallel resistance with the pure
capacitance: (Figure 4.15)
Fortunately, these stray resistances are usually of modest impact (low series resistance and
high parallel resistance), much less significant than the stray resistances present in an average
inductor.
Electrolytic capacitors, known for their relatively high capacitance and low working volt
age, are also known for their notorious lossiness, due to both the characteristics of the micro
scopically thin dielectric film and the electrolyte paste. Unless specially made for AC service,
electrolytic capacitors should never be used with AC unless it is mixed (biased) with a constant
DC voltage preventing the capacitor from ever being subjected to reverse voltage. Even then,
their resistive characteristics may be too severe a shortcoming for the application anyway.
4.6. CONTRIBUTORS 97
Ideal
capacitor
Equivalent circuit for a real capacitor
Rseries
Rparallel
Figure 4.15: Real capacitor has both series and parallel resistance.
4.6 Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
Jason Starck (June 2000): HTML document formatting, which led to a much better
looking second edition.
98 CHAPTER 4. REACTANCE AND IMPEDANCE – CAPACITIVE
Chapter 5
REACTANCE AND IMPEDANCE
– R, L, AND C
Contents
5.1 Review of R, X, and Z . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
5.2 Series R, L, and C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
5.3 Parallel R, L, and C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
5.4 Seriesparallel R, L, and C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
5.5 Susceptance and Admittance . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
5.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
5.7 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
5.1 Review of R, X, and Z
Before we begin to explore the effects of resistors, inductors, and capacitors connected together
in the same AC circuits, let’s briefly review some basic terms and facts.
Resistance is essentially friction against the motion of electrons. It is present in all con
ductors to some extent (except superconductors!), most notably in resistors. When alternating
current goes through a resistance, a voltage drop is produced that is inphase with the current.
Resistance is mathematically symbolized by the letter “R” and is measured in the unit of ohms
(Ω).
Reactance is essentially inertia against the motion of electrons. It is present anywhere
electric or magnetic fields are developed in proportion to applied voltage or current, respec
tively; but most notably in capacitors and inductors. When alternating current goes through a
pure reactance, a voltage drop is produced that is 90o out of phase with the current. Reactance
is mathematically symbolized by the letter “X” and is measured in the unit of ohms (Ω).
99
100 CHAPTER 5. REACTANCE AND IMPEDANCE – R, L, AND C
Impedance is a comprehensive expression of any and all forms of opposition to electron
flow, including both resistance and reactance. It is present in all circuits, and in all compo
nents. When alternating current goes through an impedance, a voltage drop is produced that
is somewhere between 0o and 90o out of phase with the current. Impedance is mathematically
symbolized by the letter “Z” and is measured in the unit of ohms (Ω), in complex form.
Perfect resistors (Figure 5.1) possess resistance, but not reactance. Perfect inductors and
perfect capacitors (Figure 5.1) possess reactance but no resistance. All components possess
impedance, and because of this universal quality, it makes sense to translate all component
values (resistance, inductance, capacitance) into common terms of impedance as the first step
in analyzing an AC circuit.
Resistor 100 Ω
R = 100 Ω
X = 0 Ω
Z = 100 Ω ∠ 0o
Inductor
R = 0 Ω
159.15 Hz
100 mH
X = 100 Ω
Z = 100 Ω ∠ 90o
Capacitor 10 µF
159.15 Hz
R = 0 Ω
X = 100 Ω
Z = 100 Ω ∠ 90o
Figure 5.1: Perfect resistor, inductor, and capacitor.
The impedance phase angle for any component is the phase shift between voltage across
that component and current through that component. For a perfect resistor, the voltage drop
and current are always in phase with each other, and so the impedance angle of a resistor
is said to be 0o. For an perfect inductor, voltage drop always leads current by 90o, and so
an inductor’s impedance phase angle is said to be +90o. For a perfect capacitor, voltage drop
always lags current by 90o, and so a capacitor’s impedance phase angle is said to be 90o.
Impedances in AC behave analogously to resistances in DC circuits: they add in series, and
they diminish in parallel. A revised version of Ohm’s Law, based on impedance rather than
resistance, looks like this:
Ohm’s Law for AC circuits:
E = IZ I = Z =E
Z
E
I
All quantities expressed in
complex, not scalar, form
Kirchhoff ’s Laws and all network analysis methods and theorems are true for AC circuits
as well, so long as quantities are represented in complex rather than scalar form. While this
qualified equivalence may be arithmetically challenging, it is conceptually simple and elegant.
The only real difference between DC and AC circuit calculations is in regard to power. Because
reactance doesn’t dissipate power as resistance does, the concept of power in AC circuits is
radically different from that of DC circuits. More on this subject in a later chapter!
5.2. SERIES R, L, AND C 101
5.2 Series R, L, and C
Let’s take the following example circuit and analyze it: (Figure 5.2)
120 V
60 Hz
250 Ω
R
L
C
650 mH
1.5 µF
Figure 5.2: Example series R, L, and C circuit.
The first step is to determine the reactances (in ohms) for the inductor and the capacitor.
XL = 2pifL
XL = (2)(pi)(60 Hz)(650 mH)
XL = 245.04 Ω
XC = 2pifC
1
XC = (2)(pi)(60 Hz)(1.5 µF)
1
XC = 1.7684 kΩ
The next step is to express all resistances and reactances in a mathematically common
form: impedance. (Figure 5.3) Remember that an inductive reactance translates into a positive
imaginary impedance (or an impedance at +90o), while a capacitive reactance translates into a
negative imaginary impedance (impedance at 90o). Resistance, of course, is still regarded as
a purely “real” impedance (polar angle of 0o):
ZR = 250 + j0 Ω or 250 Ω ∠ 0o
ZL = 0 + j245.04 Ω or 245.04 Ω ∠ 90o
ZC = 0  j1.7684k Ω or 1.7684 kΩ ∠ 90o
102 CHAPTER 5. REACTANCE AND IMPEDANCE – R, L, AND C
120 V
60 Hz
ZR
ZL
ZC
250 Ω ∠ 0o
245.04 Ω ∠ 90o
1.7684 kΩ ∠ 90o
Figure 5.3: Example series R, L, and C circuit with component values replaced by impedances.
Now, with all quantities of opposition to electric current expressed in a common, complex
number format (as impedances, and not as resistances or reactances), they can be handled in
the same way as plain resistances in a DC circuit. This is an ideal time to draw up an analysis
table for this circuit and insert all the “given” figures (total voltage, and the impedances of the
resistor, inductor, and capacitor).
E
I
Volts
Amps
OhmsZ
R L TotalC
250 + j0
250 ∠ 0o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
120 + j0
120 ∠ 0o
Unless otherwise specified, the source voltage will be our reference for phase shift, and so
will be written at an angle of 0o. Remember that there is no such thing as an “absolute” angle
of phase shift for a voltage or current, since its always a quantity relative to another wave
form. Phase angles for impedance, however (like those of the resistor, inductor, and capacitor),
are known absolutely, because the phase relationships between voltage and current at each
component are absolutely defined.
Notice that I’m assuming a perfectly reactive inductor and capacitor, with impedance phase
angles of exactly +90 and 90o, respectively. Although real components won’t be perfect in this
regard, they should be fairly close. For simplicity, I’ll assume perfectly reactive inductors and
capacitors from now on in my example calculations except where noted otherwise.
Since the above example circuit is a series circuit, we know that the total circuit impedance
is equal to the sum of the individuals, so:
Ztotal = ZR + ZL + ZC
Ztotal = (250 + j0 Ω) + (0 + j245.04 Ω) + (0  j1.7684k Ω)
Ztotal = 250  j1.5233k Ω or 1.5437 kΩ ∠ 80.680o
Inserting this figure for total impedance into our table:
5.2. SERIES R, L, AND C 103
E
I
Volts
Amps
OhmsZ
R L TotalC
250 + j0
250 ∠ 0o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
120 + j0
120 ∠ 0o
250  j1.5233k
1.5437k ∠ 80.680o
Rule of series
circuits:
Ztotal = ZR + ZL + ZC
We can now apply Ohm’s Law (I=E/R) vertically in the “Total” column to find total current
for this series circuit:
E
I
Volts
Amps
OhmsZ
R L TotalC
250 + j0
250 ∠ 0o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
120 + j0
120 ∠ 0o
250  j1.5233k
1.5437k ∠ 80.680o
12.589m + 76.708m
77.734m ∠ 80.680o
Ohm’s
Law
I = E
Z
Being a series circuit, current must be equal through all components. Thus, we can take
the figure obtained for total current and distribute it to each of the other columns:
E
I
Volts
Amps
OhmsZ
R L TotalC
250 + j0
250 ∠ 0o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
120 + j0
120 ∠ 0o
250  j1.5233k
1.5437k ∠ 80.680o
12.589m + 76.708m
77.734m ∠ 80.680o
12.589m + 76.708m
77.734m ∠ 80.680o
12.589m + 76.708m
77.734m ∠ 80.680o
12.589m + 76.708m
77.734m ∠ 80.680o
Rule of series
circuits:
Itotal = IR = IL = IC
Nowwe’re prepared to apply Ohm’s Law (E=IZ) to each of the individual component columns
in the table, to determine voltage drops:
104 CHAPTER 5. REACTANCE AND IMPEDANCE – R, L, AND C
E
I
Volts
Amps
OhmsZ
R L TotalC
250 + j0
250 ∠ 0o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
120 + j0
120 ∠ 0o
250  j1.5233k
1.5437k ∠ 80.680o
12.589m + 76.708m
77.734m ∠ 80.680o
12.589m + 76.708m
77.734m ∠ 80.680o
12.589m + 76.708m
77.734m ∠ 80.680o
12.589m + 76.708m
77.734m ∠ 80.680o
Ohm’s
Law
Ohm’s
Law
Ohm’s
Law
3.1472 + j19.177
19.434 ∠ 80.680o
18.797 + j3.0848
19.048 ∠ 170.68o
135.65  j22.262
137.46 ∠ 9.3199o
E = IZ E = IZ E = IZ
Notice something strange here: although our supply voltage is only 120 volts, the voltage
across the capacitor is 137.46 volts! How can this be? The answer lies in the interaction
between the inductive and capacitive reactances. Expressed as impedances, we can see that
the inductor opposes current in a manner precisely opposite that of the capacitor. Expressed
in rectangular form, the inductor’s impedance has a positive imaginary term and the capacitor
has a negative imaginary term. When these two contrary impedances are added (in series),
they tend to cancel each other out! Although they’re still added together to produce a sum, that
sum is actually less than either of the individual (capacitive or inductive) impedances alone.
It is analogous to adding together a positive and a negative (scalar) number: the sum is a
quantity less than either one’s individual absolute value.
If the total impedance in a series circuit with both inductive and capacitive elements is less
than the impedance of either element separately, then the total current in that circuit must be
greater than what it would be with only the inductive or only the capacitive elements there.
With this abnormally high current through each of the components, voltages greater than the
source voltage may be obtained across some of the individual components! Further conse
quences of inductors’ and capacitors’ opposite reactances in the same circuit will be explored
in the next chapter.
Once you’ve mastered the technique of reducing all component values to impedances (Z),
analyzing any AC circuit is only about as difficult as analyzing any DC circuit, except that the
quantities dealt with are vector instead of scalar. With the exception of equations dealing with
power (P), equations in AC circuits are the same as those in DC circuits, using impedances (Z)
instead of resistances (R). Ohm’s Law (E=IZ) still holds true, and so do Kirchhoff ’s Voltage and
Current Laws.
To demonstrate Kirchhoff ’s Voltage Law in an AC circuit, we can look at the answers we
derived for component voltage drops in the last circuit. KVL tells us that the algebraic sum of
the voltage drops across the resistor, inductor, and capacitor should equal the applied voltage
from the source. Even though this may not look like it is true at first sight, a bit of complex
number addition proves otherwise:
5.2. SERIES R, L, AND C 105
ER + EL + EC should equal Etotal
3.1472 + j19.177 V ER
18.797 + j3.0848 V EL
135.65  j22.262 V EC+
120 + j0 V Etotal
Aside from a bit of rounding error, the sum of these voltage drops does equal 120 volts.
Performed on a calculator (preserving all digits), the answer you will receive should be exactly
120 + j0 volts.
We can also use SPICE to verify our figures for this circuit: (Figure 5.4)
1 2
30
120 V
60 Hz
R
L
C
250 Ω
650 mH
1.5 µF
Figure 5.4: Example series R, L, and C SPICE circuit.
ac rlc circuit
v1 1 0 ac 120 sin
r1 1 2 250
l1 2 3 650m
c1 3 0 1.5u
.ac lin 1 60 60
.print ac v(1,2) v(2,3) v(3,0) i(v1)
.print ac vp(1,2) vp(2,3) vp(3,0) ip(v1)
.end
freq v(1,2) v(2,3) v(3) i(v1)
6.000E+01 1.943E+01 1.905E+01 1.375E+02 7.773E02
freq vp(1,2) vp(2,3) vp(3) ip(v1)
6.000E+01 8.068E+01 1.707E+02 9.320E+00 9.932E+01
106 CHAPTER 5. REACTANCE AND IMPEDANCE – R, L, AND C
ER = 19.43 V ∠ 80.68o
EL = 19.05 V ∠ 170.7o
EC = 137.5 V ∠ 9.320o
I = 77.73 mA ∠ 99.32o (actual phase angle = 80.68o)
Interpreted SPICE results
The SPICE simulation shows our handcalculated results to be accurate.
As you can see, there is little difference between AC circuit analysis and DC circuit analysis,
except that all quantities of voltage, current, and resistance (actually, impedance) must be
handled in complex rather than scalar form so as to account for phase angle. This is good,
since it means all you’ve learned about DC electric circuits applies to what you’re learning
here. The only exception to this consistency is the calculation of power, which is so unique that
it deserves a chapter devoted to that subject alone.
• REVIEW:
• Impedances of any kind add in series: ZTotal = Z1 + Z2 + . . . Zn
• Although impedances add in series, the total impedance for a circuit containing both
inductance and capacitance may be less than one or more of the individual impedances,
because series inductive and capacitive impedances tend to cancel each other out. This
may lead to voltage drops across components exceeding the supply voltage!
• All rules and laws of DC circuits apply to AC circuits, so long as values are expressed in
complex form rather than scalar. The only exception to this principle is the calculation of
power, which is very different for AC.
5.3 Parallel R, L, and C
We can take the same components from the series circuit and rearrange them into a parallel
configuration for an easy example circuit: (Figure 5.5)
120 V
60 Hz
R L C250 Ω 650 mH 1.5 µF
Figure 5.5: Example R, L, and C parallel circuit.
The fact that these components are connected in parallel instead of series now has ab
solutely no effect on their individual impedances. So long as the power supply is the same
5.3. PARALLEL R, L, AND C 107
120 V
60 Hz
ZR ZL ZC
250 Ω ∠ 0o
245.04 Ω ∠ 90o
1.7684 kΩ ∠ 90o
Figure 5.6: Example R, L, and C parallel circuit with impedances replacing component values.
frequency as before, the inductive and capacitive reactances will not have changed at all: (Fig
ure 5.6)
With all component values expressed as impedances (Z), we can set up an analysis table
and proceed as in the last example problem, except this time following the rules of parallel
circuits instead of series:
E
I
Volts
Amps
OhmsZ
R L TotalC
250 + j0
250 ∠ 0o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
120 + j0
120 ∠ 0o
Knowing that voltage is shared equally by all components in a parallel circuit, we can
transfer the figure for total voltage to all component columns in the table:
E
I
Volts
Amps
OhmsZ
R L TotalC
250 + j0
250 ∠ 0o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
120 + j0
120 ∠ 0o
120 + j0
120 ∠ 0o
120 + j0
120 ∠ 0o
120 + j0
120 ∠ 0o
Rule of parallel
circuits:
Etotal = ER = EL = EC
Now, we can apply Ohm’s Law (I=E/Z) vertically in each column to determine current
through each component:
108 CHAPTER 5. REACTANCE AND IMPEDANCE – R, L, AND C
E
I
Volts
Amps
OhmsZ
R L TotalC
250 + j0
250 ∠ 0o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
120 + j0
120 ∠ 0o
120 + j0
120 ∠ 0o
120 + j0
120 ∠ 0o
120 + j0
120 ∠ 0o
Ohm’s
Law
Ohm’s
Law
Ohm’s
Law
480m + j0
480m ∠ 0o
0  j489.71m
489.71m ∠ 90o
0 + j67.858m
67.858m ∠ 90o
I = E
Z
I = E
Z
I = E
Z
There are two strategies for calculating total current and total impedance. First, we could
calculate total impedance from all the individual impedances in parallel (ZTotal = 1/(1/ZR + 1/ZL
+ 1/ZC), and then calculate total current by dividing source voltage by total impedance (I=E/Z).
However, working through the parallel impedance equation with complex numbers is no easy
task, with all the reciprocations (1/Z). This is especially true if you’re unfortunate enough
not to have a calculator that handles complex numbers and are forced to do it all by hand
(reciprocate the individual impedances in polar form, then convert them all to rectangular
form for addition, then convert back to polar form for the final inversion, then invert). The
second way to calculate total current and total impedance is to add up all the branch currents
to arrive at total current (total current in a parallel circuit – AC or DC – is equal to the sum
of the branch currents), then use Ohm’s Law to determine total impedance from total voltage
and total current (Z=E/I).
E
I
Volts
Amps
OhmsZ
R L TotalC
250 + j0
250 ∠ 0o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
120 + j0
120 ∠ 0o
120 + j0
120 ∠ 0o
120 + j0
120 ∠ 0o
120 + j0
120 ∠ 0o
480m + j0
480 ∠ 0o
0  j489.71m
489.71m ∠ 90o
0 + j67.858m
67.858m ∠ 90o
480m  j421.85m
639.03m ∠ 41.311o
141.05 + j123.96
187.79 ∠ 41.311o
Either method, performed properly, will provide the correct answers. Let’s try analyzing
this circuit with SPICE and see what happens: (Figure 5.7)
5.3. PARALLEL R, L, AND C 109
2
4
5
6
1
2
3
0 0 0 0
2 2
120 V
60 Hz
R L C250 Ω 650 mH 1.5 µF
Vir Vil Vic
Vi
Rbogus
Figure 5.7: Example parallel R, L, and C SPICE circuit. Battery symbols are “dummy” voltage
sources for SPICE to use as current measurement points. All are set to 0 volts.
ac rlc circuit
v1 1 0 ac 120 sin
vi 1 2 ac 0
vir 2 3 ac 0
vil 2 4 ac 0
rbogus 4 5 1e12
vic 2 6 ac 0
r1 3 0 250
l1 5 0 650m
c1 6 0 1.5u
.ac lin 1 60 60
.print ac i(vi) i(vir) i(vil) i(vic)
.print ac ip(vi) ip(vir) ip(vil) ip(vic)
.end
freq i(vi) i(vir) i(vil) i(vic)
6.000E+01 6.390E01 4.800E01 4.897E01 6.786E02
freq ip(vi) ip(vir) ip(vil) ip(vic)
6.000E+01 4.131E+01 0.000E+00 9.000E+01 9.000E+01
110 CHAPTER 5. REACTANCE AND IMPEDANCE – R, L, AND C
Itotal = 639.0 mA ∠ 41.31o
IR = 480 mA ∠ 0o
IL = 489.7 mA ∠ 90o
IC = 67.86 mA ∠ 90o
Interpreted SPICE results
It took a little bit of trickery to get SPICE working as we would like on this circuit (installing
“dummy” voltage sources in each branch to obtain current figures and installing the “dummy”
resistor in the inductor branch to prevent a direct inductortovoltage source loop, which SPICE
cannot tolerate), but we did get the proper readings. Even more than that, by installing the
dummy voltage sources (current meters) in the proper directions, we were able to avoid that
idiosyncrasy of SPICE of printing current figures 180o out of phase. This way, our current
phase readings came out to exactly match our hand calculations.
5.4 Seriesparallel R, L, and C
Now that we’ve seen how series and parallel AC circuit analysis is not fundamentally different
than DC circuit analysis, it should come as no surprise that seriesparallel analysis would be
the same as well, just using complex numbers instead of scalar to represent voltage, current,
and impedance.
Take this seriesparallel circuit for example: (Figure 5.8)
120 V
60 Hz
C1
4.7 µF L 650 mH
R 470 Ω
C2 1.5 µF
Figure 5.8: Example seriesparallel R, L, and C circuit.
The first order of business, as usual, is to determine values of impedance (Z) for all compo
nents based on the frequency of the AC power source. To do this, we need to first determine
values of reactance (X) for all inductors and capacitors, then convert reactance (X) and resis
tance (R) figures into proper impedance (Z) form:
5.4. SERIESPARALLEL R, L, AND C 111
Reactances and Resistances:
XC1 = 2pifC1
1
XC1 = (2)(pi)(60 Hz)(4.7 µF)
1
XC1 = 564.38 Ω
XL = 2pifL
XL = (2)(pi)(60 Hz)(650 mH)
XL = 245.04 Ω
XC2 =
1
2pifC2
XC2 = (2)(pi)(60 Hz)(1.5 µF)
1
XC2 = 1.7684 kΩ
R = 470 Ω
ZC1 = 0  j564.38 Ω or 564.38 Ω ∠ 90o
ZL = 0 + j245.04 Ω or 245.04 Ω ∠ 90o
ZC2 = 0  j1.7684k Ω or 1.7684 kΩ ∠ 90o
ZR = 470 + j0 Ω or 470 Ω ∠ 0o
Now we can set up the initial values in our table:
E
I
Volts
Amps
OhmsZ
TotalC1 L C2 R
470 + j0
470 ∠ 0o
120 + j0
120 ∠ 0o
0  j564.38
564.38 ∠ 90o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
Being a seriesparallel combination circuit, we must reduce it to a total impedance in more
than one step. The first step is to combine L and C2 as a series combination of impedances,
by adding their impedances together. Then, that impedance will be combined in parallel with
the impedance of the resistor, to arrive at another combination of impedances. Finally, that
quantity will be added to the impedance of C1 to arrive at the total impedance.
In order that our table may follow all these steps, it will be necessary to add additional
columns to it so that each step may be represented. Adding more columns horizontally to the
table shown above would be impractical for formatting reasons, so I will place a new row of
columns underneath, each column designated by its respective component combination:
112 CHAPTER 5. REACTANCE AND IMPEDANCE – R, L, AND C
E
I
Volts
Amps
OhmsZ
L  C2 R // (L  C2) C1  [R // (L  C2)]
Total
Calculating these new (combination) impedances will require complex addition for series
combinations, and the “reciprocal” formula for complex impedances in parallel. This time,
there is no avoidance of the reciprocal formula: the required figures can be arrived at no other
way!
E
I
Volts
Amps
OhmsZ
L  C2 R // (L  C2) C1  [R // (L  C2)]
Total
0  j1.5233k
1.5233k ∠ 90o
429.15  j132.41
449.11 ∠ 17.147o
429.15  j696.79
818.34 ∠ 58.371o
120 + j0
120 ∠ 0o
Rule of series
circuits:
Rule of parallel
circuits:
Rule of series
circuits:
ZLC2 = ZL + ZC2
ZR//(LC2) =
ZR ZLC2
11
+
1
Ztotal = ZC1 + ZR//(LC2)
Seeing as how our second table contains a column for “Total,” we can safely discard that
column from the first table. This gives us one table with four columns and another table with
three columns.
Now that we know the total impedance (818.34 Ω 6 58.371o) and the total voltage (120
volts 6 0o), we can apply Ohm’s Law (I=E/Z) vertically in the “Total” column to arrive at a
figure for total current:
5.4. SERIESPARALLEL R, L, AND C 113
E
I
Volts
Amps
OhmsZ
L  C2 R // (L  C2) C1  [R // (L  C2)]
Total
0  j1.5233k
1.5233k ∠ 90o
429.15  j132.41
449.11 ∠ 17.147o
429.15  j696.79
818.34 ∠ 58.371o
120 + j0
120 ∠ 0o
76.899m + j124.86m
146.64m ∠ 58.371o
Ohm’s
Law
I = E
Z
At this point we ask ourselves the question: are there any components or component com
binations which share either the total voltage or the total current? In this case, both C1 and
the parallel combination R//(L−−C2) share the same (total) current, since the total impedance
is composed of the two sets of impedances in series. Thus, we can transfer the figure for total
current into both columns:
E
I
Volts
Amps
OhmsZ
C1 L C2 R
470 + j0
470 ∠ 0o
0  j564.38
564.38 ∠ 90o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
76.899m + j124.86m
146.64m ∠ 58.371o
Rule of series
circuits:
Itotal = IC1 = IR//(LC2)
E
I
Volts
Amps
OhmsZ
L  C2 R // (L  C2) C1  [R // (L  C2)]
Total
0  j1.5233k
1.5233k ∠ 90o
429.15  j132.41
449.11 ∠ 17.147o
429.15  j696.79
818.34 ∠ 58.371o
120 + j0
120 ∠ 0o
76.899m + j124.86m
146.64m ∠ 58.371o
76.899m + j124.86m
146.64m ∠ 58.371o
Rule of series
circuits:
Itotal = IC1 = IR//(LC2)
Now, we can calculate voltage drops across C1 and the seriesparallel combination of R//(L−−C2)
using Ohm’s Law (E=IZ) vertically in those table columns:
114 CHAPTER 5. REACTANCE AND IMPEDANCE – R, L, AND C
E
I
Volts
Amps
OhmsZ
C1 L C2 R
470 + j0
470 ∠ 0o
0  j564.38
564.38 ∠ 90o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
76.899m + j124.86m
146.64m ∠ 58.371o
Ohm’s
Law
70.467  j43.400
82.760 ∠ 31.629o
E = IZ
E
I
Volts
Amps
OhmsZ
L  C2 R // (L  C2) C1  [R // (L  C2)]
Total
0  j1.5233k
1.5233k ∠ 90o
429.15  j132.41
449.11 ∠ 17.147o
429.15  j696.79
818.34 ∠ 58.371o
120 + j0
120 ∠ 0o
76.899m + j124.86m
146.64m ∠ 58.371o
76.899m + j124.86m
146.64m ∠ 58.371o
Ohm’s
Law
49.533 + j43.400
65.857 ∠ 41.225o
E = IZ
A quick doublecheck of our work at this point would be to see whether or not the voltage
drops across C1 and the seriesparallel combination of R//(L−−C2) indeed add up to the total.
According to Kirchhoff ’s Voltage Law, they should!
Etotal should be equal to EC1 + ER//(LC2)
70.467  j43.400 V
49.533 + j43.400 V+
120 + j0 V Indeed, it is!
That last step was merely a precaution. In a problem with as many steps as this one has,
there is much opportunity for error. Occasional crosschecks like that one can save a person a
lot of work and unnecessary frustration by identifying problems prior to the final step of the
problem.
After having solved for voltage drops across C1 and the combination R//(L−−C2), we again
ask ourselves the question: what other components share the same voltage or current? In this
case, the resistor (R) and the combination of the inductor and the second capacitor (L−−C2)
share the same voltage, because those sets of impedances are in parallel with each other.
Therefore, we can transfer the voltage figure just solved for into the columns for R and L−−C2:
5.4. SERIESPARALLEL R, L, AND C 115
E
I
Volts
Amps
OhmsZ
C1 L C2 R
470 + j0
470 ∠ 0o
0  j564.38
564.38 ∠ 90o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
76.899m + j124.86m
146.64m ∠ 58.371o
70.467  j43.400
82.760 ∠ 31.629o
49.533 + j43.400
65.857 ∠ 41.225o
Rule of parallel
circuits:
ER//(LC2) = ER = ELC2
E
I
Volts
Amps
OhmsZ
L  C2 R // (L  C2) C1  [R // (L  C2)]
Total
0  j1.5233k
1.5233k ∠ 90o
429.15  j132.41
449.11 ∠ 17.147o
429.15  j696.79
818.34 ∠ 58.371o
120 + j0
120 ∠ 0o
76.899m + j124.86m
146.64m ∠ 58.371o
76.899m + j124.86m
146.64m ∠ 58.371o
49.533 + j43.400
65.857 ∠ 41.225o
Rule of parallel
circuits:
ER//(LC2) = ER = ELC2
49.533 + j43.400
65.857 ∠ 41.225o
Now we’re all set for calculating current through the resistor and through the series com
bination L−−C2. All we need to do is apply Ohm’s Law (I=E/Z) vertically in both of those
columns:
E
I
Volts
Amps
OhmsZ
C1 L C2 R
470 + j0
470 ∠ 0o
0  j564.38
564.38 ∠ 90o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
76.899m + j124.86m
146.64m ∠ 58.371o
70.467  j43.400
82.760 ∠ 31.629o
49.533 + j43.400
65.857 ∠ 41.225o
Ohm’s
Law
105.39m + j92.341m
140.12m ∠ 41.225o
I = E
Z
116 CHAPTER 5. REACTANCE AND IMPEDANCE – R, L, AND C
E
I
Volts
Amps
OhmsZ
L  C2 R // (L  C2) C1  [R // (L  C2)]
Total
0  j1.5233k
1.5233k ∠ 90o
429.15  j132.41
449.11 ∠ 17.147o
429.15  j696.79
818.34 ∠ 58.371o
120 + j0
120 ∠ 0o
76.899m + j124.86m
146.64m ∠ 58.371o
76.899m + j124.86m
146.64m ∠ 58.371o
49.533 + j43.400
65.857 ∠ 41.225o
49.533 + j43.400
65.857 ∠ 41.225o
28.490m + j32.516m
43.232m ∠ 131.22o
Ohm’s
Law
I = E
Z
Another quick doublecheck of our work at this point would be to see if the current figures
for L−−C2 and R add up to the total current. According to Kirchhoff ’s Current Law, they
should:
IR//(LC2) should be equal to IR + I(LC2)
105.39m + j92.341m
28.490m + j32.516m+
76.899m + j124.86m Indeed, it is!
Since the L and C2 are connected in series, and since we know the current through their
series combination impedance, we can distribute that current figure to the L and C2 columns
following the rule of series circuits whereby series components share the same current:
E
I
Volts
Amps
OhmsZ
C1 L C2 R
470 + j0
470 ∠ 0o
0  j564.38
564.38 ∠ 90o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
76.899m + j124.86m
146.64m ∠ 58.371o
70.467  j43.400
82.760 ∠ 31.629o
49.533 + j43.400
65.857 ∠ 41.225o
105.39m + j92.341m
140.12m ∠ 41.225o
28.490m + j32.516m
43.232m ∠ 131.22o
28.490m + j32.516m
43.232m ∠ 131.22o
Rule of series
circuits:
ILC2 = IL = IC2
With one last step (actually, two calculations), we can complete our analysis table for this
circuit. With impedance and current figures in place for L and C2, all we have to do is apply
Ohm’s Law (E=IZ) vertically in those two columns to calculate voltage drops.
5.4. SERIESPARALLEL R, L, AND C 117
E
I
Volts
Amps
OhmsZ
C1 L C2 R
470 + j0
470 ∠ 0o
0  j564.38
564.38 ∠ 90o
0 + j245.04
245.04 ∠ 90o
0  j1.7684k
1.7684k ∠ 90o
76.899m + j124.86m
146.64m ∠ 58.371o
70.467  j43.400
82.760 ∠ 31.629o
49.533 + j43.400
65.857 ∠ 41.225o
105.39m + j92.341m
140.12m ∠ 41.225o
28.490m + j32.516m
43.232m ∠ 131.22o
28.490m + j32.516m
43.232m ∠ 131.22o
7.968  j6.981
Ohm’s
Law
Ohm’s
Law
10.594 ∠ 221.22o
57.501 + j50.382
76.451 ∠ 41.225
E = IZ E = IZ
Now, let’s turn to SPICE for a computer verification of our work:
1
0
2 3
4
5
6
0 0
3
more "dummy" voltage sources to
act as current measurement points
in the SPICE analysis (all set to 0
volts).C1
L
C2
R 470 Ω
1.5 µF
650 mH
4.7 µF
120 V
60 Hz
Vit
Vilc Vir
Figure 5.9: Example seriesparallel R, L, C SPICE circuit.
Each line of the SPICE output listing gives the voltage, voltage phase angle, current, and
current phase angle for C1, L, C2, and R, in that order. As you can see, these figures do concur
with our handcalculated figures in the circuit analysis table.
As daunting a task as seriesparallel AC circuit analysis may appear, it must be emphasized
that there is nothing really new going on here besides the use of complex numbers. Ohm’s Law
(in its new form of E=IZ) still holds true, as do the voltage and current Laws of Kirchhoff.
While there is more potential for human error in carrying out the necessary complex number
calculations, the basic principles and techniques of seriesparallel circuit reduction are exactly
the same.
118 CHAPTER 5. REACTANCE AND IMPEDANCE – R, L, AND C
ac seriesparallel rlc circuit
v1 1 0 ac 120 sin
vit 1 2 ac 0
vilc 3 4 ac 0
vir 3 6 ac 0
c1 2 3 4.7u
l 4 5 650m
c2 5 0 1.5u
r 6 0 470
.ac lin 1 60 60
.print ac v(2,3) vp(2,3) i(vit) ip(vit)
.print ac v(4,5) vp(4,5) i(vilc) ip(vilc)
.print ac v(5,0) vp(5,0) i(vilc) ip(vilc)
.print ac v(6,0) vp(6,0) i(vir) ip(vir)
.end
freq v(2,3) vp(2,3) i(vit) ip(vit) C1
6.000E+01 8.276E+01 3.163E+01 1.466E01 5.837E+01
freq v(4,5) vp(4,5) i(vilc) ip(vilc) L
6.000E+01 1.059E+01 1.388E+02 4.323E02 1.312E+02
freq v(5) vp(5) i(vilc) ip(vilc) C2
6.000E+01 7.645E+01 4.122E+01 4.323E02 1.312E+02
freq v(6) vp(6) i(vir) ip(vir) R
6.000E+01 6.586E+01 4.122E+01 1.401E01 4.122E+01
• REVIEW:
• Analysis of seriesparallel AC circuits is much the same as seriesparallel DC circuits.
The only substantive difference is that all figures and calculations are in complex (not
scalar) form.
• It is important to remember that before seriesparallel reduction (simplification) can be
gin, you must determine the impedance (Z) of every resistor, inductor, and capacitor. That
way, all component values will be expressed in common terms (Z) instead of an incompat
ible mix of resistance (R), inductance (L), and capacitance (C).
5.5. SUSCEPTANCE AND ADMITTANCE 119
5.5 Susceptance and Admittance
In the study of DC circuits, the student of electricity comes across a term meaning the oppo
site of resistance: conductance. It is a useful term when exploring the mathematical formula
for parallel resistances: Rparallel = 1 / (1/R1 + 1/R2 + . . . 1/Rn). Unlike resistance, which
diminishes as more parallel components are included in the circuit, conductance simply adds.
Mathematically, conductance is the reciprocal of resistance, and each 1/R term in the “parallel
resistance formula” is actually a conductance.
Whereas the term “resistance” denotes the amount of opposition to flowing electrons in
a circuit, “conductance” represents the ease of which electrons may flow. Resistance is the
measure of how much a circuit resists current, while conductance is the measure of how much
a circuit conducts current. Conductance used to be measured in the unit of mhos, or “ohms”
spelled backward. Now, the proper unit of measurement is Siemens. When symbolized in a
mathematical formula, the proper letter to use for conductance is “G”.
Reactive components such as inductors and capacitors oppose the flow of electrons with
respect to time, rather than with a constant, unchanging friction as resistors do. We call this
timebased opposition, reactance, and like resistance we also measure it in the unit of ohms.
As conductance is the complement of resistance, there is also a complementary expression
of reactance, called susceptance. Mathematically, it is equal to 1/X, the reciprocal of reactance.
Like conductance, it used to be measured in the unit of mhos, but now is measured in Siemens.
Its mathematical symbol is “B”, unfortunately the same symbol used to represent magnetic
flux density.
The terms “reactance” and “susceptance” have a certain linguistic logic to them, just like
resistance and conductance. While reactance is the measure of how much a circuit reacts
against change in current over time, susceptance is the measure of how much a circuit is
susceptible to conducting a changing current.
If one were tasked with determining the total effect of several parallelconnected, pure
reactances, one could convert each reactance (X) to a susceptance (B), then add susceptances
rather than diminish reactances: Xparallel = 1/(1/X1 + 1/X2 + . . . 1/Xn). Like conductances (G),
susceptances (B) add in parallel and diminish in series. Also like conductance, susceptance is
a scalar quantity.
When resistive and reactive components are interconnected, their combined effects can no
longer be analyzed with scalar quantities of resistance (R) and reactance (X). Likewise, figures
of conductance (G) and susceptance (B) are most useful in circuits where the two types of
opposition are not mixed, i.e. either a purely resistive (conductive) circuit, or a purely reactive
(susceptive) circuit. In order to express and quantify the effects of mixed resistive and reactive
components, we had to have a new term: impedance, measured in ohms and symbolized by the
letter “Z”.
To be consistent, we need a complementary measure representing the reciprocal of impedance.
The name for this measure is admittance. Admittance is measured in (guess what?) the unit
of Siemens, and its symbol is “Y”. Like impedance, admittance is a complex quantity rather
than scalar. Again, we see a certain logic to the naming of this new term: while impedance is
a measure of how much alternating current is impeded in a circuit, admittance is a measure of
how much current is admitted.
Given a scientific calculator capable of handling complex number arithmetic in both polar
and rectangular forms, you may never have to work with figures of susceptance (B) or admit
120 CHAPTER 5. REACTANCE AND IMPEDANCE – R, L, AND C
tance (Y). Be aware, though, of their existence and their meanings.
5.6 Summary
With the notable exception of calculations for power (P), all AC circuit calculations are based
on the same general principles as calculations for DC circuits. The only significant difference
is that fact that AC calculations use complex quantities while DC calculations use scalar quan
tities. Ohm’s Law, Kirchhoff ’s Laws, and even the network theorems learned in DC still hold
true for AC when voltage, current, and impedance are all expressed with complex numbers.
The same troubleshooting strategies applied toward DC circuits also hold for AC, although AC
can certainly be more difficult to work with due to phase angles which aren’t registered by a
handheld multimeter.
Power is another subject altogether, and will be covered in its own chapter in this book.
Because power in a reactive circuit is both absorbed and released – not just dissipated as it is
with resistors – its mathematical handling requires a more direct application of trigonometry
to solve.
When faced with analyzing an AC circuit, the first step in analysis is to convert all resistor,
inductor, and capacitor component values into impedances (Z), based on the frequency of the
power source. After that, proceed with the same steps and strategies learned for analyzing DC
circuits, using the “new” form of Ohm’s Law: E=IZ ; I=E/Z ; and Z=E/I
Remember that only the calculated figures expressed in polar form apply directly to empir
ical measurements of voltage and current. Rectangular notation is merely a useful tool for us
to add and subtract complex quantities together. Polar notation, where the magnitude (length
of vector) directly relates to the magnitude of the voltage or current measured, and the an
gle directly relates to the phase shift in degrees, is the most practical way to express complex
quantities for circuit analysis.
5.7 Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
Jason Starck (June 2000): HTML document formatting, which led to a much better
looking second edition.
Chapter 6
RESONANCE
Contents
6.1 An electric pendulum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
6.2 Simple parallel (tank circuit) resonance . . . . . . . . . . . . . . . . . . . . 126
6.3 Simple series resonance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
6.4 Applications of resonance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
6.5 Resonance in seriesparallel circuits . . . . . . . . . . . . . . . . . . . . . . 136
6.6 Q and bandwidth of a resonant circuit . . . . . . . . . . . . . . . . . . . . . 145
6.6.1 Series resonant circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
6.6.2 Parallel resonant circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
6.7 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
6.1 An electric pendulum
Capacitors store energy in the form of an electric field, and electrically manifest that stored
energy as a potential: static voltage. Inductors store energy in the form of a magnetic field, and
electrically manifest that stored energy as a kinetic motion of electrons: current. Capacitors
and inductors are flipsides of the same reactive coin, storing and releasing energy in comple
mentary modes. When these two types of reactive components are directly connected together,
their complementary tendencies to store energy will produce an unusual result.
If either the capacitor or inductor starts out in a charged state, the two components will
exchange energy between them, back and forth, creating their own AC voltage and current
cycles. If we assume that both components are subjected to a sudden application of voltage
(say, from a momentarily connected battery), the capacitor will very quickly charge and the
inductor will oppose change in current, leaving the capacitor in the charged state and the
inductor in the discharged state: (Figure 6.1)
The capacitor will begin to discharge, its voltage decreasing. Meanwhile, the inductor will
begin to build up a “charge” in the form of a magnetic field as current increases in the circuit:
(Figure 6.2)
121
122 CHAPTER 6. RESONANCE
+

capacitor charged: voltage at (+) peak
inductor discharged: zero current
e
i Time
Battery momentarily
connected to start the cycle e=
i=
Figure 6.1: Capacitor charged: voltage at (+) peak, inductor discharged: zero current.
+

Time
e=
i=
capacitor discharging: voltage decreasing
inductor charging: current increasing
Figure 6.2: Capacitor discharging: voltage decreasing, Inductor charging: current increasing.
The inductor, still charging, will keep electrons flowing in the circuit until the capacitor has
been completely discharged, leaving zero voltage across it: (Figure 6.3)
Time
e=
i=
capacitor fully discharged: zero voltage
inductor fully charged: maximum current
Figure 6.3: Capacitor fully discharged: zero voltage, inductor fully charged: maximum current.
The inductor will maintain current flow even with no voltage applied. In fact, it will gen
erate a voltage (like a battery) in order to keep current in the same direction. The capacitor,
being the recipient of this current, will begin to accumulate a charge in the opposite polarity
as before: (Figure 6.4)
When the inductor is finally depleted of its energy reserve and the electrons come to a halt,
the capacitor will have reached full (voltage) charge in the opposite polarity as when it started:
(Figure 6.5)
Now we’re at a condition very similar to where we started: the capacitor at full charge
and zero current in the circuit. The capacitor, as before, will begin to discharge through the
inductor, causing an increase in current (in the opposite direction as before) and a decrease in
voltage as it depletes its own energy reserve: (Figure 6.6)
Eventually the capacitor will discharge to zero volts, leaving the inductor fully charged with
6.1. AN ELECTRIC PENDULUM 123
Time
e=
i=

+
capacitor charging: voltage increasing (in opposite polarity)
inductor discharging: current decreasing
Figure 6.4: Capacitor charging: voltage increasing (in opposite polarity), inductor discharging:
current decreasing.
Time
e=
i=

+
capacitor fully charged: voltage at () peak
inductor fully discharged: zero current
Figure 6.5: Capacitor fully charged: voltage at () peak, inductor fully discharged: zero current.
Time
e=
i=

+
capacitor discharging: voltage decreasing
inductor charging: current increasing
Figure 6.6: Capacitor discharging: voltage decreasing, inductor charging: current increasing.
124 CHAPTER 6. RESONANCE
full current through it: (Figure 6.7)
Time
e=
i=
capacitor fully discharged: zero voltage
inductor fully charged: current at () peak
Figure 6.7: Capacitor fully discharged: zero voltage, inductor fully charged: current at () peak.
The inductor, desiring to maintain current in the same direction, will act like a source again,
generating a voltage like a battery to continue the flow. In doing so, the capacitor will begin to
charge up and the current will decrease in magnitude: (Figure 6.8)
Time
e=
i=

+
capacitor charging: voltage increasing
inductor discharging: current decreasing
Figure 6.8: Capacitor charging: voltage increasing, inductor discharging: current decreasing.
Eventually the capacitor will become fully charged again as the inductor expends all of its
energy reserves trying to maintain current. The voltage will once again be at its positive peak
and the current at zero. This completes one full cycle of the energy exchange between the
capacitor and inductor: (Figure 6.9)
Time
e=
i=

+
capacitor fully charged: voltage at (+) peak
inductor fully discharged: zero current
Figure 6.9: Capacitor fully charged: voltage at (+) peak, inductor fully discharged: zero current.
This oscillation will continue with steadily decreasing amplitude due to power losses from
stray resistances in the circuit, until the process stops altogether. Overall, this behavior is akin
to that of a pendulum: as the pendulum mass swings back and forth, there is a transformation
6.1. AN ELECTRIC PENDULUM 125
of energy taking place from kinetic (motion) to potential (height), in a similar fashion to the
way energy is transferred in the capacitor/inductor circuit back and forth in the alternating
forms of current (kinetic motion of electrons) and voltage (potential electric energy).
At the peak height of each swing of a pendulum, the mass briefly stops and switches di
rections. It is at this point that potential energy (height) is at a maximum and kinetic energy
(motion) is at zero. As the mass swings back the other way, it passes quickly through a point
where the string is pointed straight down. At this point, potential energy (height) is at zero and
kinetic energy (motion) is at maximum. Like the circuit, a pendulum’s backandforth oscilla
tion will continue with a steadily dampened amplitude, the result of air friction (resistance)
dissipating energy. Also like the circuit, the pendulum’s position and velocity measurements
trace two sine waves (90 degrees out of phase) over time: (Figure 6.10)
mass
zero potential energy,
maximum kinetic energy
maximum potential energy,
zero kinetic energy
potential energy =
kinetic energy =
Figure 6.10: Pendelum transfers energy between kinetic and potential energy as it swings low
to high.
In physics, this kind of natural sinewave oscillation for a mechanical system is called Sim
ple Harmonic Motion (often abbreviated as “SHM”). The same underlying principles govern
both the oscillation of a capacitor/inductor circuit and the action of a pendulum, hence the
similarity in effect. It is an interesting property of any pendulum that its periodic time is gov
erned by the length of the string holding the mass, and not the weight of the mass itself. That
is why a pendulum will keep swinging at the same frequency as the oscillations decrease in
amplitude. The oscillation rate is independent of the amount of energy stored in it.
The same is true for the capacitor/inductor circuit. The rate of oscillation is strictly depen
dent on the sizes of the capacitor and inductor, not on the amount of voltage (or current) at
each respective peak in the waves. The ability for such a circuit to store energy in the form of
126 CHAPTER 6. RESONANCE
oscillating voltage and current has earned it the name tank circuit. Its property of maintaining
a single, natural frequency regardless of how much or little energy is actually being stored in
it gives it special significance in electric circuit design.
However, this tendency to oscillate, or resonate, at a particular frequency is not limited to
circuits exclusively designed for that purpose. In fact, nearly any AC circuit with a combination
of capacitance and inductance (commonly called an “LC circuit”) will tend to manifest unusual
effects when the AC power source frequency approaches that natural frequency. This is true
regardless of the circuit’s intended purpose.
If the power supply frequency for a circuit exactly matches the natural frequency of the
circuit’s LC combination, the circuit is said to be in a state of resonance. The unusual effects
will reach maximum in this condition of resonance. For this reason, we need to be able to
predict what the resonant frequency will be for various combinations of L and C, and be aware
of what the effects of resonance are.
• REVIEW:
• A capacitor and inductor directly connected together form something called a tank circuit,
which oscillates (or resonates) at one particular frequency. At that frequency, energy is
alternately shuffled between the capacitor and the inductor in the form of alternating
voltage and current 90 degrees out of phase with each other.
• When the power supply frequency for an AC circuit exactly matches that circuit’s natural
oscillation frequency as set by the L and C components, a condition of resonance will have
been reached.
6.2 Simple parallel (tank circuit) resonance
A condition of resonance will be experienced in a tank circuit (Figure 6.11) when the reactances
of the capacitor and inductor are equal to each other. Because inductive reactance increases
with increasing frequency and capacitive reactance decreases with increasing frequency, there
will only be one frequency where these two reactances will be equal.
10 µF 100 mH
Figure 6.11: Simple parallel resonant circuit (tank circuit).
In the above circuit, we have a 10 µF capacitor and a 100 mH inductor. Since we know the
equations for determining the reactance of each at a given frequency, and we’re looking for that
6.2. SIMPLE PARALLEL (TANK CIRCUIT) RESONANCE 127
point where the two reactances are equal to each other, we can set the two reactance formulae
equal to each other and solve for frequency algebraically:
XL = 2pifL XC = 2pifC
1
. . . setting the two equal to each other,
representing a condition of equal reactance
(resonance) . . .
2pifL = 1
2pifC
Multiplying both sides by f eliminates the f
term in the denominator of the fraction . . .
2pif2L =
2piC
1
Dividing both sides by 2piL leaves f2 by itself
on the lefthand side of the equation . . .
f2 =
2pi2piLC
1
Taking the square root of both sides of the
equation leaves f by itself on the left side . . .
f =
2pi2piLC
1
. . . simplifying . . .
f =
LC2pi
1
So there we have it: a formula to tell us the resonant frequency of a tank circuit, given the
values of inductance (L) in Henrys and capacitance (C) in Farads. Plugging in the values of L
and C in our example circuit, we arrive at a resonant frequency of 159.155 Hz.
What happens at resonance is quite interesting. With capacitive and inductive reactances
equal to each other, the total impedance increases to infinity, meaning that the tank circuit
draws no current from the AC power source! We can calculate the individual impedances of
the 10 µF capacitor and the 100 mH inductor and work through the parallel impedance formula
to demonstrate this mathematically:
128 CHAPTER 6. RESONANCE
XL = 2pifL
XL = (2)(pi)(159.155 Hz)(100 mH)
XL = 100 Ω
XC = 2pifC
1
XC = (2)(pi)(159.155 Hz)(10 µF)
1
XC = 100 Ω
As you might have guessed, I chose these component values to give resonance impedances
that were easy to work with (100 Ω even). Now, we use the parallel impedance formula to see
what happens to total Z:
Zparallel =
ZL ZC
11
+
1
Zparallel =
100 Ω ∠ 90o 100 Ω ∠ 90o
1 1
+
1
Zparallel =
1
+0.01 ∠ 90o 0.01 ∠ 90o
Zparallel =
1
0
Undefined!
We can’t divide any number by zero and arrive at a meaningful result, but we can say
that the result approaches a value of infinity as the two parallel impedances get closer to each
other. What this means in practical terms is that, the total impedance of a tank circuit is
infinite (behaving as an open circuit) at resonance. We can plot the consequences of this over a
wide power supply frequency range with a short SPICE simulation: (Figure 6.12)
The 1 picoohm (1 pΩ) resistor is placed in this SPICE analysis to overcome a limitation
of SPICE: namely, that it cannot analyze a circuit containing a direct inductorvoltage source
loop. (Figure 6.12) A very low resistance value was chosen so as to have minimal effect on
circuit behavior.
This SPICE simulation plots circuit current over a frequency range of 100 to 200 Hz in
twenty even steps (100 and 200 Hz inclusive). Current magnitude on the graph increases from
6.2. SIMPLE PARALLEL (TANK CIRCUIT) RESONANCE 129
1
0 0
1 1
2
0 0
10 uFC1 L1 100 mH
Rbogus 1 pΩ
Figure 6.12: Resonant circuit sutitable for SPICE simulation.
freq i(v1) 3.162E04 1.000E03 3.162E03 1.0E02
                                
1.000E+02 9.632E03 . . . . *
1.053E+02 8.506E03 . . . . * .
1.105E+02 7.455E03 . . . . * .
1.158E+02 6.470E03 . . . . * .
1.211E+02 5.542E03 . . . . * .
1.263E+02 4.663E03 . . . . * .
1.316E+02 3.828E03 . . . .* .
1.368E+02 3.033E03 . . . *. .
1.421E+02 2.271E03 . . . * . .
1.474E+02 1.540E03 . . . * . .
1.526E+02 8.373E04 . . * . . .
1.579E+02 1.590E04 . * . . . .
1.632E+02 4.969E04 . . * . . .
1.684E+02 1.132E03 . . . * . .
1.737E+02 1.749E03 . . . * . .
1.789E+02 2.350E03 . . . * . .
1.842E+02 2.934E03 . . . *. .
1.895E+02 3.505E03 . . . .* .
1.947E+02 4.063E03 . . . . * .
2.000E+02 4.609E03 . . . . * .
                                
130 CHAPTER 6. RESONANCE
tank circuit frequency sweep
v1 1 0 ac 1 sin
c1 1 0 10u
* rbogus is necessary to eliminate a direct loop
* between v1 and l1, which SPICE can’t handle
rbogus 1 2 1e12
l1 2 0 100m
.ac lin 20 100 200
.plot ac i(v1)
.end
left to right, while frequency increases from top to bottom. The current in this circuit takes
a sharp dip around the analysis point of 157.9 Hz, which is the closest analysis point to our
predicted resonance frequency of 159.155 Hz. It is at this point that total current from the
power source falls to zero.
The plot above is produced from the above spice circuit file ( *.cir), the command (.plot) in the
last line producing the text plot on any printer or terminal. A better looking plot is produced
by the “nutmeg” graphical postprocessor, part of the spice package. The above spice ( *.cir)
does not require the plot (.plot) command, though it does no harm. The following commands
produce the plot below: (Figure 6.13)
spice b r resonant.raw resonant.cir
( b batch mode, r raw file, input is resonant.cir)
nutmeg resonant.raw
From the nutmeg prompt:
>setplot ac1 (setplot {enter} for list of plots)
>display (for list of signals)
>plot mag(v1#branch)
(magnitude of complex current vector v1#branch)
Incidentally, the graph output produced by this SPICE computer analysis is more generally
known as a Bode plot. Such graphs plot amplitude or phase shift on one axis and frequency on
the other. The steepness of a Bode plot curve characterizes a circuit’s “frequency response,” or
how sensitive it is to changes in frequency.
• REVIEW:
• Resonance occurs when capacitive and inductive reactances are equal to each other.
• For a tank circuit with no resistance (R), resonant frequency can be calculated with the
following formula:
•
fresonant =
2pi LC
1
• The total impedance of a parallel LC circuit approaches infinity as the power supply
frequency approaches resonance.
6.3. SIMPLE SERIES RESONANCE 131
Figure 6.13: Nutmeg produces plot of current I(v1) for parallel resonant circuit.
• A Bode plot is a graph plotting waveform amplitude or phase on one axis and frequency
on the other.
6.3 Simple series resonance
A similar effect happens in series inductive/capacitive circuits. (Figure 6.14) When a state of
resonance is reached (capacitive and inductive reactances equal), the two impedances cancel
each other out and the total impedance drops to zero!
10 µF
100 mH
Figure 6.14: Simple series resonant circuit.
132 CHAPTER 6. RESONANCE
At 159.155 Hz:
ZL = 0 + j100 Ω ZC = 0  j100 Ω
Zseries = ZL + ZC
Zseries = (0 + j100 Ω) + (0  j100 Ω)
Zseries = 0 Ω
With the total series impedance equal to 0 Ω at the resonant frequency of 159.155 Hz, the
result is a short circuit across the AC power source at resonance. In the circuit drawn above,
this would not be good. I’ll add a small resistor (Figure 6.15) in series along with the capacitor
and the inductor to keep the maximum circuit current somewhat limited, and perform another
SPICE analysis over the same range of frequencies: (Figure 6.16)
1 2
3
0 0
R1
1 V
1 Ω
C1
L1 100 mH
10 µF
Figure 6.15: Series resonant circuit suitable for SPICE.
series lc circuit
v1 1 0 ac 1 sin
r1 1 2 1
c1 2 3 10u
l1 3 0 100m
.ac lin 20 100 200
.plot ac i(v1)
.end
As before, circuit current amplitude increases from bottom to top, while frequency increases
from left to right. (Figure 6.16) The peak is still seen to be at the plotted frequency point of
157.9 Hz, the closest analyzed point to our predicted resonance point of 159.155 Hz. This would
suggest that our resonant frequency formula holds as true for simple series LC circuits as it
does for simple parallel LC circuits, which is the case:
6.3. SIMPLE SERIES RESONANCE 133
Figure 6.16: Series resonant circuit plot of current I(v1).
fresonant =
2pi LC
1
A word of caution is in order with series LC resonant circuits: because of the high currents
which may be present in a series LC circuit at resonance, it is possible to produce dangerously
high voltage drops across the capacitor and the inductor, as each component possesses signifi
cant impedance. We can edit the SPICE netlist in the above example to include a plot of voltage
across the capacitor and inductor to demonstrate what happens: (Figure 6.17)
series lc circuit
v1 1 0 ac 1 sin
r1 1 2 1
c1 2 3 10u
l1 3 0 100m
.ac lin 20 100 200
.plot ac i(v1) v(2,3) v(3)
.end
According to SPICE, voltage across the capacitor and inductor reach a peak somewhere
around 70 volts! This is quite impressive for a power supply that only generates 1 volt. Need
less to say, caution is in order when experimenting with circuits such as this. This SPICE
voltage is lower than the expected value due to the small (20) number of steps in the AC anal
ysis statement (.ac lin 20 100 200). What is the expected value?
Given: fr = 159.155 Hz, L = 100mH, R = 1
XL = 2pifL = 2pi(159.155)(100mH)=j100Ω
XC = 1/(2pifC) = 1/(2pi(159.155)(10µF)) = j100Ω
134 CHAPTER 6. RESONANCE
Figure 6.17: Plot of Vc=V(2,3) 70 V peak, VL=v(3) 70 V peak, I=I(V1#branch) 0.532 A peak
Z = 1 +j100 j100 = 1 Ω
I = V/Z = (1 V)/(1 Ω) = 1 A
VL = IZ = (1 A)(j100) = j100 V
VC = IZ = (1 A)(j100) = j100 V
VR = IR = (1 A)(1)= 1 V
Vtotal = VL + VC + VR
Vtotal = j100 j100 +1 = 1 V
The expected values for capacitor and inductor voltage are 100 V. This voltage will stress
these components to that level and they must be rated accordingly. However, these voltages
are out of phase and cancel yielding a total voltage across all three components of only 1 V, the
applied voltage. The ratio of the capacitor (or inductor) voltage to the applied voltage is the “Q”
factor.
Q = VL/VR = VC/VR
• REVIEW:
• The total impedance of a series LC circuit approaches zero as the power supply frequency
approaches resonance.
• The same formula for determining resonant frequency in a simple tank circuit applies to
simple series circuits as well.
• Extremely high voltages can be formed across the individual components of series LC
circuits at resonance, due to high current flows and substantial individual component
impedances.
6.4. APPLICATIONS OF RESONANCE 135
6.4 Applications of resonance
So far, the phenomenon of resonance appears to be a useless curiosity, or at most a nuisance
to be avoided (especially if series resonance makes for a shortcircuit across our AC voltage
source!). However, this is not the case. Resonance is a very valuable property of reactive AC
circuits, employed in a variety of applications.
One use for resonance is to establish a condition of stable frequency in circuits designed
to produce AC signals. Usually, a parallel (tank) circuit is used for this purpose, with the
capacitor and inductor directly connected together, exchanging energy between each other.
Just as a pendulum can be used to stabilize the frequency of a clock mechanism’s oscillations,
so can a tank circuit be used to stabilize the electrical frequency of an AC oscillator circuit. As
was noted before, the frequency set by the tank circuit is solely dependent upon the values of L
and C, and not on the magnitudes of voltage or current present in the oscillations: (Figure 6.18)
. . .
. . .
... to the rest of
the "oscillator"
circuit
the natural frequency
of the "tank circuit"
helps to stabilize
oscillations
Figure 6.18: Resonant circuit serves as stable frequency source.
Another use for resonance is in applications where the effects of greatly increased or de
creased impedance at a particular frequency is desired. A resonant circuit can be used to
“block” (present high impedance toward) a frequency or range of frequencies, thus acting as
a sort of frequency “filter” to strain certain frequencies out of a mix of others. In fact, these
particular circuits are called filters, and their design constitutes a discipline of study all by
itself: (Figure 6.19)
load
AC source of
mixed frequencies
Tank circuit presents a
high impedance to a narrow
range of frequencies, blocking
them from getting to the load
Figure 6.19: Resonant circuit serves as filter.
In essence, this is how analog radio receiver tuner circuits work to filter, or select, one
station frequency out of the mix of different radio station frequency signals intercepted by the
antenna.
• REVIEW:
136 CHAPTER 6. RESONANCE
• Resonance can be employed to maintain AC circuit oscillations at a constant frequency,
just as a pendulum can be used to maintain constant oscillation speed in a timekeeping
mechanism.
• Resonance can be exploited for its impedance properties: either dramatically increas
ing or decreasing impedance for certain frequencies. Circuits designed to screen certain
frequencies out of a mix of different frequencies are called filters.
6.5 Resonance in seriesparallel circuits
In simple reactive circuits with little or no resistance, the effects of radically altered impedance
will manifest at the resonance frequency predicted by the equation given earlier. In a parallel
(tank) LC circuit, this means infinite impedance at resonance. In a series LC circuit, it means
zero impedance at resonance:
fresonant =
2pi LC
1
However, as soon as significant levels of resistance are introduced into most LC circuits,
this simple calculation for resonance becomes invalid. We’ll take a look at several LC circuits
with added resistance, using the same values for capacitance and inductance as before: 10 µF
and 100 mH, respectively. According to our simple equation, the resonant frequency should
be 159.155 Hz. Watch, though, where current reaches maximum or minimum in the following
SPICE analyses:
1
0 0
1 1
2
0 0
V1 1 V
C1 L1
R1 100 Ω
100 mH10 µF
Parallel LC with resistance in series with L
Figure 6.20: Parallel LC circuit with resistance in series with L.
Here, an extra resistor (Rbogus) (Figure 6.22)is necessary to prevent SPICE from encounter
ing trouble in analysis. SPICE can’t handle an inductor connected directly in parallel with any
voltage source or any other inductor, so the addition of a series resistor is necessary to “break
6.5. RESONANCE IN SERIESPARALLEL CIRCUITS 137
resonant circuit
v1 1 0 ac 1 sin
c1 1 0 10u
r1 1 2 100
l1 2 0 100m
.ac lin 20 100 200
.plot ac i(v1)
.end
Figure 6.21: Resistance in series with L produces minimum current at 136.8 Hz instead of
calculated 159.2 Hz
Minimum current at 136.8 Hz instead of 159.2 Hz!
138 CHAPTER 6. RESONANCE
1
0 0
1 1
2
0 0
3
Parallel LC with resistance in series with C
V1 1 V
R1
C1 L1 100 mH
Rbogus100 Ω
10 µF
Figure 6.22: Parallel LC with resistance in serieis with C.
up” the voltage source/inductor loop that would otherwise be formed. This resistor is chosen to
be a very low value for minimum impact on the circuit’s behavior.
resonant circuit
v1 1 0 ac 1 sin
r1 1 2 100
c1 2 0 10u
rbogus 1 3 1e12
l1 3 0 100m
.ac lin 20 100 400
.plot ac i(v1)
.end
Minimum current at roughly 180 Hz instead of 159.2 Hz!
Switching our attention to series LC circuits, (Figure 6.24) we experiment with placing
significant resistances in parallel with either L or C. In the following series circuit examples,
a 1 Ω resistor (R1) is placed in series with the inductor and capacitor to limit total current at
resonance. The “extra” resistance inserted to influence resonant frequency effects is the 100 Ω
resistor, R2. The results are shown in (Figure 6.25).
And finally, a series LC circuit with the significant resistance in parallel with the capacitor.
(Figure 6.26) The shifted resonance is shown in (Figure 6.27)
The tendency for added resistance to skew the point at which impedance reaches a maxi
mum or minimum in an LC circuit is called antiresonance. The astute observer will notice a
pattern between the four SPICE examples given above, in terms of how resistance affects the
resonant peak of a circuit:
6.5. RESONANCE IN SERIESPARALLEL CIRCUITS 139
Figure 6.23: Resistance in series with C shifts minimum current from calculated 159.2 Hz to
roughly 180 Hz.
1
0 0
2
3 3
0
Series LC with resistance in parallel with L
V1 1 V
R1
1 Ω
C1 10 µF
L1 100 mH R2 100 Ω
Figure 6.24: Series LC resonant circuit with resistance in parallel with L.
140 CHAPTER 6. RESONANCE
resonant circuit
v1 1 0 ac 1 sin
r1 1 2 1
c1 2 3 10u
l1 3 0 100m
r2 3 0 100
.ac lin 20 100 400
.plot ac i(v1)
.end
Maximum current at roughly 178.9 Hz instead of 159.2 Hz!
Figure 6.25: Series resonant circuit with resistance in parallel with L shifts maximum current
from 159.2 Hz to roughly 180 Hz.
resonant circuit
v1 1 0 ac 1 sin
r1 1 2 1
c1 2 3 10u
r2 2 3 100
l1 3 0 100m
.ac lin 20 100 200
.plot ac i(v1)
.end
Maximum current at 136.8 Hz instead of 159.2 Hz!
6.5. RESONANCE IN SERIESPARALLEL CIRCUITS 141
1
0 0
2
3 3
2
Series LC with resistance in parallel with C
V1 1 V
R1
1 Ω
C1
L1 100 mH
10 µF R2 100 Ω
Figure 6.26: Series LC resonant circuit with rsistance in parallel with C.
Figure 6.27: Resistance in parallel with C in series resonant circuit shifts curreent maximum
from calculated 159.2 Hz to about 136.8 Hz.
142 CHAPTER 6. RESONANCE
• Parallel (“tank”) LC circuit:
• R in series with L: resonant frequency shifted down
• R in series with C: resonant frequency shifted up
• Series LC circuit:
• R in parallel with L: resonant frequency shifted up
• R in parallel with C: resonant frequency shifted down
Again, this illustrates the complementary nature of capacitors and inductors: how resis
tance in series with one creates an antiresonance effect equivalent to resistance in parallel
with the other. If you look even closer to the four SPICE examples given, you’ll see that the
frequencies are shifted by the same amount, and that the shape of the complementary graphs
are mirrorimages of each other!
Antiresonance is an effect that resonant circuit designers must be aware of. The equations
for determining antiresonance “shift” are complex, and will not be covered in this brief lesson.
It should suffice the beginning student of electronics to understand that the effect exists, and
what its general tendencies are.
Added resistance in an LC circuit is no academic matter. While it is possible to manufacture
capacitors with negligible unwanted resistances, inductors are typically plagued with substan
tial amounts of resistance due to the long lengths of wire used in their construction. What
is more, the resistance of wire tends to increase as frequency goes up, due to a strange phe
nomenon known as the skin effect where AC current tends to be excluded from travel through
the very center of a wire, thereby reducing the wire’s effective crosssectional area. Thus,
inductors not only have resistance, but changing, frequencydependent resistance at that.
As if the resistance of an inductor’s wire weren’t enough to cause problems, we also have to
contend with the “core losses” of ironcore inductors, which manifest themselves as added re
sistance in the circuit. Since iron is a conductor of electricity as well as a conductor of magnetic
flux, changing flux produced by alternating current through the coil will tend to induce electric
currents in the core itself (eddy currents). This effect can be thought of as though the iron
core of the transformer were a sort of secondary transformer coil powering a resistive load: the
lessthanperfect conductivity of the iron metal. This effects can be minimized with laminated
cores, good core design and highgrade materials, but never completely eliminated.
One notable exception to the rule of circuit resistance causing a resonant frequency shift
is the case of series resistorinductorcapacitor (“RLC”) circuits. So long as all components are
connected in series with each other, the resonant frequency of the circuit will be unaffected by
the resistance. (Figure 6.28) The resulting plot is shown in (Figure 6.29).
Maximum current at 159.2 Hz once again!
Note that the peak of the current graph (Figure 6.29) has not changed from the earlier series
LC circuit (the one with the 1 Ω token resistance in it), even though the resistance is now 100
times greater. The only thing that has changed is the “sharpness” of the curve. Obviously, this
circuit does not resonate as strongly as one with less series resistance (it is said to be “less
selective”), but at least it has the same natural frequency!
6.5. RESONANCE IN SERIESPARALLEL CIRCUITS 143
1
0 0
2
3
Series LC with resistance in series
V1 1 V
R1
100 Ω
C1 10 µF
L1 100 mH
Figure 6.28: Series LC with resistance in series.
series rlc circuit
v1 1 0 ac 1 sin
r1 1 2 100
c1 2 3 10u
l1 3 0 100m
.ac lin 20 100 200
.plot ac i(v1)
.end
Figure 6.29: Resistance in series resonant circuit leaves current maximum at calculated 159.2
Hz, broadening the curve.
144 CHAPTER 6. RESONANCE
It is noteworthy that antiresonance has the effect of dampening the oscillations of free
running LC circuits such as tank circuits. In the beginning of this chapter we saw how a
capacitor and inductor connected directly together would act something like a pendulum, ex
changing voltage and current peaks just like a pendulum exchanges kinetic and potential en
ergy. In a perfect tank circuit (no resistance), this oscillation would continue forever, just as a
frictionless pendulum would continue to swing at its resonant frequency forever. But friction
less machines are difficult to find in the real world, and so are lossless tank circuits. Energy
lost through resistance (or inductor core losses or radiated electromagnetic waves or . . .) in a
tank circuit will cause the oscillations to decay in amplitude until they are no more. If enough
energy losses are present in a tank circuit, it will fail to resonate at all.
Antiresonance’s dampening effect is more than just a curiosity: it can be used quite ef
fectively to eliminate unwanted oscillations in circuits containing stray inductances and/or
capacitances, as almost all circuits do. Take note of the following L/R time delay circuit: (Fig
ure 6.30)
switch R
L
Figure 6.30: L/R time delay circuit
The idea of this circuit is simple: to “charge” the inductor when the switch is closed. The
rate of inductor charging will be set by the ratio L/R, which is the time constant of the circuit
in seconds. However, if you were to build such a circuit, you might find unexpected oscillations
(AC) of voltage across the inductor when the switch is closed. (Figure 6.31) Why is this? There’s
no capacitor in the circuit, so how can we have resonant oscillation with just an inductor,
resistor, and battery?
All inductors contain a certain amount of stray capacitance due to turntoturn and turn
tocore insulation gaps. Also, the placement of circuit conductors may create stray capacitance.
While clean circuit layout is important in eliminating much of this stray capacitance, there
will always be some that you cannot eliminate. If this causes resonant problems (unwanted
AC oscillations), added resistance may be a way to combat it. If resistor R is large enough, it
will cause a condition of antiresonance, dissipating enough energy to prohibit the inductance
and stray capacitance from sustaining oscillations for very long.
Interestingly enough, the principle of employing resistance to eliminate unwanted reso
nance is one frequently used in the design of mechanical systems, where any moving object
with mass is a potential resonator. A very common application of this is the use of shock ab
sorbers in automobiles. Without shock absorbers, cars would bounce wildly at their resonant
frequency after hitting any bump in the road. The shock absorber’s job is to introduce a strong
antiresonant effect by dissipating energy hydraulically (in the same way that a resistor dissi
6.6. Q AND BANDWIDTH OF A RESONANT CIRCUIT 145
ideal L/R voltage curve =
actual L/R voltage curve =
Figure 6.31: Inductor ringing due to resonance with stray capacitance.
pates energy electrically).
• REVIEW:
• Added resistance to an LC circuit can cause a condition known as antiresonance, where
the peak impedance effects happen at frequencies other than that which gives equal ca
pacitive and inductive reactances.
• Resistance inherent in realworld inductors can contribute greatly to conditions of an
tiresonance. One source of such resistance is the skin effect, caused by the exclusion of
AC current from the center of conductors. Another source is that of core losses in ironcore
inductors.
• In a simple series LC circuit containing resistance (an “RLC” circuit), resistance does not
produce antiresonance. Resonance still occurs when capacitive and inductive reactances
are equal.
6.6 Q and bandwidth of a resonant circuit
The Q, quality factor, of a resonant circuit is a measure of the “goodness” or quality of a reso
nant circuit. A higher value for this figure of merit correspondes to a more narrow bandwith,
which is desirable in many applications. More formally, Q is the ration of power stored to power
dissipated in the circuit reactance and resistance, respectively:
Q = Pstored/Pdissipated = I2X/I2R
Q = X/R
where: X = Capacitive or Inductive reactance at resonance
R = Series resistance.
This formula is applicable to series resonant circuits, and also parallel resonant circuits if
the resistance is in series with the inductor. This is the case in practical applications, as we
146 CHAPTER 6. RESONANCE
are mostly concerned with the resistance of the inductor limiting the Q. Note: Some text may
show X and R interchanged in the “Q” formula for a parallel resonant circuit. This is correct
for a large value of R in parallel with C and L. Our formula is correct for a small R in series
with L.
A practical application of “Q” is that voltage across L or C in a series resonant circuit is Q
times total applied voltage. In a parallel resonant circuit, current through L or C is Q times
the total applied current.
6.6.1 Series resonant circuits
A series resonant circuit looks like a resistance at the resonant frequency. (Figure 6.32) Since
the definition of resonance is XL=XC , the reactive components cancel, leaving only the resis
tance to contribute to the impedance. The impedance is also at a minimum at resonance.
(Figure 6.33) Below the resonant frequency, the series resonant circuit looks capacitive since
the impedance of the capacitor increases to a value greater than the decreasing inducitve re
actance, leaving a net capacitive value. Above resonance, the inductive rectance increases,
capacitive reactance decreases, leaving a net inductive component.
Figure 6.32: At resonance the series resonant circuit appears purely resistive. Below resonance
it looks capacitive. Above resonance it appears inductive.
Current is maximum at resonance, impedance at a minumum. Current is set by the value
of the resistance. Above or below resonance, impedance increases.
The resonant current peak may be changed by varying the series resistor, which changes
the Q. (Figure 6.34) This also affects the broadness of the curve. A low resistance, high Q
circuit has a narrow bandwidth, as compared to a high resistance, low Q circuit. Bandwidth in
terms of Q and resonant frequency:
6.6. Q AND BANDWIDTH OF A RESONANT CIRCUIT 147
Figure 6.33: Impedance is at a minumum at resonance in a series resonant circuit.
BW = fc/Q
Where fc = resonant frquency
Q = quality factor
Figure 6.34: A high Q resonant circuit has a narrow bandwidth as compared to a low Q
148 CHAPTER 6. RESONANCE
Bandwidth is measured between the 0.707 current amplitude points. The 0.707 current
points correspond to the half power points since P = I2R, (0.707)2 = (0.5). (Figure 6.35)
Figure 6.35: Bandwidth, ∆f is measured between the 70.7% amplitude points of series resonant
circuit.
BW = ∆f = fhfl = fc/Q
Where fh = high band edge, fl = low band edge
fl = fc  ∆f/2
fh = fc + ∆f/2
Where fc = center frequency (resonant frequency)
In Figure 6.35, the 100% current point is 50 mA. The 70.7% level is 0707(50 mA)=35.4 mA.
The upper and lower band edges read from the curve are 291 Hz for fl and 355 Hz for fh. The
bandwidth is 64 Hz, and the half power points are ± 32 Hz of the center resonant frequency:
BW = ∆f = fhfl = 355291 = 64
fl = fc  ∆f/2 = 32332 = 291
fh = fc + ∆f/2 = 323+32 = 355
Since BW = fc/Q:
Q = fc/BW = (323 Hz)/(64 Hz) = 5
6.6.2 Parallel resonant circuits
A parallel resonant circuit is resistive at the resonant frequency. (Figure 6.36) At resonance
XL=XC , the reactive components cancel. The impedance is maximum at resonance. (Fig
6.6. Q AND BANDWIDTH OF A RESONANT CIRCUIT 149
ure 6.37) Below the resonant frequency, the parallel resonant circuit looks inductive since the
impedance of the inductor is lower, drawing the larger proportion of current. Above resonance,
the capacitive rectance decreases, drawing the larger current, thus, taking on a capacitive
characteristic.
Figure 6.36: A parallel resonant circuit is resistive at resonance, inductive below resonance,
capacitive above resonance.
Impedance is maximum at resonance in a parallel resonant circuit, but decreases above or
below resonance. Voltage is at a peak at resonance since voltage is proportional to impedance
(E=IZ). (Figure 6.37)
A low Q due to a high resistance in series with the inductor produces a low peak on a broad
response curve for a parallel resonant circuit. (Figure 6.38) conversely, a high Q is due to a low
resistance in series with the inductor. This produces a higher peak in the narrower response
curve. The high Q is achieved by winding the inductor with larger diameter (smaller gague),
lower resistance wire.
The bandwidth of the parallel resonant response curve is measured between the half power
points. This corresponds to the 70.7% voltage points since power is proportional to E2. ((0.707)2=0.50)
Since voltage is proportional to impedance, we may use the impedance curve. (Figure 6.39)
In Figure 6.39, the 100% impedance point is 500 Ω. The 70.7% level is 0707(500)=354 Ω.
The upper and lower band edges read from the curve are 281 Hz for fl and 343 Hz for fh. The
bandwidth is 62 Hz, and the half power points are ± 31 Hz of the center resonant frequency:
BW = ∆f = fhfl = 343281 = 62
fl = fc  ∆f/2 = 31231 = 281
fh = fc + ∆f/2 = 312+31 = 343
Q = fc/BW = (312 Hz)/(62 Hz) = 5
150 CHAPTER 6. RESONANCE
Figure 6.37: Parallel resonant circuit: Impedance peaks at resonance.
Figure 6.38: Parallel resonant response varies with Q.
6.7. CONTRIBUTORS 151
Figure 6.39: Bandwidth, ∆f is measured between the 70.7% impedance points of a parallel
resonant circuit.
6.7 Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
Jason Starck (June 2000): HTML document formatting, which led to a much better
looking second edition.
152 CHAPTER 6. RESONANCE
Chapter 7
MIXEDFREQUENCY AC
SIGNALS
Contents
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
7.2 Square wave signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
7.3 Other waveshapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
7.4 More on spectrum analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
7.5 Circuit effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
7.6 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
7.1 Introduction
In our study of AC circuits thus far, we’ve explored circuits powered by a singlefrequency
sine voltage waveform. In many applications of electronics, though, singlefrequency signals
are the exception rather than the rule. Quite often we may encounter circuits where multiple
frequencies of voltage coexist simultaneously. Also, circuit waveforms may be something other
than sinewave shaped, in which case we call them nonsinusoidal waveforms.
Additionally, we may encounter situations where DC is mixed with AC: where a waveform is
superimposed on a steady (DC) signal. The result of such a mix is a signal varying in intensity,
but never changing polarity, or changing polarity asymmetrically (spending more time positive
than negative, for example). Since DC does not alternate as AC does, its “frequency” is said
to be zero, and any signal containing DC along with a signal of varying intensity (AC) may
be rightly called a mixedfrequency signal as well. In any of these cases where there is a mix
of frequencies in the same circuit, analysis is more complex than what we’ve seen up to this
point.
Sometimes mixedfrequency voltage and current signals are created accidentally. This may
be the result of unintended connections between circuits – called coupling – made possible by
153
154 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
stray capacitance and/or inductance between the conductors of those circuits. A classic example
of coupling phenomenon is seen frequently in industry where DC signal wiring is placed in
close proximity to AC power wiring. The nearby presence of high AC voltages and currents may
cause “foreign” voltages to be impressed upon the length of the signal wiring. Stray capacitance
formed by the electrical insulation separating power conductors from signal conductors may
cause voltage (with respect to earth ground) from the power conductors to be impressed upon
the signal conductors, while stray inductance formed by parallel runs of wire in conduit may
cause current from the power conductors to electromagnetically induce voltage along the signal
conductors. The result is a mix of DC and AC at the signal load. The following schematic shows
how an AC “noise” source may “couple” to a DC circuit through mutual inductance (Mstray) and
capacitance (Cstray) along the length of the conductors. (Figure 7.1)
"Noise"
source
Cstray
Mstray
Zwire Zwire Zwire
"Clean" DC voltage DC voltage + AC "noise"
Figure 7.1: Stray inductance and capacitance couple stray AC into desired DC signal.
When stray AC voltages from a “noise” source mix with DC signals conducted along signal
wiring, the results are usually undesirable. For this reason, power wiring and lowlevel signal
wiring should always be routed through separated, dedicated metal conduit, and signals should
be conducted via 2conductor “twisted pair” cable rather than through a single wire and ground
connection: (Figure 7.2)
The grounded cable shield – a wire braid or metal foil wrapped around the two insulated
conductors – isolates both conductors from electrostatic (capacitive) coupling by blocking any
external electric fields, while the parallal proximity of the two conductors effectively cancels
any electromagnetic (mutually inductive) coupling because any induced noise voltage will be
approximately equal in magnitude and opposite in phase along both conductors, canceling each
other at the receiving end for a net (differential) noise voltage of almost zero. Polarity marks
placed near each inductive portion of signal conductor length shows how the induced voltages
are phased in such a way as to cancel one another.
Coupling may also occur between two sets of conductors carrying AC signals, in which case
both signals may become “mixed” with each other: (Figure 7.3)
7.1. INTRODUCTION 155
"Noise"
source
Cstray
Mstray
Shielded cable
+
+
Figure 7.2: Shielded twisted pair minimized noise.
Cstray
Mstray
Zwire Zwire Zwire
Zwire Zwire Zwire
Signal A
Signal B
A + B
B + A
Figure 7.3: Coupling of AC signals between parallel conductors.
156 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
Coupling is but one example of how signals of different frequencies may become mixed.
Whether it be AC mixed with DC, or two AC signals mixing with each other, signal coupling via
stray inductance and capacitance is usually accidental and undesired. In other cases, mixed
frequency signals are the result of intentional design or they may be an intrinsic quality of a
signal. It is generally quite easy to create mixedfrequency signal sources. Perhaps the easiest
way is to simply connect voltage sources in series: (Figure 7.4)
mixedfrequency
AC voltage
AC + DC
voltage
60 Hz
90 Hz
Figure 7.4: Series connection of voltage sources mixes signals.
Some computer communications networks operate on the principle of superimposing high
frequency voltage signals along 60 Hz powerline conductors, so as to convey computer data
along existing lengths of power cabling. This technique has been used for years in electric
power distribution networks to communicate load data along highvoltage power lines. Cer
tainly these are examples of mixedfrequency AC voltages, under conditions that are deliber
ately established.
In some cases, mixedfrequency signals may be produced by a single voltage source. Such is
the case with microphones, which convert audiofrequency air pressure waves into correspond
ing voltage waveforms. The particular mix of frequencies in the voltage signal output by the
microphone is dependent on the sound being reproduced. If the sound waves consist of a single,
pure note or tone, the voltage waveform will likewise be a sine wave at a single frequency. If
the sound wave is a chord or other harmony of several notes, the resulting voltage waveform
produced by the microphone will consist of those frequencies mixed together. Very few natural
sounds consist of single, pure sine wave vibrations but rather are a mix of different frequency
vibrations at different amplitudes.
Musical chords are produced by blending one frequency with other frequencies of particular
fractional multiples of the first. However, investigating a little further, we find that even a
single piano note (produced by a plucked string) consists of one predominant frequency mixed
with several other frequencies, each frequency a wholenumber multiple of the first (called
harmonics, while the first frequency is called the fundamental). An illustration of these terms
is shown in Table 7.1 with a fundamental frequency of 1000 Hz (an arbitrary figure chosen for
this example).
Sometimes the term “overtone” is used to describe the a harmonic frequency produced by
a musical instrument. The “first” overtone is the first harmonic frequency greater than the
fundamental. If we had an instrument producing the entire range of harmonic frequencies
shown in the table above, the first overtone would be 2000 Hz (the 2nd harmonic), while the
second overtone would be 3000 Hz (the 3rd harmonic), etc. However, this application of the
term “overtone” is specific to particular instruments.
7.1. INTRODUCTION 157
Table 7.1: For a “base” frequency of 1000 Hz:
Frequency (Hz) Term
1000 1st harmonic, or fundamental
2000 2nd harmonic
3000 3rd harmonic
4000 4th harmonic
5000 5th harmonic
6000 6th harmonic
7000 7th harmonic
It so happens that certain instruments are incapable of producing certain types of harmonic
frequencies. For example, an instrument made from a tube that is open on one end and closed
on the other (such as a bottle, which produces sound when air is blown across the opening)
is incapable of producing evennumbered harmonics. Such an instrument set up to produce a
fundamental frequency of 1000 Hz would also produce frequencies of 3000 Hz, 5000 Hz, 7000
Hz, etc, but would not produce 2000 Hz, 4000 Hz, 6000 Hz, or any other evenmultiple fre
quencies of the fundamental. As such, we would say that the first overtone (the first frequency
greater than the fundamental) in such an instrument would be 3000 Hz (the 3rd harmonic),
while the second overtone would be 5000 Hz (the 5th harmonic), and so on.
A pure sine wave (single frequency), being entirely devoid of any harmonics, sounds very
“flat” and “featureless” to the human ear. Most musical instruments are incapable of producing
sounds this simple. What gives each instrument its distinctive tone is the same phenomenon
that gives each person a distinctive voice: the unique blending of harmonic waveforms with
each fundamental note, described by the physics of motion for each unique object producing
the sound.
Brass instruments do not possess the same “harmonic content” as woodwind instruments,
and neither produce the same harmonic content as stringed instruments. A distinctive blend
of frequencies is what gives a musical instrument its characteristic tone. As anyone who has
played guitar can tell you, steel strings have a different sound than nylon strings. Also, the
tone produced by a guitar string changes depending on where along its length it is plucked.
These differences in tone, as well, are a result of different harmonic content produced by dif
ferences in the mechanical vibrations of an instrument’s parts. All these instruments produce
harmonic frequencies (wholenumber multiples of the fundamental frequency) when a single
note is played, but the relative amplitudes of those harmonic frequencies are different for dif
ferent instruments. In musical terms, the measure of a tone’s harmonic content is called timbre
or color.
Musical tones become even more complex when the resonating element of an instrument
is a twodimensional surface rather than a onedimensional string. Instruments based on the
vibration of a string (guitar, piano, banjo, lute, dulcimer, etc.) or of a column of air in a tube
(trumpet, flute, clarinet, tuba, pipe organ, etc.) tend to produce sounds composed of a single
frequency (the “fundamental”) and a mix of harmonics. Instruments based on the vibration
of a flat plate (steel drums, and some types of bells), however, produce a much broader range
of frequencies, not limited to wholenumber multiples of the fundamental. The result is a
distinctive tone that some people find acoustically offensive.
158 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
As you can see, music provides a rich field of study for mixed frequencies and their effects.
Later sections of this chapter will refer to musical instruments as sources of waveforms for
analysis in more detail.
• REVIEW:
• A sinusoidal waveform is one shaped exactly like a sine wave.
• A nonsinusoidal waveform can be anything from a distorted sinewave shape to some
thing completely different like a square wave.
• Mixedfrequency waveforms can be accidently created, purposely created, or simply exist
out of necessity. Most musical tones, for instance, are not composed of a single frequency
sinewave, but are rich blends of different frequencies.
• When multiple sine waveforms are mixed together (as is often the case in music), the
lowest frequency sinewave is called the fundamental, and the other sinewaves whose
frequencies are wholenumber multiples of the fundamental wave are called harmonics.
• An overtone is a harmonic produced by a particular device. The “first” overtone is the first
frequency greater than the fundamental, while the “second” overtone is the next greater
frequency produced. Successive overtones may or may not correspond to incremental
harmonics, depending on the device producing the mixed frequencies. Some devices and
systems do not permit the establishment of certain harmonics, and so their overtones
would only include some (not all) harmonic frequencies.
7.2 Square wave signals
It has been found that any repeating, nonsinusoidal waveform can be equated to a combination
of DC voltage, sine waves, and/or cosine waves (sine waves with a 90 degree phase shift) at
various amplitudes and frequencies. This is true no matter how strange or convoluted the
waveform in question may be. So long as it repeats itself regularly over time, it is reducible
to this series of sinusoidal waves. In particular, it has been found that square waves are
mathematically equivalent to the sum of a sine wave at that same frequency, plus an infinite
series of oddmultiple frequency sine waves at diminishing amplitude:
7.2. SQUARE WAVE SIGNALS 159
4
pi
1 V (peak) repeating square wave at 50 Hz is equivalent to:
(1 V peak sine wave at 50 Hz)
+ 4
pi
(1/3 V peak sine wave at 150 Hz)
+ 4
pi
(1/5 V peak sine wave at 250 Hz)
+ 4
pi
(1/7 V peak sine wave at 350 Hz)
+ 4
pi
(1/9 V peak sine wave at 450 Hz)
+ . . . ad infinitum . . .
This truth about waveforms at first may seem too strange to believe. However, if a square
wave is actually an infinite series of sine wave harmonics added together, it stands to reason
that we should be able to prove this by adding together several sine wave harmonics to pro
duce a close approximation of a square wave. This reasoning is not only sound, but easily
demonstrated with SPICE.
The circuit we’ll be simulating is nothing more than several sine wave AC voltage sources
of the proper amplitudes and frequencies connected together in series. We’ll use SPICE to plot
the voltage waveforms across successive additions of voltage sources, like this: (Figure 7.5)
plot voltage waveform
plot voltage waveform
plot voltage waveform
plot voltage waveform
plot voltage waveformV1=1.27V
50Hz
V3=424mV
150Hz
V5=255mV
250Hz
V7=182mV
350Hz
V9=141mV
450Hz
Figure 7.5: A square wave is approximated by the sum of harmonics.
In this particular SPICE simulation, I’ve summed the 1st, 3rd, 5th, 7th, and 9th harmonic
160 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
voltage sources in series for a total of five AC voltage sources. The fundamental frequency is
50 Hz and each harmonic is, of course, an integer multiple of that frequency. The amplitude
(voltage) figures are not random numbers; rather, they have been arrived at through the equa
tions shown in the frequency series (the fraction 4/pi multiplied by 1, 1/3, 1/5, 1/7, etc. for each
of the increasing odd harmonics).
building a squarewave
v1 1 0 sin (0 1.27324 50 0 0) 1st harmonic (50 Hz)
v3 2 1 sin (0 424.413m 150 0 0) 3rd harmonic
v5 3 2 sin (0 254.648m 250 0 0) 5th harmonic
v7 4 3 sin (0 181.891m 350 0 0) 7th harmonic
v9 5 4 sin (0 141.471m 450 0 0) 9th harmonic
r1 5 0 10k
.tran 1m 20m
.plot tran v(1,0) Plot 1st harmonic
.plot tran v(2,0) Plot 1st + 3rd harmonics
.plot tran v(3,0) Plot 1st + 3rd + 5th harmonics
.plot tran v(4,0) Plot 1st + 3rd + 5th + 7th harmonics
.plot tran v(5,0) Plot 1st + . . . + 9th harmonics
.end
I’ll narrate the analysis step by step from here, explaining what it is we’re looking at. In
this first plot, we see the fundamentalfrequency sinewave of 50 Hz by itself. It is nothing but
a pure sine shape, with no additional harmonic content. This is the kind of waveform produced
by an ideal AC power source: (Figure 7.6)
Figure 7.6: Pure 50 Hz sinewave.
Next, we see what happens when this clean and simple waveform is combined with the
7.2. SQUARE WAVE SIGNALS 161
third harmonic (three times 50 Hz, or 150 Hz). Suddenly, it doesn’t look like a clean sine wave
any more: (Figure 7.7)
Figure 7.7: Sum of 1st (50 Hz) and 3rd (150 Hz) harmonics approximates a 50 Hz square wave.
The rise and fall times between positive and negative cycles are much steeper now, and the
crests of the wave are closer to becoming flat like a squarewave. Watch what happens as we
add the next odd harmonic frequency: (Figure 7.8)
Figure 7.8: Sum of 1st, 3rd and 5th harmonics approximates square wave.
162 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
The most noticeable change here is how the crests of the wave have flattened even more.
There are more several dips and crests at each end of the wave, but those dips and crests are
smaller in amplitude than they were before. Watch again as we add the next odd harmonic
waveform to the mix: (Figure 7.9)
Figure 7.9: Sum of 1st, 3rd, 5th, and 7th harmonics approximates square wave.
Here we can see the wave becoming flatter at each peak. Finally, adding the 9th harmonic,
the fifth sine wave voltage source in our circuit, we obtain this result: (Figure 7.10)
The end result of adding the first five odd harmonic waveforms together (all at the proper
amplitudes, of course) is a close approximation of a square wave. The point in doing this is to
illustrate how we can build a square wave up from multiple sine waves at different frequencies,
to prove that a pure square wave is actually equivalent to a series of sine waves. When a square
wave AC voltage is applied to a circuit with reactive components (capacitors and inductors),
those components react as if they were being exposed to several sine wave voltages of different
frequencies, which in fact they are.
The fact that repeating, nonsinusoidal waves are equivalent to a definite series of additive
DC voltage, sine waves, and/or cosine waves is a consequence of how waves work: a fundamen
tal property of all waverelated phenomena, electrical or otherwise. The mathematical process
of reducing a nonsinusoidal wave into these constituent frequencies is called Fourier analysis,
the details of which are well beyond the scope of this text. However, computer algorithms have
been created to perform this analysis at high speeds on real waveforms, and its application in
AC power quality and signal analysis is widespread.
SPICE has the ability to sample a waveform and reduce it into its constituent sine wave
harmonics by way of a Fourier Transform algorithm, outputting the frequency analysis as a
table of numbers. Let’s try this on a square wave, which we already know is composed of
oddharmonic sine waves:
The pulse option in the netlist line describing voltage source v1 instructs SPICE to simulate
7.2. SQUARE WAVE SIGNALS 163
Figure 7.10: Sum of 1st, 3rd, 5th, 7th and 9th harmonics approximates square wave.
squarewave analysis netlist
v1 1 0 pulse (1 1 0 .1m .1m 10m 20m)
r1 1 0 10k
.tran 1m 40m
.plot tran v(1,0)
.four 50 v(1,0)
.end
164 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
a squareshaped “pulse” waveform, in this case one that is symmetrical (equal time for each
halfcycle) and has a peak amplitude of 1 volt. First we’ll plot the square wave to be analyzed:
(Figure 7.11)
Figure 7.11: Squarewave for SPICE Fourier analysis
Next, we’ll print the Fourier analysis generated by SPICE for this square wave:
fourier components of transient response v(1)
dc component = 2.439E02
harmonic frequency fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 5.000E+01 1.274E+00 1.000000 2.195 0.000
2 1.000E+02 4.892E02 0.038415 94.390 92.195
3 1.500E+02 4.253E01 0.333987 6.585 4.390
4 2.000E+02 4.936E02 0.038757 98.780 96.585
5 2.500E+02 2.562E01 0.201179 10.976 8.780
6 3.000E+02 5.010E02 0.039337 103.171 100.976
7 3.500E+02 1.841E01 0.144549 15.366 13.171
8 4.000E+02 5.116E02 0.040175 107.561 105.366
9 4.500E+02 1.443E01 0.113316 19.756 17.561
total harmonic distortion = 43.805747 percent
Here, (Figure 7.12) SPICE has broken the waveform down into a spectrum of sinusoidal
frequencies up to the ninth harmonic, plus a small DC voltage labelled DC component. I
had to inform SPICE of the fundamental frequency (for a square wave with a 20 millisecond
period, this frequency is 50 Hz), so it knew how to classify the harmonics. Note how small the
figures are for all the even harmonics (2nd, 4th, 6th, 8th), and how the amplitudes of the odd
harmonics diminish (1st is largest, 9th is smallest).
7.2. SQUARE WAVE SIGNALS 165
Figure 7.12: Plot of Fourier analysis esults.
This same technique of “Fourier Transformation” is often used in computerized power in
strumentation, sampling the AC waveform(s) and determining the harmonic content thereof.
A common computer algorithm (sequence of program steps to perform a task) for this is the
Fast Fourier Transform or FFT function. You need not be concerned with exactly how these
computer routines work, but be aware of their existence and application.
This samemathematical technique used in SPICE to analyze the harmonic content of waves
can be applied to the technical analysis of music: breaking up any particular sound into its con
stituent sinewave frequencies. In fact, you may have already seen a device designed to do just
that without realizing what it was! A graphic equalizer is a piece of highfidelity stereo equip
ment that controls (and sometimes displays) the nature of music’s harmonic content. Equipped
with several knobs or slide levers, the equalizer is able to selectively attenuate (reduce) the
amplitude of certain frequencies present in music, to “customize” the sound for the listener’s
benefit. Typically, there will be a “bar graph” display next to each control lever, displaying the
amplitude of each particular frequency. (Figure 7.13)
50 150 300 500 750 1 1.5 3.5 5 7.5 10 12.5
Hz Hz Hz Hz Hz kHz kHz kHz kHz kHz kHz kHz
Graphic Equalizer
amplitude of each
frequency
Control levers set
the attenuation factor
for each frequency
Bargraph displays the
Figure 7.13: HiFi audio graphic equalizer.
166 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
A device built strictly to display – not control – the amplitudes of each frequency range for
a mixedfrequency signal is typically called a spectrum analyzer. The design of spectrum ana
lyzers may be as simple as a set of “filter” circuits (see the next chapter for details) designed to
separate the different frequencies from each other, or as complex as a specialpurpose digital
computer running an FFT algorithm to mathematically split the signal into its harmonic com
ponents. Spectrum analyzers are often designed to analyze extremely highfrequency signals,
such as those produced by radio transmitters and computer network hardware. In that form,
they often have an appearance like that of an oscilloscope: (Figure 7.14)
Spectrum Analyzer
frequency
amplitude
Figure 7.14: Spectrum analyzer shows amplitude as a function of frequency.
Like an oscilloscope, the spectrum analyzer uses a CRT (or a computer display mimicking a
CRT) to display a plot of the signal. Unlike an oscilloscope, this plot is amplitude over frequency
rather than amplitude over time. In essence, a frequency analyzer gives the operator a Bode
plot of the signal: something an engineer might call a frequencydomain rather than a time
domain analysis.
The term “domain” is mathematical: a sophisticated word to describe the horizontal axis of
a graph. Thus, an oscilloscope’s plot of amplitude (vertical) over time (horizontal) is a “time
domain” analysis, whereas a spectrum analyzer’s plot of amplitude (vertical) over frequency
(horizontal) is a “frequencydomain” analysis. When we use SPICE to plot signal amplitude
(either voltage or current amplitude) over a range of frequencies, we are performing frequency
domain analysis.
Please take note of how the Fourier analysis from the last SPICE simulation isn’t “perfect.”
Ideally, the amplitudes of all the even harmonics should be absolutely zero, and so should the
DC component. Again, this is not so much a quirk of SPICE as it is a property of waveforms
in general. A waveform of infinite duration (infinite number of cycles) can be analyzed with
absolute precision, but the less cycles available to the computer for analysis, the less precise
the analysis. It is only when we have an equation describing a waveform in its entirety that
7.2. SQUARE WAVE SIGNALS 167
Fourier analysis can reduce it to a definite series of sinusoidal waveforms. The fewer times
that a wave cycles, the less certain its frequency is. Taking this concept to its logical extreme,
a short pulse – a waveform that doesn’t even complete a cycle – actually has no frequency, but
rather acts as an infinite range of frequencies. This principle is common to all wavebased
phenomena, not just AC voltages and currents.
Suffice it to say that the number of cycles and the certainty of a waveform’s frequency com
ponent(s) are directly related. We could improve the precision of our analysis here by letting
the wave oscillate on and on for many cycles, and the result would be a spectrum analysis more
consistent with the ideal. In the following analysis, I’ve omitted the waveform plot for brevity’s
sake – its just a really long square wave:
squarewave
v1 1 0 pulse (1 1 0 .1m .1m 10m 20m)
r1 1 0 10k
.option limpts=1001
.tran 1m 1
.plot tran v(1,0)
.four 50 v(1,0)
.end
fourier components of transient response v(1)
dc component = 9.999E03
harmonic frequency fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 5.000E+01 1.273E+00 1.000000 1.800 0.000
2 1.000E+02 1.999E02 0.015704 86.382 88.182
3 1.500E+02 4.238E01 0.332897 5.400 3.600
4 2.000E+02 1.997E02 0.015688 82.764 84.564
5 2.500E+02 2.536E01 0.199215 9.000 7.200
6 3.000E+02 1.994E02 0.015663 79.146 80.946
7 3.500E+02 1.804E01 0.141737 12.600 10.800
8 4.000E+02 1.989E02 0.015627 75.529 77.329
9 4.500E+02 1.396E01 0.109662 16.199 14.399
Notice how this analysis (Figure 7.15) shows less of a DC component voltage and lower
amplitudes for each of the even harmonic frequency sine waves, all because we let the computer
sample more cycles of the wave. Again, the imprecision of the first analysis is not so much a
flaw in SPICE as it is a fundamental property of waves and of signal analysis.
• REVIEW:
• Square waves are equivalent to a sine wave at the same (fundamental) frequency added
to an infinite series of oddmultiple sinewave harmonics at decreasing amplitudes.
• Computer algorithms exist which are able to sample waveshapes and determine their
constituent sinusoidal components. The Fourier Transform algorithm (particularly the
168 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
Figure 7.15: Improved fourier analysis.
Fast Fourier Transform, or FFT) is commonly used in computer circuit simulation pro
grams such as SPICE and in electronic metering equipment for determining power qual
ity.
7.3 Other waveshapes
As strange as it may seem, any repeating, nonsinusoidal waveform is actually equivalent to a
series of sinusoidal waveforms of different amplitudes and frequencies added together. Square
waves are a very common and wellunderstood case, but not the only one.
Electronic power control devices such as transistors and siliconcontrolled rectifiers (SCRs)
often produce voltage and current waveforms that are essentially choppedup versions of the
otherwise “clean” (pure) sinewave AC from the power supply. These devices have the ability
to suddenly change their resistance with the application of a control signal voltage or cur
rent, thus “turning on” or “turning off” almost instantaneously, producing current waveforms
bearing little resemblance to the source voltage waveform powering the circuit. These current
waveforms then produce changes in the voltage waveform to other circuit components, due to
voltage drops created by the nonsinusoidal current through circuit impedances.
Circuit components that distort the normal sinewave shape of AC voltage or current are
called nonlinear. Nonlinear components such as SCRs find popular use in power electronics
due to their ability to regulate large amounts of electrical power without dissipating much
heat. While this is an advantage from the perspective of energy efficiency, the waveshape
distortions they introduce can cause problems.
These nonsinusoidal waveforms, regardless of their actual shape, are equivalent to a series
of sinusoidal waveforms of higher (harmonic) frequencies. If not taken into consideration by
the circuit designer, these harmonic waveforms created by electronic switching components
may cause erratic circuit behavior. It is becoming increasingly common in the electric power
industry to observe overheating of transformers and motors due to distortions in the sine
7.3. OTHER WAVESHAPES 169
wave shape of the AC power line voltage stemming from “switching” loads such as computers
and highefficiency lights. This is no theoretical exercise: it is very real and potentially very
troublesome.
In this section, I will investigate a few of the more common waveshapes and show their
harmonic components by way of Fourier analysis using SPICE.
One very common way harmonics are generated in an AC power system is when AC is
converted, or “rectified” into DC. This is generally done with components called diodes, which
only allow the passage of current in one direction. The simplest type of AC/DC rectification is
halfwave, where a single diode blocks half of the AC current (over time) from passing through
the load. (Figure 7.16) Oddly enough, the conventional diode schematic symbol is drawn such
that electrons flow against the direction of the symbol’s arrowhead:
load
diode
1 2
0 0
+

The diode only allows electron
flow in a counterclockwise
direction.
Figure 7.16: Halfwave rectifier.
halfwave rectifier
v1 1 0 sin(0 15 60 0 0)
rload 2 0 10k
d1 1 2 mod1
.model mod1 d
.tran .5m 17m
.plot tran v(1,0) v(2,0)
.four 60 v(1,0) v(2,0)
.end
halfwave rectifier
First, we’ll see how SPICE analyzes the source waveform, a pure sine wave voltage: (Fig
ure 7.18)
Notice the extremely small harmonic and DC components of this sinusoidal waveform in
the table above, though, too small to show on the harmonic plot above. Ideally, there would be
nothing but the fundamental frequency showing (being a perfect sine wave), but our Fourier
analysis figures aren’t perfect because SPICE doesn’t have the luxury of sampling a wave
form of infinite duration. Next, we’ll compare this with the Fourier analysis of the halfwave
170 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
Figure 7.17: Halfwave rectifier waveforms. V(1)+0.4 shifts the sinewave input V(1) up for
clarity. This is not part of the simulation.
fourier components of transient response v(1)
dc component = 8.016E04
harmonic frequency fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 1.482E+01 1.000000 0.005 0.000
2 1.200E+02 2.492E03 0.000168 104.347 104.342
3 1.800E+02 6.465E04 0.000044 86.663 86.658
4 2.400E+02 1.132E03 0.000076 61.324 61.319
5 3.000E+02 1.185E03 0.000080 70.091 70.086
6 3.600E+02 1.092E03 0.000074 63.607 63.602
7 4.200E+02 1.220E03 0.000082 56.288 56.283
8 4.800E+02 1.354E03 0.000091 54.669 54.664
9 5.400E+02 1.467E03 0.000099 52.660 52.655
7.3. OTHER WAVESHAPES 171
Figure 7.18: Fourier analysis of the sine wave input.
“rectified” voltage across the load resistor: (Figure 7.19)
fourier components of transient response v(2)
dc component = 4.456E+00
harmonic frequency fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 7.000E+00 1.000000 0.195 0.000
2 1.200E+02 3.016E+00 0.430849 89.765 89.570
3 1.800E+02 1.206E01 0.017223 168.005 167.810
4 2.400E+02 5.149E01 0.073556 87.295 87.100
5 3.000E+02 6.382E02 0.009117 152.790 152.595
6 3.600E+02 1.727E01 0.024676 79.362 79.167
7 4.200E+02 4.492E02 0.006417 132.420 132.224
8 4.800E+02 7.493E02 0.010703 61.479 61.284
9 5.400E+02 4.051E02 0.005787 115.085 114.889
Notice the relatively large evenmultiple harmonics in this analysis. By cutting out half of
our AC wave, we’ve introduced the equivalent of several higherfrequency sinusoidal (actually,
cosine) waveforms into our circuit from the original, pure sinewave. Also take note of the
large DC component: 4.456 volts. Because our AC voltage waveform has been “rectified” (only
allowed to push in one direction across the load rather than backandforth), it behaves a lot
more like DC.
Another method of AC/DC conversion is called fullwave (Figure 7.20), which as you may
have guessed utilizes the full cycle of AC power from the source, reversing the polarity of
half the AC cycle to get electrons to flow through the load the same direction all the time.
I won’t bore you with details of exactly how this is done, but we can examine the waveform
(Figure 7.21) and its harmonic analysis through SPICE: (Figure 7.22)
What a difference! According to SPICE’s Fourier transform, we have a 2nd harmonic com
ponent to this waveform that’s over 85 times the amplitude of the original AC source frequency!
172 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
Figure 7.19: Fourier analysis halfwave output.
+ 
1
0
32
1
0
V1 15 V60 Hz
D1 D3
D2 D4
Rload
10 kΩ
Figure 7.20: Fullwave rectifier circuit.
fullwave bridge rectifier
v1 1 0 sin(0 15 60 0 0)
rload 2 3 10k
d1 1 2 mod1
d2 0 2 mod1
d3 3 1 mod1
d4 3 0 mod1
.model mod1 d
.tran .5m 17m
.plot tran v(1,0) v(2,3)
.four 60 v(2,3)
.end
7.3. OTHER WAVESHAPES 173
Figure 7.21: Waveforms for fullwave rectifier
fourier components of transient response v(2,3)
dc component = 8.273E+00
harmonic frequency fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 7.000E02 1.000000 93.519 0.000
2 1.200E+02 5.997E+00 85.669415 90.230 3.289
3 1.800E+02 7.241E02 1.034465 93.787 0.267
4 2.400E+02 1.013E+00 14.465161 92.492 1.027
5 3.000E+02 7.364E02 1.052023 95.026 1.507
6 3.600E+02 3.337E01 4.767350 100.271 6.752
7 4.200E+02 7.496E02 1.070827 94.023 0.504
8 4.800E+02 1.404E01 2.006043 118.839 25.319
9 5.400E+02 7.457E02 1.065240 90.907 2.612
174 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
Figure 7.22: Fourier analysis of fullwave rectifier output.
The DC component of this wave shows up as being 8.273 volts (almost twice what is was for the
halfwave rectifier circuit) while the second harmonic is almost 6 volts in amplitude. Notice
all the other harmonics further on down the table. The odd harmonics are actually stronger at
some of the higher frequencies than they are at the lower frequencies, which is interesting.
As you can see, what may begin as a neat, simple AC sinewave may end up as a complex
mess of harmonics after passing through just a few electronic components. While the complex
mathematics behind all this Fourier transformation is not necessary for the beginning student
of electric circuits to understand, it is of the utmost importance to realize the principles at work
and to grasp the practical effects that harmonic signals may have on circuits. The practical
effects of harmonic frequencies in circuits will be explored in the last section of this chapter,
but before we do that we’ll take a closer look at waveforms and their respective harmonics.
• REVIEW:
• Any waveform at all, so long as it is repetitive, can be reduced to a series of sinusoidal
waveforms added together. Different waveshapes consist of different blends of sinewave
harmonics.
• Rectification of AC to DC is a very common source of harmonics within industrial power
systems.
7.4 More on spectrum analysis
Computerized Fourier analysis, particularly in the form of the FFT algorithm, is a powerful
tool for furthering our understanding of waveforms and their related spectral components. This
same mathematical routine programmed into the SPICE simulator as the .fourier option is
also programmed into a variety of electronic test instruments to perform realtime Fourier
analysis on measured signals. This section is devoted to the use of such tools and the analysis
of several different waveforms.
7.4. MORE ON SPECTRUM ANALYSIS 175
First we have a simple sine wave at a frequency of 523.25 Hz. This particular frequency
value is a “C” pitch on a piano keyboard, one octave above “middle C”. Actually, the signal
measured for this demonstration was created by an electronic keyboard set to produce the tone
of a panflute, the closest instrument “voice” I could find resembling a perfect sine wave. The
plot below was taken from an oscilloscope display, showing signal amplitude (voltage) over
time: (Figure 7.23)
Figure 7.23: Oscilloscope display: voltage vs time.
Viewed with an oscilloscope, a sine wave looks like a wavy curve traced horizontally on the
screen. The horizontal axis of this oscilloscope display is marked with the word “Time” and an
arrow pointing in the direction of time’s progression. The curve itself, of course, represents the
cyclic increase and decrease of voltage over time.
Close observation reveals imperfections in the sinewave shape. This, unfortunately, is a
result of the specific equipment used to analyze the waveform. Characteristics like these due
to quirks of the test equipment are technically known as artifacts: phenomena existing solely
because of a peculiarity in the equipment used to perform the experiment.
If we view this same AC voltage on a spectrum analyzer, the result is quite different: (Fig
ure 7.24)
As you can see, the horizontal axis of the display is marked with the word “Frequency,”
denoting the domain of this measurement. The single peak on the curve represents the pre
dominance of a single frequency within the range of frequencies covered by the width of the
display. If the scale of this analyzer instrument were marked with numbers, you would see
that this peak occurs at 523.25 Hz. The height of the peak represents the signal amplitude
(voltage).
If we mix three different sinewave tones together on the electronic keyboard (CEG, a C
176 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
Figure 7.24: Spectrum analyzer display: voltage vs frequency.
major chord) and measure the result, both the oscilloscope display and the spectrum analyzer
display reflect this increased complexity: (Figure 7.25)
The oscilloscope display (timedomain) shows a waveform with many more peaks and val
leys than before, a direct result of the mixing of these three frequencies. As you will notice,
some of these peaks are higher than the peaks of the original singlepitch waveform, while
others are lower. This is a result of the three different waveforms alternately reinforcing and
canceling each other as their respective phase shifts change in time.
The spectrum display (frequencydomain) is much easier to interpret: each pitch is rep
resented by its own peak on the curve. (Figure 7.26) The difference in height between these
three peaks is another artifact of the test equipment: a consequence of limitations within the
equipment used to generate and analyze these waveforms, and not a necessary characteristic
of the musical chord itself.
As was stated before, the device used to generate these waveforms is an electronic keyboard:
a musical instrument designed to mimic the tones of many different instruments. The panflute
“voice” was chosen for the first demonstrations because it most closely resembled a pure sine
wave (a single frequency on the spectrum analyzer display). Other musical instrument “voices”
are not as simple as this one, though. In fact, the unique tone produced by any instrument is a
function of its waveshape (or spectrum of frequencies). For example, let’s view the signal for a
trumpet tone: (Figure 7.27)
The fundamental frequency of this tone is the same as in the first panflute example: 523.25
Hz, one octave above “middle C.” The waveform itself is far from a pure and simple sine
wave form. Knowing that any repeating, nonsinusoidal waveform is equivalent to a series of
sinusoidal waveforms at different amplitudes and frequencies, we should expect to see multiple
7.4. MORE ON SPECTRUM ANALYSIS 177
Figure 7.25: Oscilloscape display: three tones.
Figure 7.26: Spectrum analyzer display: three tones.
178 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
Figure 7.27: Oscilloscope display: waveshape of a trumpet tone.
peaks on the spectrum analyzer display: (Figure 7.28)
Figure 7.28: Spectrum of a trumpet tone.
7.4. MORE ON SPECTRUM ANALYSIS 179
Indeed we do! The fundamental frequency component of 523.25 Hz is represented by the
leftmost peak, with each successive harmonic represented as its own peak along the width of
the analyzer screen. The second harmonic is twice the frequency of the fundamental (1046.5
Hz), the third harmonic three times the fundamental (1569.75 Hz), and so on. This display
only shows the first six harmonics, but there are many more comprising this complex tone.
Trying a different instrument voice (the accordion) on the keyboard, we obtain a simi
larly complex oscilloscope (timedomain) plot (Figure 7.29) and spectrum analyzer (frequency
domain) display: (Figure 7.30)
Figure 7.29: Oscilloscope display: waveshape of accordion tone.
Note the differences in relative harmonic amplitudes (peak heights) on the spectrum dis
plays for trumpet and accordion. Both instrument tones contain harmonics all the way from
1st (fundamental) to 6th (and beyond!), but the proportions aren’t the same. Each instrument
has a unique harmonic “signature” to its tone. Bear in mind that all this complexity is in ref
erence to a single note played with these two instrument “voices.” Multiple notes played on an
accordion, for example, would create a much more complex mixture of frequencies than what
is seen here.
The analytical power of the oscilloscope and spectrum analyzer permit us to derive gen
eral rules about waveforms and their harmonic spectra from real waveform examples. We
already know that any deviation from a pure sinewave results in the equivalent of a mixture
of multiple sinewave waveforms at different amplitudes and frequencies. However, close ob
servation allows us to be more specific than this. Note, for example, the time (Figure 7.31) and
frequencydomain (Figure 7.32) plots for a waveform approximating a square wave:
According to the spectrum analysis, this waveform contains no even harmonics, only odd.
Although this display doesn’t show frequencies past the sixth harmonic, the pattern of oddonly
180 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
Figure 7.30: Spectrum of accordion tone.
Figure 7.31: Oscilloscope timedomain display of a square wave
7.4. MORE ON SPECTRUM ANALYSIS 181
Figure 7.32: Spectrum (frequencydomain) of a square wave.
harmonics in descending amplitude continues indefinitely. This should come as no surprise, as
we’ve already seen with SPICE that a square wave is comprised of an infinitude of odd har
monics. The trumpet and accordion tones, however, contained both even and odd harmonics.
This difference in harmonic content is noteworthy. Let’s continue our investigation with an
analysis of a triangle wave: (Figure 7.33)
In this waveform there are practically no even harmonics: (Figure 7.34) the only significant
frequency peaks on the spectrum analyzer display belong to oddnumbered multiples of the
fundamental frequency. Tiny peaks can be seen for the second, fourth, and sixth harmonics,
but this is due to imperfections in this particular triangle waveshape (once again, artifacts
of the test equipment used in this analysis). A perfect triangle waveshape produces no even
harmonics, just like a perfect square wave. It should be obvious from inspection that the
harmonic spectrum of the triangle wave is not identical to the spectrum of the square wave:
the respective harmonic peaks are of different heights. However, the two different waveforms
are common in their lack of even harmonics.
Let’s examine another waveform, this one very similar to the triangle wave, except that
its risetime is not the same as its falltime. Known as a sawtooth wave, its oscilloscope plot
reveals it to be aptly named: (Figure 7.35)
When the spectrum analysis of this waveform is plotted, we see a result that is quite dif
ferent from that of the regular triangle wave, for this analysis shows the strong presence of
evennumbered harmonics (second and fourth): (Figure 7.36)
The distinction between a waveform having even harmonics versus no even harmonics re
sides in the difference between a triangle waveshape and a sawtooth waveshape. That differ
ence is symmetry above and below the horizontal centerline of the wave. A waveform that is
182 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
Figure 7.33: Oscilloscope timedomain display of a triangle wave.
Figure 7.34: Spectrum of a triangle wave.
7.4. MORE ON SPECTRUM ANALYSIS 183
Figure 7.35: Timedomain display of a sawtooth wave.
Figure 7.36: Frequencydomain display of a sawtooth wave.
184 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
symmetrical above and below its centerline (the shape on both sides mirror each other pre
cisely) will contain no evennumbered harmonics. (Figure 7.37)
Pure sine wave =
1st harmonic only
Figure 7.37: Waveforms symmetric about their xaxis center line contain only odd harmonics.
Square waves, triangle waves, and pure sine waves all exhibit this symmetry, and all are de
void of even harmonics. Waveforms like the trumpet tone, the accordion tone, and the sawtooth
wave are unsymmetrical around their centerlines and therefore do contain even harmonics.
(Figure 7.38)
Figure 7.38: Asymmetric waveforms contain even harmonics.
This principle of centerline symmetry should not be confused with symmetry around the
zero line. In the examples shown, the horizontal centerline of the waveform happens to be zero
volts on the timedomain graph, but this has nothing to do with harmonic content. This rule
of harmonic content (even harmonics only with unsymmetrical waveforms) applies whether
or not the waveform is shifted above or below zero volts with a “DC component.” For further
clarification, I will show the same sets of waveforms, shifted with DC voltage, and note that
their harmonic contents are unchanged. (Figure 7.39)
Pure sine wave =
1st harmonic only
Figure 7.39: These waveforms are composed exclusively of odd harmonics.
Again, the amount of DC voltage present in a waveform has nothing to do with that wave
form’s harmonic frequency content. (Figure 7.40)
7.5. CIRCUIT EFFECTS 185
Figure 7.40: These waveforms contain even harmonics.
Why is this harmonic ruleofthumb an important rule to know? It can help us comprehend
the relationship between harmonics in AC circuits and specific circuit components. Since most
sources of sinewave distortion in AC power circuits tend to be symmetrical, evennumbered
harmonics are rarely seen in those applications. This is good to know if you’re a power system
designer and are planning ahead for harmonic reduction: you only have to concern yourself
with mitigating the odd harmonic frequencies, even harmonics being practically nonexistent.
Also, if you happen to measure even harmonics in an AC circuit with a spectrum analyzer or
frequency meter, you know that something in that circuit must be unsymmetrically distorting
the sinewave voltage or current, and that clue may be helpful in locating the source of a prob
lem (look for components or conditions more likely to distort one halfcycle of the AC waveform
more than the other).
Now that we have this rule to guide our interpretation of nonsinusoidal waveforms, it
makes more sense that a waveform like that produced by a rectifier circuit should contain
such strong even harmonics, there being no symmetry at all above and below center.
• REVIEW:
• Waveforms that are symmetrical above and below their horizontal centerlines contain no
evennumbered harmonics.
• The amount of DC “bias” voltage present (a waveform’s “DC component”) has no impact
on that wave’s harmonic frequency content.
7.5 Circuit effects
The principle of nonsinusoidal, repeating waveforms being equivalent to a series of sine waves
at different frequencies is a fundamental property of waves in general and it has great practical
import in the study of AC circuits. It means that any time we have a waveform that isn’t
perfectly sinewaveshaped, the circuit in question will react as though its having an array of
different frequency voltages imposed on it at once.
When an AC circuit is subjected to a source voltage consisting of a mixture of frequencies,
the components in that circuit respond to each constituent frequency in a different way. Any
reactive component such as a capacitor or an inductor will simultaneously present a unique
amount of impedance to each and every frequency present in a circuit. Thankfully, the analysis
of such circuits is made relatively easy by applying the Superposition Theorem, regarding the
multiplefrequency source as a set of singlefrequency voltage sources connected in series, and
186 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
analyzing the circuit for one source at a time, summing the results at the end to determine the
aggregate total:
60 Hz
90 Hz
5 V
5 V
R
2.2 kΩ
C 1 µF
Figure 7.41: Circuit driven by a combination of frequencies: 60 Hz and 90 Hz.
Analyzing circuit for 60 Hz source alone:
5 V
60 Hz
R
C 1 µF
2.2 kΩ
XC = 2.653 kΩ
Figure 7.42: Circuit for solving 60 Hz.
E
I
Volts
Amps
OhmsZ
R TotalC
2.2k + j0
2.2k ∠ 0o
0  j2.653k
2.653k ∠ 90o
2.2k  j2.653k
3.446k ∠ 50.328o
5 + j0
5 ∠ 0o
926.22µ + j1.1168m
1.4509m ∠ 50.328o
926.22µ + j1.1168m
1.4509m ∠ 50.328o
926.22µ + j1.1168m
1.4509m ∠ 50.328o
2.0377 + j2.4569 2.9623  j2.4569
3.8486 ∠ 39.6716o3.1919 ∠ 50.328o
Analyzing the circuit for 90 Hz source alone:
5 V
90 Hz
R
C 1 µF
2.2 kΩ
XC = 1.768 kΩ
Figure 7.43: Circuit of solving 90 Hz.
7.5. CIRCUIT EFFECTS 187
E
I
Volts
Amps
OhmsZ
R TotalC
2.2k + j0
2.2k ∠ 0o
5 + j0
5 ∠ 0o
0  j1.768k
1.768k ∠ 90o
2.2k  j1.768k
2.823k ∠ 38.793o
1.3807m + j1.1098m
1.7714m ∠ 38.793o
1.3807m + j1.1098m
1.7714m ∠ 38.793o
1.3807m + j1.1098m
1.7714m ∠ 38.793o
3.0375 + j2.4415
3.8971 ∠ 38.793o
1.9625  j2.4415
3.1325 ∠ 51.207o
Superimposing the voltage drops across R and C, we get:
ER = [3.1919 V ∠ 50.328o (60 Hz)] + [3.8971 V ∠ 38.793o (90 Hz)]
EC = [3.8486 V ∠ 39.6716o (60 Hz)] + [3.1325 V ∠ 51.207o (90 Hz)]
Because the two voltages across each component are at different frequencies, we cannot con
solidate them into a single voltage figure as we could if we were adding together two voltages
of different amplitude and/or phase angle at the same frequency. Complex number notation
give us the ability to represent waveform amplitude (polar magnitude) and phase angle (polar
angle), but not frequency.
What we can tell from this application of the superposition theorem is that there will be a
greater 60 Hz voltage dropped across the capacitor than a 90 Hz voltage. Just the opposite is
true for the resistor’s voltage drop. This is worthy to note, especially in light of the fact that the
two source voltages are equal. It is this kind of unequal circuit response to signals of differing
frequency that will be our specific focus in the next chapter.
We can also apply the superposition theorem to the analysis of a circuit powered by a non
sinusoidal voltage, such as a square wave. If we know the Fourier series (multiple sine/cosine
wave equivalent) of that wave, we can regard it as originating from a seriesconnected string
of multiple sinusoidal voltage sources at the appropriate amplitudes, frequencies, and phase
shifts. Needless to say, this can be a laborious task for some waveforms (an accurate square
wave Fourier Series is considered to be expressed out to the ninth harmonic, or five sine waves
in all!), but it is possible. I mention this not to scare you, but to inform you of the potential
complexity lurking behind seemingly simple waveforms. A reallife circuit will respond just the
same to being powered by a square wave as being powered by an infinite series of sine waves
of oddmultiple frequencies and diminishing amplitudes. This has been known to translate
into unexpected circuit resonances, transformer and inductor core overheating due to eddy
currents, electromagnetic noise over broad ranges of the frequency spectrum, and the like.
Technicians and engineers need to be made aware of the potential effects of nonsinusoidal
waveforms in reactive circuits.
Harmonics are known to manifest their effects in the form of electromagnetic radiation
as well. Studies have been performed on the potential hazards of using portable computers
aboard passenger aircraft, citing the fact that computers’ high frequency squarewave “clock”
voltage signals are capable of generating radio waves that could interfere with the operation
of the aircraft’s electronic navigation equipment. It’s bad enough that typical microprocessor
clock signal frequencies are within the range of aircraft radio frequency bands, but worse yet
is the fact that the harmonic multiples of those fundamental frequencies span an even larger
range, due to the fact that clock signal voltages are squarewave in shape and not sinewave.
188 CHAPTER 7. MIXEDFREQUENCY AC SIGNALS
Electromagnetic “emissions” of this nature can be a problem in industrial applications, too,
with harmonics abounding in very large quantities due to (nonlinear) electronic control of mo
tor and electric furnace power. The fundamental power line frequency may only be 60 Hz, but
those harmonic frequency multiples theoretically extend into infinitely high frequency ranges.
Low frequency power line voltage and current doesn’t radiate into space very well as electro
magnetic energy, but high frequencies do.
Also, capacitive and inductive “coupling” caused by closeproximity conductors is usually
more severe at high frequencies. Signal wiring nearby power wiring will tend to “pick up”
harmonic interference from the power wiring to a far greater extent than pure sinewave in
terference. This problem can manifest itself in industry when old motor controls are replaced
with new, solidstate electronic motor controls providing greater energy efficiency. Suddenly
there may be weird electrical noise being impressed upon signal wiring that never used to be
there, because the old controls never generated harmonics, and those highfrequency harmonic
voltages and currents tend to inductively and capacitively “couple” better to nearby conductors
than any 60 Hz signals from the old controls used to.
• REVIEW:
• Any regular (repeating), nonsinusoidal waveform is equivalent to a particular series
of sine/cosine waves of different frequencies, phases, and amplitudes, plus a DC offset
voltage if necessary. The mathematical process for determining the sinusoidal waveform
equivalent for any waveform is called Fourier analysis.
• Multiplefrequency voltage sources can be simulated for analysis by connecting several
singlefrequency voltage sources in series. Analysis of voltages and currents is accom
plished by using the superposition theorem. NOTE: superimposed voltages and currents
of different frequencies cannot be added together in complex number form, since complex
numbers only account for amplitude and phase shift, not frequency!
• Harmonics can cause problems by impressing unwanted (“noise”) voltage signals upon
nearby circuits. These unwanted signals may come by way of capacitive coupling, induc
tive coupling, electromagnetic radiation, or a combination thereof.
7.6 Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
Jason Starck (June 2000): HTML document formatting, which led to a much better
looking second edition.
Chapter 8
FILTERS
Contents
8.1 What is a filter? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
8.2 Lowpass filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
8.3 Highpass filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
8.4 Bandpass filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
8.5 Bandstop filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
8.6 Resonant filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
8.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
8.8 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
8.1 What is a filter?
It is sometimes desirable to have circuits capable of selectively filtering one frequency or range
of frequencies out of a mix of different frequencies in a circuit. A circuit designed to perform
this frequency selection is called a filter circuit, or simply a filter. A common need for filter
circuits is in highperformance stereo systems, where certain ranges of audio frequencies need
to be amplified or suppressed for best sound quality and power efficiency. You may be familiar
with equalizers, which allow the amplitudes of several frequency ranges to be adjusted to suit
the listener’s taste and acoustic properties of the listening area. You may also be familiar
with crossover networks, which block certain ranges of frequencies from reaching speakers. A
tweeter (highfrequency speaker) is inefficient at reproducing lowfrequency signals such as
drum beats, so a crossover circuit is connected between the tweeter and the stereo’s output
terminals to block lowfrequency signals, only passing highfrequency signals to the speaker’s
connection terminals. This gives better audio system efficiency and thus better performance.
Both equalizers and crossover networks are examples of filters, designed to accomplish filtering
of certain frequencies.
189
190 CHAPTER 8. FILTERS
Another practical application of filter circuits is in the “conditioning” of nonsinusoidal volt
age waveforms in power circuits. Some electronic devices are sensitive to the presence of har
monics in the power supply voltage, and so require power conditioning for proper operation. If
a distorted sinewave voltage behaves like a series of harmonic waveforms added to the fun
damental frequency, then it should be possible to construct a filter circuit that only allows the
fundamental waveform frequency to pass through, blocking all (higherfrequency) harmonics.
We will be studying the design of several elementary filter circuits in this lesson. To re
duce the load of math on the reader, I will make extensive use of SPICE as an analysis tool,
displaying Bode plots (amplitude versus frequency) for the various kinds of filters. Bear in
mind, though, that these circuits can be analyzed over several points of frequency by repeated
seriesparallel analysis, much like the previous example with two sources (60 and 90 Hz), if
the student is willing to invest a lot of time working and reworking circuit calculations for
each frequency.
• REVIEW:
• A filter is an AC circuit that separates some frequencies from others within mixedfrequency
signals.
• Audio equalizers and crossover networks are two wellknown applications of filter circuits.
• A Bode plot is a graph plotting waveform amplitude or phase on one axis and frequency
on the other.
8.2 Lowpass filters
By definition, a lowpass filter is a circuit offering easy passage to lowfrequency signals and
difficult passage to highfrequency signals. There are two basic kinds of circuits capable of
accomplishing this objective, and many variations of each one: The inductive lowpass filter in
Figure 8.1 and the capacitive lowpass filter in Figure 8.3
1
0 0
2
L1
3 H
Rload 1 kΩV1 1 V
Figure 8.1: Inductive lowpass filter
The inductor’s impedance increases with increasing frequency. This high impedance in
series tends to block highfrequency signals from getting to the load. This can be demonstrated
with a SPICE analysis: (Figure 8.2)
8.2. LOWPASS FILTERS 191
inductive lowpass filter
v1 1 0 ac 1 sin
l1 1 2 3
rload 2 0 1k
.ac lin 20 1 200
.plot ac v(2)
.end
Figure 8.2: The response of an inductive lowpass filter falls off with increasing frequency.
1
0 0
2
R1
500 Ω
C1
7 µF
Rload 1 kΩV1 1 V
Figure 8.3: Capacitive lowpass filter.
192 CHAPTER 8. FILTERS
The capacitor’s impedance decreases with increasing frequency. This low impedance in
parallel with the load resistance tends to short out highfrequency signals, dropping most of
the voltage across series resistor R1. (Figure 8.4)
capacitive lowpass filter
v1 1 0 ac 1 sin
r1 1 2 500
c1 2 0 7u
rload 2 0 1k
.ac lin 20 30 150
.plot ac v(2)
.end
Figure 8.4: The response of a capacitive lowpass filter falls off with increasing frequency.
The inductive lowpass filter is the pinnacle of simplicity, with only one component com
prising the filter. The capacitive version of this filter is not that much more complex, with
only a resistor and capacitor needed for operation. However, despite their increased complex
ity, capacitive filter designs are generally preferred over inductive because capacitors tend to
be “purer” reactive components than inductors and therefore are more predictable in their be
havior. By “pure” I mean that capacitors exhibit little resistive effects than inductors, making
them almost 100% reactive. Inductors, on the other hand, typically exhibit significant dissi
pative (resistorlike) effects, both in the long lengths of wire used to make them, and in the
magnetic losses of the core material. Capacitors also tend to participate less in “coupling” ef
fects with other components (generate and/or receive interference from other components via
mutual electric or magnetic fields) than inductors, and are less expensive.
However, the inductive lowpass filter is often preferred in ACDC power supplies to filter
out the AC “ripple” waveform created when AC is converted (rectified) into DC, passing only
8.2. LOWPASS FILTERS 193
the pure DC component. The primary reason for this is the requirement of low filter resistance
for the output of such a power supply. A capacitive lowpass filter requires an extra resistance
in series with the source, whereas the inductive lowpass filter does not. In the design of a
highcurrent circuit like a DC power supply where additional series resistance is undesirable,
the inductive lowpass filter is the better design choice. On the other hand, if low weight and
compact size are higher priorities than low internal supply resistance in a power supply design,
the capacitive lowpass filter might make more sense.
All lowpass filters are rated at a certain cutoff frequency. That is, the frequency above
which the output voltage falls below 70.7% of the input voltage. This cutoff percentage of 70.7
is not really arbitrary, all though it may seem so at first glance. In a simple capacitive/resistive
lowpass filter, it is the frequency at which capacitive reactance in ohms equals resistance in
ohms. In a simple capacitive lowpass filter (one resistor, one capacitor), the cutoff frequency
is given as:
fcutoff = 2piRC
1
Inserting the values of R and C from the last SPICE simulation into this formula, we arrive
at a cutoff frequency of 45.473 Hz. However, when we look at the plot generated by the SPICE
simulation, we see the load voltage well below 70.7% of the source voltage (1 volt) even at a
frequency as low as 30 Hz, below the calculated cutoff point. What’s wrong? The problem
here is that the load resistance of 1 kΩ affects the frequency response of the filter, skewing it
down from what the formula told us it would be. Without that load resistance in place, SPICE
produces a Bode plot whose numbers make more sense: (Figure 8.5)
capacitive lowpass filter
v1 1 0 ac 1 sin
r1 1 2 500
c1 2 0 7u
* note: no load resistor!
.ac lin 20 40 50
.plot ac v(2)
.end
fcutoff = 1/(2piRC) = 1/(2pi(500 Ω)(7 µF)) = 45.473 Hz
When dealing with filter circuits, it is always important to note that the response of the filter
depends on the filter’s component values and the impedance of the load. If a cutoff frequency
equation fails to give consideration to load impedance, it assumes no load and will fail to give
accurate results for a reallife filter conducting power to a load.
One frequent application of the capacitive lowpass filter principle is in the design of circuits
having components or sections sensitive to electrical “noise.” As mentioned at the beginning of
the last chapter, sometimes AC signals can “couple” from one circuit to another via capacitance
(Cstray) and/or mutual inductance (Mstray) between the two sets of conductors. A prime exam
ple of this is unwanted AC signals (“noise”) becoming impressed on DC power lines supplying
sensitive circuits: (Figure 8.6)
194 CHAPTER 8. FILTERS
Figure 8.5: For the capacitive lowpass filter with R = 500 Ω and C = 7 µF, the Output should
be 70.7% at 45.473 Hz.
Load
"Noise"
source
Cstray
Mstray
"Clean" DC power
"Dirty" or "noisy" DC power
Zwire Zwire Zwire
Esupply Eload
Figure 8.6: Noise is coupled by stray capacitance and mutual inductance into “clean” DC power.
8.2. LOWPASS FILTERS 195
The oscilloscopemeter on the left shows the “clean” power from the DC voltage source.
After coupling with the AC noise source via stray mutual inductance and stray capacitance,
though, the voltage as measured at the load terminals is now a mix of AC and DC, the AC
being unwanted. Normally, one would expect Eload to be precisely identical to Esource, because
the uninterrupted conductors connecting them should make the two sets of points electrically
common. However, power conductor impedance allows the two voltages to differ, which means
the noise magnitude can vary at different points in the DC system.
If we wish to prevent such “noise” from reaching the DC load, all we need to do is connect a
lowpass filter near the load to block any coupled signals. In its simplest form, this is nothing
more than a capacitor connected directly across the power terminals of the load, the capacitor
behaving as a very low impedance to any AC noise, and shorting it out. Such a capacitor is
called a decoupling capacitor: (Figure 8.7)
Load
"Noise"
source
Cstray
Mstray
"Clean" DC power
Zwire Zwire Zwire
Esupply
Eload
"Cleaner" DC power with
decoupling capacitor
Figure 8.7: Decoupling capacitor, applied to load, filters noise from DC power supply.
A cursory glance at a crowded printedcircuit board (PCB) will typically reveal decoupling
capacitors scattered throughout, usually located as close as possible to the sensitive DC loads.
Capacitor size is usually 0.1 µF or more, a minimum amount of capacitance needed to produce
a low enough impedance to short out any noise. Greater capacitance will do a better job at
filtering noise, but size and economics limit decoupling capacitors to meager values.
• REVIEW:
• A lowpass filter allows for easy passage of lowfrequency signals from source to load, and
difficult passage of highfrequency signals.
• Inductive lowpass filters insert an inductor in series with the load; capacitive lowpass
filters insert a resistor in series and a capacitor in parallel with the load. The former
196 CHAPTER 8. FILTERS
filter design tries to “block” the unwanted frequency signal while the latter tries to short
it out.
• The cutoff frequency for a lowpass filter is that frequency at which the output (load)
voltage equals 70.7% of the input (source) voltage. Above the cutoff frequency, the output
voltage is lower than 70.7% of the input, and vice versa.
8.3 Highpass filters
A highpass filter’s task is just the opposite of a lowpass filter: to offer easy passage of a
highfrequency signal and difficult passage to a lowfrequency signal. As one might expect, the
inductive (Figure 8.10) and capacitive (Figure 8.8) versions of the highpass filter are just the
opposite of their respective lowpass filter designs:
1
0 0
2
V1 1 V
C1
0.5 µF
Rload 1 kΩ
Figure 8.8: Capacitive highpass filter.
The capacitor’s impedance (Figure 8.8) increases with decreasing frequency. (Figure 8.9)
This high impedance in series tends to block lowfrequency signals from getting to load.
capacitive highpass filter
v1 1 0 ac 1 sin
c1 1 2 0.5u
rload 2 0 1k
.ac lin 20 1 200
.plot ac v(2)
.end
The inductor’s impedance (Figure 8.10) decreases with decreasing frequency. (Figure 8.11)
This low impedance in parallel tends to short out lowfrequency signals from getting to the
load resistor. As a consequence, most of the voltage gets dropped across series resistor R1.
This time, the capacitive design is the simplest, requiring only one component above and
beyond the load. And, again, the reactive purity of capacitors over inductors tends to favor
their use in filter design, especially with highpass filters where high frequencies commonly
cause inductors to behave strangely due to the skin effect and electromagnetic core losses.
As with lowpass filters, highpass filters have a rated cutoff frequency, above which the
output voltage increases above 70.7% of the input voltage. Just as in the case of the capacitive
8.3. HIGHPASS FILTERS 197
Figure 8.9: The response of the capacitive highpass filter increases with frequency.
1
0 0
2
R1
L1
200 Ω
100 mH
Rload
1 kΩV1 1 V
Figure 8.10: Inductive highpass filter.
inductive highpass filter
v1 1 0 ac 1 sin
r1 1 2 200
l1 2 0 100m
rload 2 0 1k
.ac lin 20 1 200
.plot ac v(2)
.end
198 CHAPTER 8. FILTERS
Figure 8.11: The response of the inductive highpass filter increases with frequency.
lowpass filter circuit, the capacitive highpass filter’s cutoff frequency can be found with the
same formula:
fcutoff = 2piRC
1
In the example circuit, there is no resistance other than the load resistor, so that is the
value for R in the formula.
Using a stereo system as a practical example, a capacitor connected in series with the
tweeter (treble) speaker will serve as a highpass filter, imposing a high impedance to low
frequency bass signals, thereby preventing that power from being wasted on a speaker inef
ficient for reproducing such sounds. In like fashion, an inductor connected in series with the
woofer (bass) speaker will serve as a lowpass filter for the low frequencies that particular
speaker is designed to reproduce. In this simple example circuit, the midrange speaker is
subjected to the full spectrum of frequencies from the stereo’s output. More elaborate filter
networks are sometimes used, but this should give you the general idea. Also bear in mind
that I’m only showing you one channel (either left or right) on this stereo system. A real stereo
would have six speakers: 2 woofers, 2 midranges, and 2 tweeters.
For better performance yet, we might like to have some kind of filter circuit capable of
passing frequencies that are between low (bass) and high (treble) to the midrange speaker
so that none of the low or highfrequency signal power is wasted on a speaker incapable of
efficiently reproducing those sounds. What we would be looking for is called a bandpass filter,
which is the topic of the next section.
• REVIEW:
• A highpass filter allows for easy passage of highfrequency signals from source to load,
8.4. BANDPASS FILTERS 199
Stereo Midrange
Woofer
Tweeter
lowpass
highpass
Figure 8.12: Highpass filter routes high frequencies to tweeter, while lowpass filter routes
lows to woofer.
and difficult passage of lowfrequency signals.
• Capacitive highpass filters insert a capacitor in series with the load; inductive highpass
filters insert a resistor in series and an inductor in parallel with the load. The former
filter design tries to “block” the unwanted frequency signal while the latter tries to short
it out.
• The cutoff frequency for a highpass filter is that frequency at which the output (load)
voltage equals 70.7% of the input (source) voltage. Above the cutoff frequency, the output
voltage is greater than 70.7% of the input, and vice versa.
8.4 Bandpass filters
There are applications where a particular band, or spread, or frequencies need to be filtered
from a wider range of mixed signals. Filter circuits can be designed to accomplish this task
by combining the properties of lowpass and highpass into a single filter. The result is called
a bandpass filter. Creating a bandpass filter from a lowpass and highpass filter can be
illustrated using block diagrams: (Figure 8.14)
Lowpass filter Highpass filterSignalinput
Signal
output
blocks frequencies
that are too high
blocks frequencies
that are too low
Figure 8.13: System level block diagram of a bandpass filter.
200 CHAPTER 8. FILTERS
What emerges from the series combination of these two filter circuits is a circuit that will
only allow passage of those frequencies that are neither too high nor too low. Using real com
ponents, here is what a typical schematic might look like Figure 8.14. The response of the
bandpass filter is shown in (Figure 8.15)
Source Lowpass
filter section
Highpass
filter section
1
0
2 3
0 0
V1 1 V
R1
200 Ω
C1 2.5 µF
C2
1 µF
Rload 1 kΩ
Figure 8.14: Capacitive bandpass filter.
capacitive bandpass filter
v1 1 0 ac 1 sin
r1 1 2 200
c1 2 0 2.5u
c2 2 3 1u
rload 3 0 1k
.ac lin 20 100 500
.plot ac v(3)
.end
Bandpass filters can also be constructed using inductors, but as mentioned before, the
reactive “purity” of capacitors gives them a design advantage. If we were to design a bandpass
filter using inductors, it might look something like Figure 8.16.
The fact that the highpass section comes “first” in this design instead of the lowpass sec
tion makes no difference in its overall operation. It will still filter out all frequencies too high
or too low.
While the general idea of combining lowpass and highpass filters together to make a band
pass filter is sound, it is not without certain limitations. Because this type of bandpass filter
works by relying on either section to block unwanted frequencies, it can be difficult to design
such a filter to allow unhindered passage within the desired frequency range. Both the low
pass and highpass sections will always be blocking signals to some extent, and their combined
effort makes for an attenuated (reduced amplitude) signal at best, even at the peak of the
“passband” frequency range. Notice the curve peak on the previous SPICE analysis: the load
voltage of this filter never rises above 0.59 volts, although the source voltage is a full volt.
8.4. BANDPASS FILTERS 201
Figure 8.15: The response of a capacitive bandpass filter peaks within a narrow frequency
range.
Source Lowpass
filter section
Highpass
filter section
R1 L2
L1 Rload
Figure 8.16: Inductive bandpass filter.
202 CHAPTER 8. FILTERS
This signal attenuation becomes more pronounced if the filter is designed to be more selective
(steeper curve, narrower band of passable frequencies).
There are other methods to achieve bandpass operation without sacrificing signal strength
within the passband. We will discuss those methods a little later in this chapter.
• REVIEW:
• A bandpass filter works to screen out frequencies that are too low or too high, giving
easy passage only to frequencies within a certain range.
• Bandpass filters can be made by stacking a lowpass filter on the end of a highpass filter,
or vice versa.
• “Attenuate” means to reduce or diminish in amplitude. When you turn down the volume
control on your stereo, you are “attenuating” the signal being sent to the speakers.
8.5 Bandstop filters
Also called bandelimination, bandreject, or notch filters, this kind of filter passes all frequen
cies above and below a particular range set by the component values. Not surprisingly, it can
be made out of a lowpass and a highpass filter, just like the bandpass design, except that
this time we connect the two filter sections in parallel with each other instead of in series.
(Figure 8.17)
Lowpass filter
Highpass filter
Signal
input
Signal
output
passes low frequencies
passes high frequencies
Figure 8.17: System level block diagram of a bandstop filter.
Constructed using two capacitive filter sections, it looks something like (Figure 8.18).
The lowpass filter section is comprised of R1, R2, and C1 in a “T” configuration. The high
pass filter section is comprised of C2, C3, and R3 in a “T” configuration as well. Together,
this arrangement is commonly known as a “TwinT” filter, giving sharp response when the
component values are chosen in the following ratios:
R1 = R2 = 2(R3)
Component value ratios for
the "TwinT" bandstop filter
C2 = C3 = (0.5)C1
8.5. BANDSTOP FILTERS 203
source
R1 R2
C1C2 C3
R3 Rload
Figure 8.18: “TwinT” bandstop filter.
Given these component ratios, the frequency of maximum rejection (the “notch frequency”)
can be calculated as follows:
fnotch = 4piR3C3
1
The impressive bandstopping ability of this filter is illustrated by the following SPICE
analysis: (Figure 8.19)
twint bandstop filter
v1 1 0 ac 1 sin
r1 1 2 200
c1 2 0 2u
r2 2 3 200
c2 1 4 1u
r3 4 0 100
c3 4 3 1u
rload 3 0 1k
.ac lin 20 200 1.5k
.plot ac v(3)
.end
• REVIEW:
• A bandstop filter works to screen out frequencies that are within a certain range, giving
easy passage only to frequencies outside of that range. Also known as bandelimination,
bandreject, or notch filters.
• Bandstop filters can be made by placing a lowpass filter in parallel with a highpass
filter. Commonly, both the lowpass and highpass filter sections are of the “T” configura
tion, giving the name “TwinT” to the bandstop combination.
• The frequency of maximum attenuation is called the notch frequency.
204 CHAPTER 8. FILTERS
Figure 8.19: Response of “twinT” bandstop filter.
8.6 Resonant filters
So far, the filter designs we’ve concentrated on have employed either capacitors or inductors,
but never both at the same time. We should know by now that combinations of L and C will
tend to resonate, and this property can be exploited in designing bandpass and bandstop filter
circuits.
Series LC circuits give minimum impedance at resonance, while parallel LC (“tank”) cir
cuits give maximum impedance at their resonant frequency. Knowing this, we have two basic
strategies for designing either bandpass or bandstop filters.
For bandpass filters, the two basic resonant strategies are this: series LC to pass a signal
(Figure 8.20), or parallel LC (Figure 8.22) to short a signal. The two schemes will be contrasted
and simulated here:
filter
1 2 3
0 0
L1
1 H
C1
1 µF
Rload 1 kΩV1 1 V
Figure 8.20: Series resonant LC bandpass filter.
8.6. RESONANT FILTERS 205
Series LC components pass signal at resonance, and block signals of any other frequencies
from getting to the load. (Figure 8.21)
series resonant bandpass filter
v1 1 0 ac 1 sin
l1 1 2 1
c1 2 3 1u
rload 3 0 1k
.ac lin 20 50 250
.plot ac v(3)
.end
Figure 8.21: Series resonant bandpass filter: voltage peaks at resonant frequency of 159.15
Hz.
A couple of points to note: see how there is virtually no signal attenuation within the “pass
band” (the range of frequencies near the load voltage peak), unlike the bandpass filters made
from capacitors or inductors alone. Also, since this filter works on the principle of series LC
resonance, the resonant frequency of which is unaffected by circuit resistance, the value of the
load resistor will not skew the peak frequency. However, different values for the load resistor
will change the “steepness” of the Bode plot (the “selectivity” of the filter).
The other basic style of resonant bandpass filters employs a tank circuit (parallel LC com
bination) to short out signals too high or too low in frequency from getting to the load: (Fig
ure 8.22)
The tank circuit will have a lot of impedance at resonance, allowing the signal to get to the
load with minimal attenuation. Under or over resonant frequency, however, the tank circuit
will have a low impedance, shorting out the signal and dropping most of it across series resistor
R1. (Figure 8.23)
206 CHAPTER 8. FILTERS
filter
1
0 0
2 2 2
00
R1
500 Ω
L1
100
mH
C1
10
µF
Rload 1 kΩV11 V
Figure 8.22: Parallel resonant bandpass filter.
parallel resonant bandpass filter
v1 1 0 ac 1 sin
r1 1 2 500
l1 2 0 100m
c1 2 0 10u
rload 2 0 1k
.ac lin 20 50 250
.plot ac v(2)
.end
Figure 8.23: Parallel resonant filter: voltage peaks a resonant frequency of 159.15 Hz.
8.6. RESONANT FILTERS 207
Just like the lowpass and highpass filter designs relying on a series resistance and a
parallel “shorting” component to attenuate unwanted frequencies, this resonant circuit can
never provide full input (source) voltage to the load. That series resistance will always be
dropping some amount of voltage so long as there is a load resistance connected to the output
of the filter.
It should be noted that this form of bandpass filter circuit is very popular in analog radio
tuning circuitry, for selecting a particular radio frequency from the multitudes of frequencies
available from the antenna. In most analog radio tuner circuits, the rotating dial for station
selection moves a variable capacitor in a tank circuit.
Figure 8.24: Variable capacitor tunes radio receiver tank circuit to select one out of many
broadcast stations.
The variable capacitor and aircore inductor shown in Figure 8.24 photograph of a simple
radio comprise the main elements in the tank circuit filter used to discriminate one radio
station’s signal from another.
Just as we can use series and parallel LC resonant circuits to pass only those frequencies
within a certain range, we can also use them to block frequencies within a certain range,
creating a bandstop filter. Again, we have two major strategies to follow in doing this, to use
either series or parallel resonance. First, we’ll look at the series variety: (Figure 8.25)
When the series LC combination reaches resonance, its very low impedance shorts out the
signal, dropping it across resistor R1 and preventing its passage on to the load. (Figure 8.26)
Next, we will examine the parallel resonant bandstop filter: (Figure 8.27)
The parallel LC components present a high impedance at resonant frequency, thereby block
ing the signal from the load at that frequency. Conversely, it passes signals to the load at any
other frequencies. (Figure 8.28)
Once again, notice how the absence of a series resistor makes for minimum attenuation for
all the desired (passed) signals. The amplitude at the notch frequency, on the other hand, is
208 CHAPTER 8. FILTERS
1
0
2
3
2
0 0
V1 1 V
R1
500 Ω
L1 100 mH
C1 10 µF
Rload 1 kΩ
Figure 8.25: Series resonant bandstop filter.
series resonant bandstop filter
v1 1 0 ac 1 sin
r1 1 2 500
l1 2 3 100m
c1 3 0 10u
rload 2 0 1k
.ac lin 20 70 230
.plot ac v(2)
.end
Figure 8.26: Series resonant bandstop filter: Notch frequency = LC resonant frequency (159.15
Hz).
8.6. RESONANT FILTERS 209
1
0 0
2
V1 1 V
C1 10 µF
L1 100 mH Rload 1 kΩ
Figure 8.27: Parallel resonant bandstop filter.
parallel resonant bandstop filter
v1 1 0 ac 1 sin
l1 1 2 100m
c1 1 2 10u
rload 2 0 1k
.ac lin 20 100 200
.plot ac v(2)
.end
Figure 8.28: Parallel resonant bandstop filter: Notch frequency = LC resonant frequency
(159.15 Hz).
210 CHAPTER 8. FILTERS
very low. In other words, this is a very “selective” filter.
In all these resonant filter designs, the selectivity depends greatly upon the “purity” of
the inductance and capacitance used. If there is any stray resistance (especially likely in the
inductor), this will diminish the filter’s ability to finely discriminate frequencies, as well as
introduce antiresonant effects that will skew the peak/notch frequency.
A word of caution to those designing lowpass and highpass filters is in order at this point.
After assessing the standard RC and LR lowpass and highpass filter designs, it might occur
to a student that a better, more effective design of lowpass or highpass filter might be realized
by combining capacitive and inductive elements together like Figure 8.29.
filter
V1 1 V
L11
0 0 0
C1
L2
Rload 1 kΩ
100 mH 100 mH
1 µF
2 3
Figure 8.29: Capacitive Inductive lowpass filter.
The inductors should block any high frequencies, while the capacitor should short out any
high frequencies as well, both working together to allow only low frequency signals to reach
the load.
At first, this seems to be a good strategy, and eliminates the need for a series resistance.
However, the more insightful student will recognize that any combination of capacitors and in
ductors together in a circuit is likely to cause resonant effects to happen at a certain frequency.
Resonance, as we have seen before, can cause strange things to happen. Let’s plot a SPICE
analysis and see what happens over a wide frequency range: (Figure 8.30)
lc lowpass filter
v1 1 0 ac 1 sin
l1 1 2 100m
c1 2 0 1u
l2 2 3 100m
rload 3 0 1k
.ac lin 20 100 1k
.plot ac v(3)
.end
What was supposed to be a lowpass filter turns out to be a bandpass filter with a peak
somewhere around 526 Hz! The capacitance and inductance in this filter circuit are attaining
resonance at that point, creating a large voltage drop around C1, which is seen at the load,
regardless of L2’s attenuating influence. The output voltage to the load at this point actually
8.6. RESONANT FILTERS 211
Figure 8.30: Unexpected response of LC lowpass filter.
exceeds the input (source) voltage! A little more reflection reveals that if L1 and C2 are at
resonance, they will impose a very heavy (very low impedance) load on the AC source, which
might not be good either. We’ll run the same analysis again, only this time plotting C1’s voltage,
vm(2) in Figure 8.31, and the source current, I(v1), along with load voltage, vm(3):
Figure 8.31: Current inceases at the unwanted resonance of the LC lowpass filter.
212 CHAPTER 8. FILTERS
Sure enough, we see the voltage across C1 and the source current spiking to a high point
at the same frequency where the load voltage is maximum. If we were expecting this filter to
provide a simple lowpass function, we might be disappointed by the results.
The problem is that an LC filter has a input impedance and an output impedance which
must be matched. The voltage source impedance must match the input impedance of the filter,
and the filter output impedance must be matched by “rload” for a flat response. The input and
output impedance is given by the square root of (L/C).
Z = (L/C)1/2
Taking the component values from (Figure 8.29), we can find the impedance of the filter,
and the required , Rg and Rload to match it.
For L= 100 mH, C= 1µF
Z = (L/C)1/2=((100 mH)/(1 µF))1/2 = 316 Ω
In Figure 8.32 we have added Rg = 316 Ω to the generator, and changed the load Rload from
1000 Ω to 316 Ω. Note that if we needed to drive a 1000 Ω load, the L/C ratio could have been
adjusted to match that resistance.
filter
1 V
L1
C1
L2
316Ω
Rg
316Ω
Vpp
Voffset
1Hz 1.0uF
100mH100mH
Rload
0
32
1
4
Figure 8.32: Circuit of source and load matched LC lowpass filter.
LC matched lowpass filter
V1 1 0 ac 1 SIN
Rg 1 4 316
L1 4 2 100m
C1 2 0 1.0u
L2 2 3 100m
Rload 3 0 316
.ac lin 20 100 1k
.plot ac v(3)
.end
8.6. RESONANT FILTERS 213
Figure 8.33 shows the “flat” response of the LC low pass filter when the source and load
impedance match the filter input and output impedances.
Figure 8.33: The response of impedance matched LC lowpass filter is nearly flat up to the
cutoff frequency.
The point to make in comparing the response of the unmatched filter (Figure 8.30) to the
matched filter (Figure 8.33) is that variable load on the filter produces a considerable change
in voltage. This property is directly applicable to LC filtered power supplies– the regulation is
poor. The power supply voltage changes with a change in load. This is undesirable.
This poor load regulation can be mitigated by a swinging choke. This is a choke, inductor,
designed to saturate when a large DC current passes through it. By saturate, we mean that
the DC current creates a “too” high level of flux in the magnetic core, so that the AC compo
nent of current cannot vary the flux. Since induction is proportional to dΦ/dt, the inductance is
decreased by the heavy DC current. The decrease in inductance decreases reactance XL. De
creasing reactance, reduces the voltage drop across the inductor; thus, increasing the voltage
at the filter output. This improves the voltage regulation with respect to variable loads.
Despite the unintended resonance, lowpass filters made up of capacitors and inductors are
frequently used as final stages in AC/DC power supplies to filter the unwanted AC “ripple”
voltage out of the DC converted from AC. Why is this, if this particular filter design possesses
a potentially troublesome resonant point?
The answer lies in the selection of filter component sizes and the frequencies encountered
from an AC/DC converter (rectifier). What we’re trying to do in an AC/DC power supply filter
is separate DC voltage from a small amount of relatively highfrequency AC voltage. The
filter inductors and capacitors are generally quite large (several Henrys for the inductors and
thousands of µF for the capacitors is typical), making the filter’s resonant frequency very, very
low. DC of course, has a “frequency” of zero, so there’s no way it can make an LC circuit
resonate. The ripple voltage, on the other hand, is a nonsinusoidal AC voltage consisting
214 CHAPTER 8. FILTERS
of a fundamental frequency at least twice the frequency of the converted AC voltage, with
harmonics many times that in addition. For pluginthewall power supplies running on 60 Hz
AC power (60 Hz United States; 50 Hz in Europe), the lowest frequency the filter will ever see
is 120 Hz (100 Hz in Europe), which is well above its resonant point. Therefore, the potentially
troublesome resonant point in a such a filter is completely avoided.
The following SPICE analysis calculates the voltage output (AC and DC) for such a filter,
with series DC and AC (120 Hz) voltage sources providing a rough approximation of the mixed
frequency output of an AC/DC converter.
1
0
2 3 4
0 0
V1 1 V
120 Hz
V2 12 V
L1
3 H
L2
2 H
C1 9500µF
Rload 1 kΩ
Figure 8.34: AC/DC power suppply filter provides “ripple free” DC power.
ac/dc power supply filter
v1 1 0 ac 1 sin
v2 2 1 dc
l1 2 3 3
c1 3 0 9500u
l2 3 4 2
rload 4 0 1k
.dc v2 12 12 1
.ac lin 1 120 120
.print dc v(4)
.print ac v(4)
.end
v2 v(4)
1.200E+01 1.200E+01 DC voltage at load = 12 volts
freq v(4)
1.200E+02 3.412E05 AC voltage at load = 34.12 microvolts
With a full 12 volts DC at the load and only 34.12 µV of AC left from the 1 volt AC source
imposed across the load, this circuit design proves itself to be a very effective power supply
filter.
8.7. SUMMARY 215
The lesson learned here about resonant effects also applies to the design of highpass filters
using both capacitors and inductors. So long as the desired and undesired frequencies are well
to either side of the resonant point, the filter will work OK. But if any signal of significant
magnitude close to the resonant frequency is applied to the input of the filter, strange things
will happen!
• REVIEW:
• Resonant combinations of capacitance and inductance can be employed to create very
effective bandpass and bandstop filters without the need for added resistance in a circuit
that would diminish the passage of desired frequencies.
•
fresonant =
2pi LC
1
8.7 Summary
As lengthy as this chapter has been up to this point, it only begins to scratch the surface of
filter design. A quick perusal of any advanced filter design textbook is sufficient to prove my
point. The mathematics involved with component selection and frequency response prediction
is daunting to say the least – well beyond the scope of the beginning electronics student. It has
been my intent here to present the basic principles of filter design with as little math as possi
ble, leaning on the power of the SPICE circuit analysis program to explore filter performance.
The benefit of such computer simulation software cannot be understated, for the beginning
student or for the working engineer.
Circuit simulation software empowers the student to explore circuit designs far beyond
the reach of their math skills. With the ability to generate Bode plots and precise figures,
an intuitive understanding of circuit concepts can be attained, which is something often lost
when a student is burdened with the task of solving lengthy equations by hand. If you are not
familiar with the use of SPICE or other circuit simulation programs, take the time to become
so! It will be of great benefit to your study. To see SPICE analyses presented in this book is an
aid to understanding circuits, but to actually set up and analyze your own circuit simulations
is a much more engaging and worthwhile endeavor as a student.
8.8 Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
Jason Starck (June 2000): HTML document formatting, which led to a much better
looking second edition.
216 CHAPTER 8. FILTERS
Chapter 9
TRANSFORMERS
Contents
9.1 Mutual inductance and basic operation . . . . . . . . . . . . . . . . . . . . . 218
9.2 Stepup and stepdown transformers . . . . . . . . . . . . . . . . . . . . . . 232
9.3 Electrical isolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
9.4 Phasing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
9.5 Winding configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
9.6 Voltage regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248
9.7 Special transformers and applications . . . . . . . . . . . . . . . . . . . . . 251
9.7.1 Impedance matching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
9.7.2 Potential transformers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
9.7.3 Current transformers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
9.7.4 Air core transformers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
9.7.5 Tesla Coil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
9.7.6 Saturable reactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262
9.7.7 ScottT transformer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
9.7.8 Linear Variable Differential Transformer . . . . . . . . . . . . . . . . . . 267
9.8 Practical considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
9.8.1 Power capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
9.8.2 Energy losses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
9.8.3 Stray capacitance and inductance . . . . . . . . . . . . . . . . . . . . . . . 271
9.8.4 Core saturation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
9.8.5 Inrush current . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
9.8.6 Heat and Noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
9.9 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
217
218 CHAPTER 9. TRANSFORMERS
9.1 Mutual inductance and basic operation
Suppose we were to wrap a coil of insulated wire around a loop of ferromagnetic material and
energize this coil with an AC voltage source: (Figure 9.1 (a))
wire
coil
iron core
resistor
(a) (b)
Figure 9.1: Insulated winding on ferromagnetic loop has inductive reactance, limiting AC cur
rent.
As an inductor, we would expect this ironcore coil to oppose the applied voltage with its
inductive reactance, limiting current through the coil as predicted by the equations XL = 2pifL
and I=E/X (or I=E/Z). For the purposes of this example, though, we need to take a more detailed
look at the interactions of voltage, current, and magnetic flux in the device.
Kirchhoff ’s voltage law describes how the algebraic sum of all voltages in a loop must equal
zero. In this example, we could apply this fundamental law of electricity to describe the respec
tive voltages of the source and of the inductor coil. Here, as in any onesource, oneload circuit,
the voltage dropped across the load must equal the voltage supplied by the source, assuming
zero voltage dropped along the resistance of any connecting wires. In other words, the load
(inductor coil) must produce an opposing voltage equal in magnitude to the source, in order
that it may balance against the source voltage and produce an algebraic loop voltage sum of
zero. From where does this opposing voltage arise? If the load were a resistor (Figure 9.1
(b)), the voltage drop originates from electrical energy loss, the “friction” of electrons flowing
through the resistance. With a perfect inductor (no resistance in the coil wire), the opposing
voltage comes from another mechanism: the reaction to a changing magnetic flux in the iron
core. When AC current changes, flux Φ changes. Changing flux induces a counter EMF.
Michael Faraday discovered the mathematical relationship between magnetic flux (Φ) and
induced voltage with this equation:
dΦ
dt
Where,
N
N =
Φ =
t =
Number of turns in wire coil (straight wire = 1)
Magnetic flux in Webers
Time in seconds
e =
e = (Instantaneous) induced voltage in volts
9.1. MUTUAL INDUCTANCE AND BASIC OPERATION 219
The instantaneous voltage (voltage dropped at any instant in time) across a wire coil is
equal to the number of turns of that coil around the core (N) multiplied by the instantaneous
rateofchange in magnetic flux (dΦ/dt) linking with the coil. Graphed, (Figure 9.2) this shows
itself as a set of sine waves (assuming a sinusoidal voltage source), the flux wave 90o lagging
behind the voltage wave:
e = voltage Φ = magnetic flux
e Φ
Figure 9.2: Magnetic flux lags applied voltage by 90o because flux is proportional to a rate of
change, dΦ/dt.
Magnetic flux through a ferromagnetic material is analogous to current through a conduc
tor: it must be motivated by some force in order to occur. In electric circuits, this motivating
force is voltage (a.k.a. electromotive force, or EMF). In magnetic “circuits,” this motivating
force is magnetomotive force, or mmf. Magnetomotive force (mmf) and magnetic flux (Φ) are
related to each other by a property of magnetic materials known as reluctance (the latter quan
tity symbolized by a strangelooking letter “R”):
Electrical Magnetic
E = IR mmf = Φℜ
A comparison of "Ohm’s Law" for
electric and magnetic circuits:
In our example, the mmf required to produce this changing magnetic flux (Φ) must be sup
plied by a changing current through the coil. Magnetomotive force generated by an electro
magnet coil is equal to the amount of current through that coil (in amps) multiplied by the
number of turns of that coil around the core (the SI unit for mmf is the ampturn). Because
the mathematical relationship between magnetic flux and mmf is directly proportional, and
because the mathematical relationship between mmf and current is also directly proportional
(no ratesofchange present in either equation), the current through the coil will be inphase
with the flux wave as in (Figure 9.3)
This is why alternating current through an inductor lags the applied voltage waveform
by 90o: because that is what is required to produce a changing magnetic flux whose rateof
change produces an opposing voltage inphase with the applied voltage. Due to its function in
providing magnetizing force (mmf) for the core, this current is sometimes referred to as the
magnetizing current.
It should be mentioned that the current through an ironcore inductor is not perfectly sinu
soidal (sinewave shaped), due to the nonlinear B/H magnetization curve of iron. In fact, if the
220 CHAPTER 9. TRANSFORMERS
e = voltage Φ = magnetic flux i = coil current
e Φ
i
Figure 9.3: Magnetic flux, like current, lags applied voltage by 90o.
inductor is cheaply built, using as little iron as possible, the magnetic flux density might reach
high levels (approaching saturation), resulting in a magnetizing current waveform that looks
something like Figure 9.4
e = voltage
Φ = magnetic flux
i = coil current
e Φ
i
Figure 9.4: As flux density approaches saturation, the magnetizing current waveform becomes
distorted.
When a ferromagnetic material approaches magnetic flux saturation, disproportionately
greater levels of magnetic field force (mmf) are required to deliver equal increases in magnetic
field flux (Φ). Because mmf is proportional to current through the magnetizing coil (mmf = NI,
where “N” is the number of turns of wire in the coil and “I” is the current through it), the large
increases of mmf required to supply the needed increases in flux results in large increases
in coil current. Thus, coil current increases dramatically at the peaks in order to maintain
a flux waveform that isn’t distorted, accounting for the bellshaped halfcycles of the current
waveform in the above plot.
The situation is further complicated by energy losses within the iron core. The effects of
hysteresis and eddy currents conspire to further distort and complicate the current waveform,
making it even less sinusoidal and altering its phase to be lagging slightly less than 90o behind
the applied voltage waveform. This coil current resulting from the sum total of all magnetic
effects in the core (dΦ/dt magnetization plus hysteresis losses, eddy current losses, etc.) is
called the exciting current. The distortion of an ironcore inductor’s exciting current may be
minimized if it is designed for and operated at very low flux densities. Generally speaking, this
9.1. MUTUAL INDUCTANCE AND BASIC OPERATION 221
requires a core with large crosssectional area, which tends to make the inductor bulky and
expensive. For the sake of simplicity, though, we’ll assume that our example core is far from
saturation and free from all losses, resulting in a perfectly sinusoidal exciting current.
As we’ve seen already in the inductors chapter, having a current waveform 90o out of phase
with the voltage waveform creates a condition where power is alternately absorbed and re
turned to the circuit by the inductor. If the inductor is perfect (no wire resistance, no magnetic
core losses, etc.), it will dissipate zero power.
Let us now consider the same inductor device, except this time with a second coil (Fig
ure 9.5) wrapped around the same iron core. The first coil will be labeled the primary coil,
while the second will be labeled the secondary:
wire
coil
iron core
wire
coil
Figure 9.5: Ferromagnetic core with primary coil (AC driven) and secondary coil.
If this secondary coil experiences the same magnetic flux change as the primary (which
it should, assuming perfect containment of the magnetic flux through the common core), and
has the same number of turns around the core, a voltage of equal magnitude and phase to
the applied voltage will be induced along its length. In the following graph, (Figure 9.6) the
induced voltage waveform is drawn slightly smaller than the source voltage waveform simply
to distinguish one from the other:
This effect is called mutual inductance: the induction of a voltage in one coil in response to
a change in current in the other coil. Like normal (self) inductance, it is measured in the unit
of Henrys, but unlike normal inductance it is symbolized by the capital letter “M” rather than
the letter “L”:
222 CHAPTER 9. TRANSFORMERS
Φ = magnetic flux
Φ
ep = primary coil voltage
es = secondary coil voltage
ep
es
ip = primary coil current
ip
Figure 9.6: Open circuited secondary sees the same flux Φ as the primary. Therefore induced
secondary voltage es is the same magnitude and phase as the primary voltage ep.
e = L di
dt
Inductance Mutual inductance
e2 = M
di1
dt
Where,
e2 = voltage induced in
secondary coil
i1 = current in primary
coil
No current will exist in the secondary coil, since it is opencircuited. However, if we connect
a load resistor to it, an alternating current will go through the coil, inphase with the induced
voltage (because the voltage across a resistor and the current through it are always inphase
with each other). (Figure 9.7)
At first, one might expect this secondary coil current to cause additional magnetic flux in the
core. In fact, it does not. If more flux were induced in the core, it would cause more voltage to
be induced voltage in the primary coil (remember that e = dΦ/dt). This cannot happen, because
the primary coil’s induced voltage must remain at the same magnitude and phase in order to
balance with the applied voltage, in accordance with Kirchhoff ’s voltage law. Consequently,
the magnetic flux in the core cannot be affected by secondary coil current. However, what does
change is the amount of mmf in the magnetic circuit.
Magnetomotive force is produced any time electrons move through a wire. Usually, this
mmf is accompanied by magnetic flux, in accordance with the mmf=ΦR “magnetic Ohm’s Law”
equation. In this case, though, additional flux is not permitted, so the only way the secondary
coil’s mmf may exist is if a counteracting mmf is generated by the primary coil, of equal mag
nitude and opposite phase. Indeed, this is what happens, an alternating current forming in
the primary coil – 180o out of phase with the secondary coil’s current – to generate this coun
teracting mmf and prevent additional core flux. Polarity marks and current direction arrows
have been added to the illustration to clarify phase relations: (Figure 9.8)
If you find this process a bit confusing, do not worry. Transformer dynamics is a complex
9.1. MUTUAL INDUCTANCE AND BASIC OPERATION 223
wire
coil
iron core
wire
coil
Figure 9.7: Resistive load on secondary has voltage and current inphase.
wire
coil
iron core
wire
coil
+

+

+

+

mmfprimary
mmfsecondary
Figure 9.8: Flux remains constant with application of a load. However, a counteracting mmf is
produced by the loaded secondary.
224 CHAPTER 9. TRANSFORMERS
subject. What is important to understand is this: when an AC voltage is applied to the primary
coil, it creates a magnetic flux in the core, which induces AC voltage in the secondary coil in
phase with the source voltage. Any current drawn through the secondary coil to power a load
induces a corresponding current in the primary coil, drawing current from the source.
Notice how the primary coil is behaving as a load with respect to the AC voltage source,
and how the secondary coil is behaving as a source with respect to the resistor. Rather than
energy merely being alternately absorbed and returned the primary coil circuit, energy is now
being coupled to the secondary coil where it is delivered to a dissipative (energyconsuming)
load. As far as the source “knows,” its directly powering the resistor. Of course, there is also an
additional primary coil current lagging the applied voltage by 90o, just enough to magnetize
the core to create the necessary voltage for balancing against the source (the exciting current).
We call this type of device a transformer, because it transforms electrical energy into mag
netic energy, then back into electrical energy again. Because its operation depends on electro
magnetic induction between two stationary coils and a magnetic flux of changing magnitude
and “polarity,” transformers are necessarily AC devices. Its schematic symbol looks like two
inductors (coils) sharing the same magnetic core: (Figure 9.9)
Transformer
Figure 9.9: Schematic symbol for transformer consists of two inductor symbols, separated by
lines indicating a ferromagnetic core.
The two inductor coils are easily distinguished in the above symbol. The pair of verti
cal lines represent an iron core common to both inductors. While many transformers have
ferromagnetic core materials, there are some that do not, their constituent inductors being
magnetically linked together through the air.
The following photograph shows a power transformer of the type used in gasdischarge
lighting. Here, the two inductor coils can be clearly seen, wound around an iron core. While
most transformer designs enclose the coils and core in a metal frame for protection, this partic
ular transformer is open for viewing and so serves its illustrative purpose well: (Figure 9.10)
Both coils of wire can be seen here with coppercolored varnish insulation. The top coil is
larger than the bottom coil, having a greater number of “turns” around the core. In trans
formers, the inductor coils are often referred to as windings, in reference to the manufacturing
process where wire is wound around the core material. As modeled in our initial example, the
powered inductor of a transformer is called the primary winding, while the unpowered coil is
called the secondary winding.
In the next photograph, Figure 9.11, a transformer is shown cut in half, exposing the cross
section of the iron core as well as both windings. Like the transformer shown previously, this
unit also utilizes primary and secondary windings of differing turn counts. The wire gauge can
also be seen to differ between primary and secondary windings. The reason for this disparity
in wire gauge will be made clear in the next section of this chapter. Additionally, the iron core
can be seen in this photograph to be made of many thin sheets (laminations) rather than a
9.1. MUTUAL INDUCTANCE AND BASIC OPERATION 225
Figure 9.10: Example of a gasdischarge lighting transformer.
solid piece. The reason for this will also be explained in a later section of this chapter.
Figure 9.11: Transformer crosssection cut shows core and windings.
It is easy to demonstrate simple transformer action using SPICE, setting up the primary
and secondary windings of the simulated transformer as a pair of “mutual” inductors. (Fig
226 CHAPTER 9. TRANSFORMERS
ure 9.12) The coefficient of magnetic field coupling is given at the end of the “k” line in the
SPICE circuit description, this example being set very nearly at perfection (1.000). This co
efficient describes how closely “linked” the two inductors are, magnetically. The better these
two inductors are magnetically coupled, the more efficient the energy transfer between them
should be.
(for SPICE to measure current)
1
0
2
0
3 4
5 5
L1 L2
Vi1
0 V
Rload 1 kΩV1 10 V
100 H 100 H
Rbogus1
Rbogus2
(very small)
(very large)
Figure 9.12: Spice circuit for coupled inductors.
transformer
v1 1 0 ac 10 sin
rbogus1 1 2 1e12
rbogus2 5 0 9e12
l1 2 0 100
l2 3 5 100
** This line tells SPICE that the two inductors
** l1 and l2 are magnetically ‘‘linked’’ together
k l1 l2 0.999
vi1 3 4 ac 0
rload 4 5 1k
.ac lin 1 60 60
.print ac v(2,0) i(v1)
.print ac v(3,5) i(vi1)
.end
Note: the Rbogus resistors are required to satisfy certain quirks of SPICE. The first breaks
the otherwise continuous loop between the voltage source and L1 which would not be permitted
by SPICE. The second provides a path to ground (node 0) from the secondary circuit, necessary
because SPICE cannot function with any ungrounded circuits.
Note that with equal inductances for both windings (100 Henrys each), the AC voltages and
currents are nearly equal for the two. The difference between primary and secondary currents
is the magnetizing current spoken of earlier: the 90o lagging current necessary to magnetize
the core. As is seen here, it is usually very small compared to primary current induced by the
load, and so the primary and secondary currents are almost equal. What you are seeing here
9.1. MUTUAL INDUCTANCE AND BASIC OPERATION 227
freq v(2) i(v1)
6.000E+01 1.000E+01 9.975E03 Primary winding
freq v(3,5) i(vi1)
6.000E+01 9.962E+00 9.962E03 Secondary winding
is quite typical of transformer efficiency. Anything less than 95% efficiency is considered poor
for modern power transformer designs, and this transfer of power occurs with no moving parts
or other components subject to wear.
If we decrease the load resistance so as to draw more current with the same amount of volt
age, we see that the current through the primary winding increases in response. Even though
the AC power source is not directly connected to the load resistance (rather, it is electromag
netically “coupled”), the amount of current drawn from the source will be almost the same as
the amount of current that would be drawn if the load were directly connected to the source.
Take a close look at the next two SPICE simulations, showing what happens with different
values of load resistors:
transformer
v1 1 0 ac 10 sin
rbogus1 1 2 1e12
rbogus2 5 0 9e12
l1 2 0 100
l2 3 5 100
k l1 l2 0.999
vi1 3 4 ac 0
** Note load resistance value of 200 ohms
rload 4 5 200
.ac lin 1 60 60
.print ac v(2,0) i(v1)
.print ac v(3,5) i(vi1)
.end
freq v(2) i(v1)
6.000E+01 1.000E+01 4.679E02
freq v(3,5) i(vi1)
6.000E+01 9.348E+00 4.674E02
Notice how the primary current closely follows the secondary current. In our first simula
tion, both currents were approximately 10 mA, but now they are both around 47 mA. In this
second simulation, the two currents are closer to equality, because the magnetizing current
remains the same as before while the load current has increased. Note also how the secondary
voltage has decreased some with the heavier (greater current) load. Let’s try another simula
tion with an even lower value of load resistance (15 Ω):
Our load current is now 0.13 amps, or 130 mA, which is substantially higher than the
last time. The primary current is very close to being the same, but notice how the secondary
228 CHAPTER 9. TRANSFORMERS
transformer
v1 1 0 ac 10 sin
rbogus1 1 2 1e12
rbogus2 5 0 9e12
l1 2 0 100
l2 3 5 100
k l1 l2 0.999
vi1 3 4 ac 0
rload 4 5 15
.ac lin 1 60 60
.print ac v(2,0) i(v1)
.print ac v(3,5) i(vi1)
.end
freq v(2) i(v1)
6.000E+01 1.000E+01 1.301E01
freq v(3,5) i(vi1)
6.000E+01 1.950E+00 1.300E01
voltage has fallen well below the primary voltage (1.95 volts versus 10 volts at the primary).
The reason for this is an imperfection in our transformer design: because the primary and
secondary inductances aren’t perfectly linked (a k factor of 0.999 instead of 1.000) there is
“stray” or “leakage” inductance. In other words, some of the magnetic field isn’t linking with
the secondary coil, and thus cannot couple energy to it: (Figure 9.13)
wire
coil
wire
coil
leakage
flux
leakage
flux
core flux
Figure 9.13: Leakage inductance is due to magnetic flux not cutting both windings.
Consequently, this “leakage” flux merely stores and returns energy to the source circuit
via selfinductance, effectively acting as a series impedance in both primary and secondary
circuits. Voltage gets dropped across this series impedance, resulting in a reduced load voltage:
9.1. MUTUAL INDUCTANCE AND BASIC OPERATION 229
voltage across the load “sags” as load current increases. (Figure 9.14)
ideal
transformer
leakage
inductance
leakage
inductance
Source Load
Figure 9.14: Equivalent circuit models leakage inductance as series inductors independent of
the “ideal transformer”.
If we change the transformer design to have better magnetic coupling between the primary
and secondary coils, the figures for voltage between primary and secondary windings will be
much closer to equality again:
transformer
v1 1 0 ac 10 sin
rbogus1 1 2 1e12
rbogus2 5 0 9e12
l1 2 0 100
l2 3 5 100
** Coupling factor = 0.99999 instead of 0.999
k l1 l2 0.99999
vi1 3 4 ac 0
rload 4 5 15
.ac lin 1 60 60
.print ac v(2,0) i(v1)
.print ac v(3,5) i(vi1)
.end
freq v(2) i(v1)
6.000E+01 1.000E+01 6.658E01
freq v(3,5) i(vi1)
6.000E+01 9.987E+00 6.658E01
Here we see that our secondary voltage is back to being equal with the primary, and the
secondary current is equal to the primary current as well. Unfortunately, building a real
transformer with coupling this complete is very difficult. A compromise solution is to design
both primary and secondary coils with less inductance, the strategy being that less inductance
overall leads to less “leakage” inductance to cause trouble, for any given degree of magnetic
coupling inefficiency. This results in a load voltage that is closer to ideal with the same (high
current heavy) load and the same coupling factor:
Simply by using primary and secondary coils of less inductance, the load voltage for this
heavy load (high current) has been brought back up to nearly ideal levels (9.977 volts). At this
230 CHAPTER 9. TRANSFORMERS
transformer
v1 1 0 ac 10 sin
rbogus1 1 2 1e12
rbogus2 5 0 9e12
** inductance = 1 henry instead of 100 henrys
l1 2 0 1
l2 3 5 1
k l1 l2 0.999
vi1 3 4 ac 0
rload 4 5 15
.ac lin 1 60 60
.print ac v(2,0) i(v1)
.print ac v(3,5) i(vi1)
.end
freq v(2) i(v1)
6.000E+01 1.000E+01 6.664E01
freq v(3,5) i(vi1)
6.000E+01 9.977E+00 6.652E01
point, one might ask, “If less inductance is all that’s needed to achieve nearideal performance
under heavy load, then why worry about coupling efficiency at all? If its impossible to build a
transformer with perfect coupling, but easy to design coils with low inductance, then why not
just build all transformers with lowinductance coils and have excellent efficiency even with
poor magnetic coupling?”
The answer to this question is found in another simulation: the same lowinductance trans
former, but this time with a lighter load (less current) of 1 kΩ instead of 15 Ω:
transformer
v1 1 0 ac 10 sin
rbogus1 1 2 1e12
rbogus2 5 0 9e12
l1 2 0 1
l2 3 5 1
k l1 l2 0.999
vi1 3 4 ac 0
rload 4 5 1k
.ac lin 1 60 60
.print ac v(2,0) i(v1)
.print ac v(3,5) i(vi1)
.end
With lower winding inductances, the primary and secondary voltages are closer to being
equal, but the primary and secondary currents are not. In this particular case, the primary
current is 28.35 mA while the secondary current is only 9.990 mA: almost three times as much
current in the primary as the secondary. Why is this? With less inductance in the primary
winding, there is less inductive reactance, and consequently a much larger magnetizing cur
9.1. MUTUAL INDUCTANCE AND BASIC OPERATION 231
freq v(2) i(v1)
6.000E+01 1.000E+01 2.835E02
freq v(3,5) i(vi1)
6.000E+01 9.990E+00 9.990E03
rent. A substantial amount of the current through the primary winding merely works to mag
netize the core rather than transfer useful energy to the secondary winding and load.
An ideal transformer with identical primary and secondary windings would manifest equal
voltage and current in both sets of windings for any load condition. In a perfect world, trans
formers would transfer electrical power from primary to secondary as smoothly as though the
load were directly connected to the primary power source, with no transformer there at all.
However, you can see this ideal goal can only be met if there is perfect coupling of magnetic
flux between primary and secondary windings. Being that this is impossible to achieve, trans
formers must be designed to operate within certain expected ranges of voltages and loads in
order to perform as close to ideal as possible. For now, the most important thing to keep in
mind is a transformer’s basic operating principle: the transfer of power from the primary to
the secondary circuit via electromagnetic coupling.
• REVIEW:
• Mutual inductance is where the magnetic flux of two or more inductors are “linked” so
that voltage is induced in one coil proportional to the rateofchange of current in another.
• A transformer is a device made of two or more inductors, one of which is powered by AC,
inducing an AC voltage across the second inductor. If the second inductor is connected to
a load, power will be electromagnetically coupled from the first inductor’s power source
to that load.
• The powered inductor in a transformer is called the primary winding. The unpowered
inductor in a transformer is called the secondary winding.
• Magnetic flux in the core (Φ) lags 90o behind the source voltage waveform. The current
drawn by the primary coil from the source to produce this flux is called the magnetizing
current, and it also lags the supply voltage by 90o.
• Total primary current in an unloaded transformer is called the exciting current, and is
comprised of magnetizing current plus any additional current necessary to overcome core
losses. It is never perfectly sinusoidal in a real transformer, but may be made more so
if the transformer is designed and operated so that magnetic flux density is kept to a
minimum.
• Core flux induces a voltage in any coil wrapped around the core. The induces voltage(s)
are ideally in phase with the primary winding source voltage and share the same wave
shape.
• Any current drawn through the secondary winding by a load will be “reflected” to the pri
mary winding and drawn from the voltage source, as if the source were directly powering
a similar load.
232 CHAPTER 9. TRANSFORMERS
9.2 Stepup and stepdown transformers
So far, we’ve observed simulations of transformers where the primary and secondary windings
were of identical inductance, giving approximately equal voltage and current levels in both
circuits. Equality of voltage and current between the primary and secondary sides of a trans
former, however, is not the norm for all transformers. If the inductances of the two windings
are not equal, something interesting happens:
transformer
v1 1 0 ac 10 sin
rbogus1 1 2 1e12
rbogus2 5 0 9e12
l1 2 0 10000
l2 3 5 100
k l1 l2 0.999
vi1 3 4 ac 0
rload 4 5 1k
.ac lin 1 60 60
.print ac v(2,0) i(v1)
.print ac v(3,5) i(vi1)
.end
freq v(2) i(v1)
6.000E+01 1.000E+01 9.975E05 Primary winding
freq v(3,5) i(vi1)
6.000E+01 9.962E01 9.962E04 Secondary winding
Notice how the secondary voltage is approximately ten times less than the primary voltage
(0.9962 volts compared to 10 volts), while the secondary current is approximately ten times
greater (0.9962 mA compared to 0.09975 mA). What we have here is a device that steps voltage
down by a factor of ten and current up by a factor of ten: (Figure 9.15)
Primary
winding
Secondary
winding
Figure 9.15: Turns ratio of 10:1 yields 10:1 primary:secondary voltage ratio and 1:10 pri
mary:secondary current ratio.
This is a very useful device, indeed. With it, we can easily multiply or divide voltage and
current in AC circuits. Indeed, the transformer has made longdistance transmission of elec
tric power a practical reality, as AC voltage can be “stepped up” and current “stepped down”
for reduced wire resistance power losses along power lines connecting generating stations with
9.2. STEPUP AND STEPDOWN TRANSFORMERS 233
loads. At either end (both the generator and at the loads), voltage levels are reduced by trans
formers for safer operation and less expensive equipment. A transformer that increases volt
age from primary to secondary (more secondary winding turns than primary winding turns)
is called a stepup transformer. Conversely, a transformer designed to do just the opposite is
called a stepdown transformer.
Let’s reexamine a photograph shown in the previous section: (Figure 9.16)
Figure 9.16: Transformer crosssection showing primary and secondary windings is a few
inches tall (approximately 10 cm).
This is a stepdown transformer, as evidenced by the high turn count of the primary winding
and the low turn count of the secondary. As a stepdown unit, this transformer converts high
voltage, lowcurrent power into lowvoltage, highcurrent power. The largergauge wire used
in the secondary winding is necessary due to the increase in current. The primary winding,
which doesn’t have to conduct as much current, may be made of smallergauge wire.
In case you were wondering, it is possible to operate either of these transformer types back
wards (powering the secondary winding with an AC source and letting the primary winding
power a load) to perform the opposite function: a stepup can function as a stepdown and
visaversa. However, as we saw in the first section of this chapter, efficient operation of a
transformer requires that the individual winding inductances be engineered for specific op
erating ranges of voltage and current, so if a transformer is to be used “backwards” like this
it must be employed within the original design parameters of voltage and current for each
winding, lest it prove to be inefficient (or lest it be damaged by excessive voltage or current!).
Transformers are often constructed in such a way that it is not obvious which wires lead
to the primary winding and which lead to the secondary. One convention used in the electric
power industry to help alleviate confusion is the use of “H” designations for the highervoltage
winding (the primary winding in a stepdown unit; the secondary winding in a stepup) and “X”
designations for the lowervoltage winding. Therefore, a simple power transformer will have
234 CHAPTER 9. TRANSFORMERS
wires labeled “H1”, “H2”, “X1”, and “X2”. There is usually significance to the numbering of the
wires (H1 versus H2, etc.), which we’ll explore a little later in this chapter.
The fact that voltage and current get “stepped” in opposite directions (one up, the other
down) makes perfect sense when you recall that power is equal to voltage times current, and
realize that transformers cannot produce power, only convert it. Any device that could output
more power than it took in would violate the Law of Energy Conservation in physics, namely
that energy cannot be created or destroyed, only converted. As with the first transformer
example we looked at, power transfer efficiency is very good from the primary to the secondary
sides of the device.
The practical significance of this is made more apparent when an alternative is consid
ered: before the advent of efficient transformers, voltage/current level conversion could only be
achieved through the use of motor/generator sets. A drawing of a motor/generator set reveals
the basic principle involved: (Figure 9.17)
Motor Generator
Shaft
coupling
Power
in
Power
out
A motor/generator set
Figure 9.17: Motor generator illustrates the basic principle of the transformer.
In such amachine, a motor is mechanically coupled to a generator, the generator designed to
produce the desired levels of voltage and current at the rotating speed of the motor. While both
motors and generators are fairly efficient devices, the use of both in this fashion compounds
their inefficiencies so that the overall efficiency is in the range of 90% or less. Furthermore,
because motor/generator sets obviously require moving parts, mechanical wear and balance
are factors influencing both service life and performance. Transformers, on the other hand, are
able to convert levels of AC voltage and current at very high efficiencies with no moving parts,
making possible the widespread distribution and use of electric power we take for granted.
In all fairness it should be noted that motor/generator sets have not necessarily been ob
soleted by transformers for all applications. While transformers are clearly superior over
motor/generator sets for AC voltage and current level conversion, they cannot convert one
frequency of AC power to another, or (by themselves) convert DC to AC or visaversa. Mo
tor/generator sets can do all these things with relative simplicity, albeit with the limitations of
efficiency and mechanical factors already described. Motor/generator sets also have the unique
property of kinetic energy storage: that is, if the motor’s power supply is momentarily inter
rupted for any reason, its angular momentum (the inertia of that rotating mass) will maintain
rotation of the generator for a short duration, thus isolating any loads powered by the genera
9.2. STEPUP AND STEPDOWN TRANSFORMERS 235
tor from “glitches” in the main power system.
Looking closely at the numbers in the SPICE analysis, we should see a correspondence
between the transformer’s ratio and the two inductances. Notice how the primary inductor (l1)
has 100 times more inductance than the secondary inductor (10000 H versus 100 H), and that
the measured voltage stepdown ratio was 10 to 1. The winding with more inductance will have
higher voltage and less current than the other. Since the two inductors are wound around the
same core material in the transformer (for the most efficient magnetic coupling between the
two), the parameters affecting inductance for the two coils are equal except for the number of
turns in each coil. If we take another look at our inductance formula, we see that inductance
is proportional to the square of the number of coil turns:
Where,
N = Number of turns in wire coil (straight wire = 1)
L = N
2µA
l
L =
µ =
A =
l =
Inductance of coil in Henrys
Permeability of core material (absolute, not relative)
Area of coil in square meters
Average length of coil in meters
So, it should be apparent that our two inductors in the last SPICE transformer example cir
cuit – with inductance ratios of 100:1 – should have coil turn ratios of 10:1, because 10 squared
equals 100. This works out to be the same ratio we found between primary and secondary volt
ages and currents (10:1), so we can say as a rule that the voltage and current transformation
ratio is equal to the ratio of winding turns between primary and secondary.
load
many turns few turns
high voltage
low current
low voltage
high current
Stepdown transformer
Figure 9.18: Stepdown transformer: (many turns :few turns).
The stepup/stepdown effect of coil turn ratios in a transformer (Figure 9.18) is analogous
to gear tooth ratios in mechanical gear systems, transforming values of speed and torque in
much the same way: (Figure 9.19)
Stepup and stepdown transformers for power distribution purposes can be gigantic in pro
portion to the power transformers previously shown, some units standing as tall as a home.
The following photograph shows a substation transformer standing about twelve feet tall: (Fig
ure 9.20)
236 CHAPTER 9. TRANSFORMERS
LARGE GEAR
SMALL GEAR
(many teeth)
(few teeth)
high torque
low speed
low torque
high speed
Figure 9.19: Torque reducing gear train steps torque down, while stepping speed up.
Figure 9.20: Substation transformer.
9.3. ELECTRICAL ISOLATION 237
• REVIEW:
• Transformers “step up” or “step down” voltage according to the ratios of primary to sec
ondary wire turns.
•
Voltage transformation ratio =
Nsecondary
Nprimary
Current transformation ratio =
Nprimary
Nsecondary
N = number of turns in winding
Where,
• A transformer designed to increase voltage from primary to secondary is called a step
up transformer. A transformer designed to reduce voltage from primary to secondary is
called a stepdown transformer.
• The transformation ratio of a transformer will be equal to the square root of its primary
to secondary inductance (L) ratio.
•
Voltage transformation ratio =
Lsecondary
Lprimary
9.3 Electrical isolation
Aside from the ability to easily convert between different levels of voltage and current in AC
and DC circuits, transformers also provide an extremely useful feature called isolation, which
is the ability to couple one circuit to another without the use of direct wire connections. We can
demonstrate an application of this effect with another SPICE simulation: this time showing
“ground” connections for the two circuits, imposing a high DC voltage between one circuit and
ground through the use of an additional voltage source:(Figure 9.21)
v1 1 0 ac 10 sin
rbogus1 1 2 1e12
v2 5 0 dc 250
l1 2 0 10000
l2 3 5 100
k l1 l2 0.999
vi1 3 4 ac 0
rload 4 5 1k
.ac lin 1 60 60
.print ac v(2,0) i(v1)
.print ac v(3,5) i(vi1)
.end
238 CHAPTER 9. TRANSFORMERS
(for SPICE to measure current)
0 V
1
0
2
0
3 4
5 5
0
Rbogus
V1 10 V L1 L2
10 kH 100 H
Vi1
Rload 1 kΩ
V2 250 V
Figure 9.21: Transformer isolates 10 Vac at V1 from 250 VDC at V2.
DC voltages referenced to ground (node 0):
(1) 0.0000 (2) 0.0000 (3) 250.0000
(4) 250.0000 (5) 250.0000
AC voltages:
freq v(2) i(v1)
6.000E+01 1.000E+01 9.975E05 Primary winding
freq v(3,5) i(vi1)
6.000E+01 9.962E01 9.962E04 Secondary winding
9.4. PHASING 239
SPICE shows the 250 volts DC being impressed upon the secondary circuit elements with
respect to ground, (Figure 9.21) but as you can see there is no effect on the primary circuit (zero
DC voltage) at nodes 1 and 2, and the transformation of AC power from primary to secondary
circuits remains the same as before. The impressed voltage in this example is often called a
commonmode voltage because it is seen at more than one point in the circuit with reference
to the common point of ground. The transformer isolates the commonmode voltage so that it
is not impressed upon the primary circuit at all, but rather isolated to the secondary side. For
the record, it does not matter that the commonmode voltage is DC, either. It could be AC, even
at a different frequency, and the transformer would isolate it from the primary circuit all the
same.
There are applications where electrical isolation is needed between two AC circuit without
any transformation of voltage or current levels. In these instances, transformers called isola
tion transformers having 1:1 transformation ratios are used. A benchtop isolation transformer
is shown in Figure 9.22.
Figure 9.22: Isolation transformer isolates power out from the power line.
• REVIEW:
• By being able to transfer power from one circuit to another without the use of intercon
necting conductors between the two circuits, transformers provide the useful feature of
electrical isolation.
• Transformers designed to provide electrical isolation without stepping voltage and cur
rent either up or down are called isolation transformers.
9.4 Phasing
Since transformers are essentially AC devices, we need to be aware of the phase relationships
between the primary and secondary circuits. Using our SPICE example from before, we can
240 CHAPTER 9. TRANSFORMERS
plot the waveshapes (Figure 9.23) for the primary and secondary circuits and see the phase
relations for ourselves:
spice transient analysis file for use with nutmeg:
transformer
v1 1 0 sin(0 15 60 0 0)
rbogus1 1 2 1e12
v2 5 0 dc 250
l1 2 0 10000
l2 3 5 100
k l1 l2 0.999
vi1 3 4 ac 0
rload 4 5 1k
.tran 0.5m 17m
.end
nutmeg commands:
setplot tran1
plot v(2) v(3,5)
Figure 9.23: Secondary voltage V(3,5) is inphase with primary voltage V(2), and stepped down
by factor of ten.
In going from primary, V(2), to secondary, V(3,5), the voltage was stepped down by a factor
of ten, (Figure 9.23) , and the current was stepped up by a factor of 10. (Figure 9.24) Both
9.4. PHASING 241
current (Figure 9.24) and voltage (Figure 9.23) waveforms are inphase in going from primary
to secondary.
nutmeg commands:
setplot tran1
plot I(L1#branch) I(L2#branch)
Figure 9.24: Primary and secondary currents are inphase. Secondary current is stepped up
by a factor of ten.
It would appear that both voltage and current for the two transformer windings are in
phase with each other, at least for our resistive load. This is simple enough, but it would be
nice to know which way we should connect a transformer in order to ensure the proper phase
relationships be kept. After all, a transformer is nothing more than a set of magnetically
linked inductors, and inductors don’t usually come with polarity markings of any kind. If we
were to look at an unmarked transformer, we would have no way of knowing which way to hook
it up to a circuit to get inphase (or 180o outofphase) voltage and current: (Figure 9.25)
+

+
 +

or ???
Figure 9.25: As a practical matter, the polarity of a transformer can be ambiguous.
242 CHAPTER 9. TRANSFORMERS
Since this is a practical concern, transformer manufacturers have come up with a sort of
polarity marking standard to denote phase relationships. It is called the dot convention, and
is nothing more than a dot placed next to each corresponding leg of a transformer winding:
(Figure 9.26)
Figure 9.26: A pair of dots indicates like polarity.
Typically, the transformer will come with some kind of schematic diagram labeling the wire
leads for primary and secondary windings. On the diagram will be a pair of dots similar to
what is seen above. Sometimes dots will be omitted, but when “H” and “X” labels are used
to label transformer winding wires, the subscript numbers are supposed to represent winding
polarity. The “1” wires (H1 and X1) represent where the polaritymarking dots would normally
be placed.
The similar placement of these dots next to the top ends of the primary and secondary
windings tells us that whatever instantaneous voltage polarity seen across the primary wind
ing will be the same as that across the secondary winding. In other words, the phase shift from
primary to secondary will be zero degrees.
On the other hand, if the dots on each winding of the transformer do not match up, the
phase shift will be 180o between primary and secondary, like this: (Figure 9.27)
Figure 9.27: Out of phase: primary red to dot, secondary black to dot.
Of course, the dot convention only tells you which end of each winding is which, relative to
the other winding(s). If you want to reverse the phase relationship yourself, all you have to do
is swap the winding connections like this: (Figure 9.28)
• REVIEW:
9.5. WINDING CONFIGURATIONS 243
Figure 9.28: In phase: primary red to dot, secondary red to dot.
• The phase relationships for voltage and current between primary and secondary circuits
of a transformer are direct: ideally, zero phase shift.
• The dot convention is a type of polarity marking for transformer windings showing which
end of the winding is which, relative to the other windings.
9.5 Winding configurations
Transformers are very versatile devices. The basic concept of energy transfer between mutual
inductors is useful enough between a single primary and single secondary coil, but transform
ers don’t have to be made with just two sets of windings. Consider this transformer circuit:
(Figure 9.29)
load #1
load #2
Figure 9.29: Transformer with multiple secondaries, provides multiple output voltages.
Here, three inductor coils share a common magnetic core, magnetically “coupling” or “link
ing” them together. The relationship of winding turn ratios and voltage ratios seen with a
single pair of mutual inductors still holds true here for multiple pairs of coils. It is entirely
possible to assemble a transformer such as the one above (one primary winding, two secondary
windings) in which one secondary winding is a stepdown and the other is a stepup. In fact,
this design of transformer was quite common in vacuum tube power supply circuits, which
were required to supply low voltage for the tubes’ filaments (typically 6 or 12 volts) and high
voltage for the tubes’ plates (several hundred volts) from a nominal primary voltage of 110
volts AC. Not only are voltages and currents of completely different magnitudes possible with
such a transformer, but all circuits are electrically isolated from one another.
244 CHAPTER 9. TRANSFORMERS
Figure 9.30: Photograph of multiplewinding transformer with six windings, a primary and
five secondaries.
The transformer in Figure 9.30 is intended to provide both high and low voltages necessary
in an electronic system using vacuum tubes. Low voltage is required to power the filaments
of vacuum tubes, while high voltage is required to create the potential difference between the
plate and cathode elements of each tube. One transformer with multiple windings suffices
elegantly to provide all the necessary voltage levels from a single 115 V source. The wires for
this transformer (15 of them!) are not shown in the photograph, being hidden from view.
If electrical isolation between secondary circuits is not of great importance, a similar effect
can be obtained by “tapping” a single secondary winding at multiple points along its length,
like Figure 9.31.
load #1
load #2
Figure 9.31: A single tapped secondary provides multiple voltages.
A tap is nothing more than a wire connection made at some point on a winding between
the very ends. Not surprisingly, the winding turn/voltage magnitude relationship of a normal
transformer holds true for all tapped segments of windings. This fact can be exploited to
produce a transformer capable of multiple ratios: (Figure 9.32)
Carrying the concept of winding taps further, we end up with a “variable transformer,”
9.5. WINDING CONFIGURATIONS 245
load
multipole
switch
Figure 9.32: A tapped secondary using a switch to select one of many possible voltages.
where a sliding contact is moved along the length of an exposed secondary winding, able to
connect with it at any point along its length. The effect is equivalent to having a winding tap
at every turn of the winding, and a switch with poles at every tap position: (Figure 9.33)
load
Variable transformer
Figure 9.33: A sliding contact on the secondary continuously varies the secondary voltage.
One consumer application of the variable transformer is in speed controls for model train
sets, especially the train sets of the 1950’s and 1960’s. These transformers were essentially
stepdown units, the highest voltage obtainable from the secondary winding being substan
tially less than the primary voltage of 110 to 120 volts AC. The variablesweep contact provided
a simple means of voltage control with little wasted power, much more efficient than control
using a variable resistor!
Movingslide contacts are too impractical to be used in large industrial power transformer
designs, but multipole switches and winding taps are common for voltage adjustment. Adjust
ments need to be made periodically in power systems to accommodate changes in loads over
months or years in time, and these switching circuits provide a convenient means. Typically,
246 CHAPTER 9. TRANSFORMERS
such “tap switches” are not engineered to handle fullload current, but must be actuated only
when the transformer has been deenergized (no power).
Seeing as how we can tap any transformer winding to obtain the equivalent of several
windings (albeit with loss of electrical isolation between them), it makes sense that it should be
possible to forego electrical isolation altogether and build a transformer from a single winding.
Indeed this is possible, and the resulting device is called an autotransformer: (Figure 9.34)
load
Autotransformer
Figure 9.34: This autotransformer steps voltage up with a single tapped winding, saving cop
per, sacrificing isolation.
The autotransformer depicted above performs a voltage stepup function. A stepdown au
totransformer would look something like Figure 9.35.
load
Autotransformer
Figure 9.35: This auto transformer steps voltage down with a single coppersaving tapped
winding.
9.5. WINDING CONFIGURATIONS 247
Autotransformers find popular use in applications requiring a slight boost or reduction in
voltage to a load. The alternative with a normal (isolated) transformer would be to either have
just the right primary/secondary winding ratio made for the job or use a stepdown configu
ration with the secondary winding connected in seriesaiding (“boosting”) or seriesopposing
(“bucking”) fashion. Primary, secondary, and load voltages are given to illustrate how this
would work.
First, the “boosting” configuration. In Figure 9.36 the secondary coil’s polarity is oriented
so that its voltage directly adds to the primary voltage.
"boosting"
120 V 30 V 150 V
Figure 9.36: Ordinary transformer wired as an autotransformer to boost the line voltage.
Next, the “bucking” configuration. In Figure 9.37 the secondary coil’s polarity is oriented so
that its voltage directly subtracts from the primary voltage:
"bucking"
120 V 30 V 90 V
Figure 9.37: Ordinary transformer wired as an autotransformer to buck the line voltage down.
The prime advantage of an autotransformer is that the same boosting or bucking function
is obtained with only a single winding, making it cheaper and lighter to manufacture than a
regular (isolating) transformer having both primary and secondary windings.
Like regular transformers, autotransformer windings can be tapped to provide variations
in ratio. Additionally, they can be made continuously variable with a sliding contact to tap
the winding at any point along its length. The latter configuration is popular enough to have
earned itself its own name: the Variac. (Figure 9.38)
Small variacs for benchtop use are popular pieces of equipment for the electronics experi
menter, being able to step household AC voltage down (or sometimes up as well) with a wide,
fine range of control by a simple twist of a knob.
248 CHAPTER 9. TRANSFORMERS
load
The "Variac"
variable autotransformer
Figure 9.38: A variac is an autotransformer with a sliding tap.
• REVIEW:
• Transformers can be equipped with more than just a single primary and single secondary
winding pair. This allows for multiple stepup and/or stepdown ratios in the same device.
• Transformer windings can also be “tapped:” that is, intersected at many points to seg
ment a single winding into sections.
• Variable transformers can be made by providing a movable arm that sweeps across the
length of a winding, making contact with the winding at any point along its length. The
winding, of course, has to be bare (no insulation) in the area where the arm sweeps.
• An autotransformer is a single, tapped inductor coil used to step up or step down voltage
like a transformer, except without providing electrical isolation.
• A Variac is a variable autotransformer.
9.6 Voltage regulation
As we saw in a few SPICE analyses earlier in this chapter, the output voltage of a transformer
varies some with varying load resistances, even with a constant voltage input. The degree
of variance is affected by the primary and secondary winding inductances, among other fac
tors, not the least of which includes winding resistance and the degree of mutual inductance
(magnetic coupling) between the primary and secondary windings. For power transformer ap
plications, where the transformer is seen by the load (ideally) as a constant source of voltage,
it is good to have the secondary voltage vary as little as possible for wide variances in load
current.
The measure of how well a power transformer maintains constant secondary voltage over a
range of load currents is called the transformer’s voltage regulation. It can be calculated from
the following formula:
Regulation percentage =
Enoload  Efullload (100%)
Efullload
9.6. VOLTAGE REGULATION 249
“Fullload” means the point at which the transformer is operating at maximum permissible
secondary current. This operating point will be determined primarily by the winding wire
size (ampacity) and the method of transformer cooling. Taking our first SPICE transformer
simulation as an example, let’s compare the output voltage with a 1 kΩ load versus a 200 Ω
load (assuming that the 200 Ω load will be our “full load” condition). Recall if you will that our
constant primary voltage was 10.00 volts AC:
freq v(3,5) i(vi1)
6.000E+01 9.962E+00 9.962E03 Output with 1k ohm load
freq v(3,5) i(vi1)
6.000E+01 9.348E+00 4.674E02 Output with 200 ohm load
Notice how the output voltage decreases as the load gets heavier (more current). Now let’s
take that same transformer circuit and place a load resistance of extremely high magnitude
across the secondary winding to simulate a “noload” condition: (See ”transformer” spice list”)
transformer
v1 1 0 ac 10 sin
rbogus1 1 2 1e12
rbogus2 5 0 9e12
l1 2 0 100
l2 3 5 100
k l1 l2 0.999
vi1 3 4 ac 0
rload 4 5 9e12
.ac lin 1 60 60
.print ac v(2,0) i(v1)
.print ac v(3,5) i(vi1)
.end
freq v(2) i(v1)
6.000E+01 1.000E+01 2.653E04
freq v(3,5) i(vi1)
6.000E+01 9.990E+00 1.110E12 Output with (almost) no load
So, we see that our output (secondary) voltage spans a range of 9.990 volts at (virtually) no
load and 9.348 volts at the point we decided to call “full load.” Calculating voltage regulation
with these figures, we get:
Regulation percentage = (100%)9.990 V  9.348 V
Regulation percentage = 6.8678 %
9.348 V
Incidentally, this would be considered rather poor (or “loose”) regulation for a power trans
former. Powering a simple resistive load like this, a good power transformer should exhibit
250 CHAPTER 9. TRANSFORMERS
a regulation percentage of less than 3%. Inductive loads tend to create a condition of worse
voltage regulation, so this analysis with purely resistive loads was a “bestcase” condition.
There are some applications, however, where poor regulation is actually desired. One such
case is in discharge lighting, where a stepup transformer is required to initially generate a
high voltage (necessary to “ignite” the lamps), then the voltage is expected to drop off once the
lamp begins to draw current. This is because discharge lamps’ voltage requirements tend to be
much lower after a current has been established through the arc path. In this case, a stepup
transformer with poor voltage regulation suffices nicely for the task of conditioning power to
the lamp.
Another application is in current control for AC arc welders, which are nothing more than
stepdown transformers supplying lowvoltage, highcurrent power for the welding process. A
high voltage is desired to assist in “striking” the arc (getting it started), but like the discharge
lamp, an arc doesn’t require as much voltage to sustain itself once the air has been heated to
the point of ionization. Thus, a decrease of secondary voltage under high load current would
be a good thing. Some arc welder designs provide arc current adjustment by means of a mov
able iron core in the transformer, cranked in or out of the winding assembly by the operator.
Moving the iron slug away from the windings reduces the strength of magnetic coupling be
tween the windings, which diminishes noload secondary voltage and makes for poorer voltage
regulation.
No exposition on transformer regulation could be called complete without mention of an un
usual device called a ferroresonant transformer. “Ferroresonance” is a phenomenon associated
with the behavior of iron cores while operating near a point of magnetic saturation (where the
core is so strongly magnetized that further increases in winding current results in little or no
increase in magnetic flux).
While being somewhat difficult to describe without going deep into electromagnetic the
ory, the ferroresonant transformer is a power transformer engineered to operate in a condition
of persistent core saturation. That is, its iron core is “stuffed full” of magnetic lines of flux
for a large portion of the AC cycle so that variations in supply voltage (primary winding cur
rent) have little effect on the core’s magnetic flux density, which means the secondary winding
outputs a nearly constant voltage despite significant variations in supply (primary winding)
voltage. Normally, core saturation in a transformer results in distortion of the sinewave shape,
and the ferroresonant transformer is no exception. To combat this side effect, ferroresonant
transformers have an auxiliary secondary winding paralleled with one or more capacitors,
forming a resonant circuit tuned to the power supply frequency. This “tank circuit” serves as
a filter to reject harmonics created by the core saturation, and provides the added benefit of
storing energy in the form of AC oscillations, which is available for sustaining output winding
voltage for brief periods of input voltage loss (milliseconds’ worth of time, but certainly better
than nothing). (Figure 9.39)
In addition to blocking harmonics created by the saturated core, this resonant circuit also
“filters out” harmonic frequencies generated by nonlinear (switching) loads in the secondary
winding circuit and any harmonics present in the source voltage, providing “clean” power to
the load.
Ferroresonant transformers offer several features useful in AC power conditioning: con
stant output voltage given substantial variations in input voltage, harmonic filtering between
the power source and the load, and the ability to “ride through” brief losses in power by keeping
a reserve of energy in its resonant tank circuit. These transformers are also highly tolerant
9.7. SPECIAL TRANSFORMERS AND APPLICATIONS 251
AC power
input
AC power
output
Resonant LC circuit
Figure 9.39: Ferroresonant transformer provides voltage regulation of the output.
of excessive loading and transient (momentary) voltage surges. They are so tolerant, in fact,
that some may be briefly paralleled with unsynchronized AC power sources, allowing a load
to be switched from one source of power to another in a “makebeforebreak” fashion with no
interruption of power on the secondary side!
Unfortunately, these devices have equally noteworthy disadvantages: they waste a lot of
energy (due to hysteresis losses in the saturated core), generating significant heat in the pro
cess, and are intolerant of frequency variations, which means they don’t work very well when
powered by small enginedriven generators having poor speed regulation. Voltages produced in
the resonant winding/capacitor circuit tend to be very high, necessitating expensive capacitors
and presenting the service technician with very dangerous working voltages. Some applica
tions, though, may prioritize the ferroresonant transformer’s advantages over its disadvan
tages. Semiconductor circuits exist to “condition” AC power as an alternative to ferroresonant
devices, but none can compete with this transformer in terms of sheer simplicity.
• REVIEW:
• Voltage regulation is the measure of how well a power transformer can maintain constant
secondary voltage given a constant primary voltage and wide variance in load current.
The lower the percentage (closer to zero), the more stable the secondary voltage and the
better the regulation it will provide.
• A ferroresonant transformer is a special transformer designed to regulate voltage at a
stable level despite wide variation in input voltage.
9.7 Special transformers and applications
9.7.1 Impedance matching
Because transformers can step voltage and current to different levels, and because power is
transferred equivalently between primary and secondary windings, they can be used to “con
vert” the impedance of a load to a different level. That last phrase deserves some explanation,
so let’s investigate what it means.
The purpose of a load (usually) is to do something productive with the power it dissipates.
In the case of a resistive heating element, the practical purpose for the power dissipated is to
252 CHAPTER 9. TRANSFORMERS
heat something up. Loads are engineered to safely dissipate a certain maximum amount of
power, but two loads of equal power rating are not necessarily identical. Consider these two
1000 watt resistive heating elements: (Figure 9.40)
250 V
I = 4 A
62.5 ΩRload
Pload = 1000 W
125 V
I = 8 A
15.625 ΩRload
Pload = 1000 W
Figure 9.40: Heating elements dissipate 1000 watts, at different voltage and current ratings.
Both heaters dissipate exactly 1000 watts of power, but they do so at different voltage and
current levels (either 250 volts and 4 amps, or 125 volts and 8 amps). Using Ohm’s Law to
determine the necessary resistance of these heating elements (R=E/I), we arrive at figures of
62.5 Ω and 15.625 Ω, respectively. If these are AC loads, we might refer to their opposition
to current in terms of impedance rather than plain resistance, although in this case that’s all
they’re composed of (no reactance). The 250 volt heater would be said to be a higher impedance
load than the 125 volt heater.
If we desired to operate the 250 volt heater element directly on a 125 volt power system,
we would end up being disappointed. With 62.5 Ω of impedance (resistance), the current would
only be 2 amps (I=E/R; 125/62.5), and the power dissipation would only be 250 watts (P=IE;
125 x 2), or onequarter of its rated power. The impedance of the heater and the voltage of our
source would be mismatched, and we couldn’t obtain the full rated power dissipation from the
heater.
All hope is not lost, though. With a stepup transformer, we could operate the 250 volt
heater element on the 125 volt power system like Figure 9.41.
1000 watts dissipation at the load resistor
125 V
I = 8 A
I = 4 A
250 V 62.5 Ω
Rload
!
Figure 9.41: Stepup transformer operates 1000 watt 250 V heater from 125 V power source
The ratio of the transformer’s windings provides the voltage stepup and current stepdown
we need for the otherwise mismatched load to operate properly on this system. Take a close
look at the primary circuit figures: 125 volts at 8 amps. As far as the power supply “knows,”
its powering a 15.625 Ω (R=E/I) load at 125 volts, not a 62.5 Ω load! The voltage and current
figures for the primary winding are indicative of 15.625 Ω load impedance, not the actual 62.5
Ω of the load itself. In other words, not only has our stepup transformer transformed voltage
and current, but it has transformed impedance as well.
The transformation ratio of impedance is the square of the voltage/current transformation
ratio, the same as the winding inductance ratio:
9.7. SPECIAL TRANSFORMERS AND APPLICATIONS 253
Voltage transformation ratio =
Nsecondary
Nprimary
Current transformation ratio =
Nprimary
Nsecondary
N = number of turns in winding
Where,
Impedance transformation ratio =
Nprimary
Nsecondary
2
Inductance ratio =
Nsecondary
Nprimary
2
This concurs with our example of the 2:1 stepup transformer and the impedance ratio of
62.5 Ω to 15.625 Ω (a 4:1 ratio, which is 2:1 squared). Impedance transformation is a highly
useful ability of transformers, for it allows a load to dissipate its full rated power even if the
power system is not at the proper voltage to directly do so.
Recall from our study of network analysis the Maximum Power Transfer Theorem, which
states that the maximum amount of power will be dissipated by a load resistance when that
load resistance is equal to the Thevenin/Norton resistance of the network supplying the power.
Substitute the word “impedance” for “resistance” in that definition and you have the AC version
of that Theorem. If we’re trying to obtain theoretical maximum power dissipation from a
load, we must be able to properly match the load impedance and source (Thevenin/Norton)
impedance together. This is generally more of a concern in specialized electric circuits such as
radio transmitter/antenna and audio amplifier/speaker systems. Let’s take an audio amplifier
system and see how it works: (Figure 9.42)
With an internal impedance of 500 Ω, the amplifier can only deliver full power to a load
(speaker) also having 500 Ω of impedance. Such a load would drop higher voltage and draw
less current than an 8 Ω speaker dissipating the same amount of power. If an 8 Ω speaker were
connected directly to the 500 Ω amplifier as shown, the impedance mismatch would result in
very poor (low peak power) performance. Additionally, the amplifier would tend to dissipate
more than its fair share of power in the form of heat trying to drive the low impedance speaker.
To make this system work better, we can use a transformer to match these mismatched
impedances. Since we’re going from a high impedance (high voltage, low current) supply to
a low impedance (low voltage, high current) load, we’ll need to use a stepdown transformer:
(Figure 9.43)
To obtain an impedance transformation ratio of 500:8, we would need a winding ratio equal
to the square root of 500:8 (the square root of 62.5:1, or 7.906:1). With such a transformer in
place, the speaker will load the amplifier to just the right degree, drawing power at the correct
voltage and current levels to satisfy the Maximum Power Transfer Theorem and make for the
254 CHAPTER 9. TRANSFORMERS
Speaker
Audio amplifier
Thevenin/Norton
Z = 500 Ω
Z = 8 Ω
Speaker
Z = 8 Ω
ZThevenin
EThevenin
. . . equivalent to . . .
500 Ω
Figure 9.42: Amplifier with impedance of 500 Ω drives 8 Ω at much less than maximum power.
Speaker
Audio amplifier
Thevenin/Norton
"matching"
transformer
impedance
impedance ratio = 500 : 8 winding ratio = 7.906 : 1
Z = 500 Ω
Z = 8 Ω
Figure 9.43: Impedance matching transformer matches 500 Ω amplifier to 8 Ω speaker for
maximum efficiency.
9.7. SPECIAL TRANSFORMERS AND APPLICATIONS 255
most efficient power delivery to the load. The use of a transformer in this capacity is called
impedance matching.
Anyone who has ridden a multispeed bicycle can intuitively understand the principle of
impedance matching. A human’s legs will produce maximum power when spinning the bicycle
crank at a particular speed (about 60 to 90 revolution per minute). Above or below that ro
tational speed, human leg muscles are less efficient at generating power. The purpose of the
bicycle’s “gears” is to impedancematch the rider’s legs to the riding conditions so that they
always spin the crank at the optimum speed.
If the rider attempts to start moving while the bicycle is shifted into its “top” gear, he or she
will find it very difficult to get moving. Is it because the rider is weak? No, its because the high
stepup ratio of the bicycle’s chain and sprockets in that top gear presents a mismatch between
the conditions (lots of inertia to overcome) and their legs (needing to spin at 6090 RPM for
maximum power output). On the other hand, selecting a gear that is too low will enable the
rider to get moving immediately, but limit the top speed they will be able to attain. Again, is
the lack of speed an indication of weakness in the bicyclist’s legs? No, its because the lower
speed ratio of the selected gear creates another type of mismatch between the conditions (low
load) and the rider’s legs (losing power if spinning faster than 90 RPM). It is much the same
with electric power sources and loads: there must be an impedance match for maximum system
efficiency. In AC circuits, transformers perform the same matching function as the sprockets
and chain (“gears”) on a bicycle to match otherwise mismatched sources and loads.
Impedance matching transformers are not fundamentally different from any other type of
transformer in construction or appearance. A small impedancematching transformer (about
two centimeters in width) for audiofrequency applications is shown in the following photo
graph: (Figure 9.44)
Figure 9.44: Audio frequency impedance matching transformer.
Another impedancematching transformer can be seen on this printed circuit board, in the
upper right corner, to the immediate left of resistors R2 and R1. It is labeled “T1”: (Figure 9.45)
256 CHAPTER 9. TRANSFORMERS
Figure 9.45: Printed circuit board mounted audio impedance matching transformer, top right.
9.7.2 Potential transformers
Transformers can also be used in electrical instrumentation systems. Due to transformers’
ability to step up or step down voltage and current, and the electrical isolation they provide,
they can serve as a way of connecting electrical instrumentation to highvoltage, high current
power systems. Suppose we wanted to accurately measure the voltage of a 13.8 kV power
system (a very common power distribution voltage in American industry): (Figure 9.46)
loadhighvoltage
power source 13.8 kV
Figure 9.46: Direct measurement of high voltage by a voltmeter is a potential safety hazard.
Designing, installing, and maintaining a voltmeter capable of directly measuring 13,800
volts AC would be no easy task. The safety hazard alone of bringing 13.8 kV conductors into an
instrument panel would be severe, not to mention the design of the voltmeter itself. However,
by using a precision stepdown transformer, we can reduce the 13.8 kV down to a safe level of
voltage at a constant ratio, and isolate it from the instrument connections, adding an additional
level of safety to the metering system: (Figure 9.47)
9.7. SPECIAL TRANSFORMERS AND APPLICATIONS 257
loadhighvoltage
power source
fusefuse
V
precision
stepdown
ratio
PT
grounded for
safety
13.8 kV
0120 VAC voltmeter range
Figure 9.47: Instrumentation application:“Potential transformer” precisely scales dangerous
high voltage to a safe value applicable to a conventional voltmeter.
Now the voltmeter reads a precise fraction, or ratio, of the actual system voltage, its scale
set to read as though it were measuring the voltage directly. The transformer keeps the in
strument voltage at a safe level and electrically isolates it from the power system, so there is
no direct connection between the power lines and the instrument or instrument wiring. When
used in this capacity, the transformer is called a Potential Transformer, or simply PT.
Potential transformers are designed to provide as accurate a voltage stepdown ratio as
possible. To aid in precise voltage regulation, loading is kept to a minimum: the voltmeter is
made to have high input impedance so as to draw as little current from the PT as possible. As
you can see, a fuse has been connected in series with the PTs primary winding, for safety and
ease of disconnecting the PT from the circuit.
A standard secondary voltage for a PT is 120 volts AC, for fullrated power line voltage.
The standard voltmeter range to accompany a PT is 150 volts, fullscale. PTs with custom
winding ratios can be manufactured to suit any application. This lends itself well to industry
standardization of the actual voltmeter instruments themselves, since the PT will be sized to
step the system voltage down to this standard instrument level.
9.7.3 Current transformers
Following the same line of thinking, we can use a transformer to step down current through
a power line so that we are able to safely and easily measure high system currents with inex
pensive ammeters. Of course, such a transformer would be connected in series with the power
line, like (Figure 9.48).
Note that while the PT is a stepdown device, the Current Transformer (or CT) is a stepup
device (with respect to voltage), which is what is needed to step down the power line current.
Quite often, CTs are built as donutshaped devices through which the power line conductor is
run, the power line itself acting as a singleturn primary winding: (Figure 9.49)
Some CTs are made to hinge open, allowing insertion around a power conductor without
258 CHAPTER 9. TRANSFORMERS
load
fusefuse
V
precision
stepdown
ratio
PT
grounded for
safety
13.8 kV
0120 VAC voltmeter range
Instrument application: the "Current Transformer"
A
grounded for
safety
05 A ammeter range
CT
Figure 9.48: Instrumentation application: “Currrent transformer” steps high current down to
a value applicable to a conventional ammeter.
Figure 9.49: Current conductor to be measured is threaded through the opening. Scaled down
current is available on wire leads.
9.7. SPECIAL TRANSFORMERS AND APPLICATIONS 259
disturbing the conductor at all. The industry standard secondary current for a CT is a range of
0 to 5 amps AC. Like PTs, CTs can be made with custom winding ratios to fit almost any appli
cation. Because their “full load” secondary current is 5 amps, CT ratios are usually described
in terms of fullload primary amps to 5 amps, like this:
600 : 5 ratio (for measuring up to 600 A line current)
100 : 5 ratio (for measuring up to 100 A line current)
1k : 5 ratio (for measuring up to 1000 A line current)
The “donut” CT shown in the photograph has a ratio of 50:5. That is, when the conductor
through the center of the torus is carrying 50 amps of current (AC), there will be 5 amps of
current induced in the CT’s winding.
Because CTs are designed to be powering ammeters, which are lowimpedance loads, and
they are wound as voltage stepup transformers, they should never, ever be operated with an
opencircuited secondary winding. Failure to heed this warning will result in the CT producing
extremely high secondary voltages, dangerous to equipment and personnel alike. To facili
tate maintenance of ammeter instrumentation, shortcircuiting switches are often installed in
parallel with the CT’s secondary winding, to be closed whenever the ammeter is removed for
service: (Figure 9.50)
power conductor current
CT
close switch BEFORE
disconnecting ammeter!
ground connection(for safety)
05 A meter movement range
Figure 9.50: Shortcircuit switch allows ammeter to be removed from an active current trans
former circuit.
Though it may seem strange to intentionally shortcircuit a power system component, it is
perfectly proper and quite necessary when working with current transformers.
9.7.4 Air core transformers
Another kind of special transformer, seen often in radiofrequency circuits, is the air core trans
former. (Figure 9.51) True to its name, an air core transformer has its windings wrapped
around a nonmagnetic form, usually a hollow tube of some material. The degree of coupling
(mutual inductance) between windings in such a transformer is many times less than that
260 CHAPTER 9. TRANSFORMERS
of an equivalent ironcore transformer, but the undesirable characteristics of a ferromagnetic
core (eddy current losses, hysteresis, saturation, etc.) are completely eliminated. It is in high
frequency applications that these effects of iron cores are most problematic.
(b)(a)
Figure 9.51: Air core transformers may be wound on cylindrical (a) or toroidal (b) forms. Center
tapped primary with secondary (a). Bifilar winding on toroidal form (b).
The inside tapped solenoid winding, (Figure (a) 9.51), without the over winding, could match
unequal impedances when DC isolation is not required. When isolation is required the over
winding is added over one end of the main winding. Air core transformers are used at radio
frequencies when iron core losses are too high. Frequently air core transformers are paralleled
with a capacitor to tune it to resonance. The over winding is connected between a radio antenna
and ground for one such application. The secondary is tuned to resonance with a variable
capacitor. The output may be taken from the tap point for amplification or detection. Small
millimeter size air core transformers are used in radio receivers. The largest radio transmitters
may use meter sized coils. Unshielded air core solenoid transformers are mounted at right
angles to each other to prevent stray coupling.
Stray coupling is minimized when the transformer is wound on a toroid form. (Figure
(b) 9.51) Toroidal air core transformers also show a higher degree of coupling, particularly
for bifilar windings. Bifilar windings are wound from a slightly twisted pair of wires. This
implies a 1:1 turns ratio. Three or four wires may be grouped for 1:2 and other integral ratios.
Windings do not have to be bifilar. This allows arbitrary turns ratios. However, the degree of
coupling suffers. Toroidal air core transformers are rare except for VHF (Very High Frequency)
work. Core materials other than air such as powdered iron or ferrite are preferred for lower
radio frequencies.
9.7.5 Tesla Coil
One notable example of an aircore transformer is the Tesla Coil, named after the Serbian
electrical genius Nikola Tesla, who was also the inventor of the rotating magnetic field AC
motor, polyphase AC power systems, and many elements of radio technology. The Tesla Coil
is a resonant, highfrequency stepup transformer used to produce extremely high voltages.
One of Tesla’s dreams was to employ his coil technology to distribute electric power without
the need for wires, simply broadcasting it in the form of radio waves which could be received
and conducted to loads by means of antennas. The basic schematic for a Tesla Coil is shown in
Figure 9.52.
9.7. SPECIAL TRANSFORMERS AND APPLICATIONS 261
discharge terminal
"Tesla Coil"
Figure 9.52: Tesla Coil: A few heavy primary turns, many secondary turns.
The capacitor, in conjunction with the transformer’s primary winding, forms a tank circuit.
The secondary winding is wound in close proximity to the primary, usually around the same
nonmagnetic form. Several options exist for “exciting” the primary circuit, the simplest being
a highvoltage, lowfrequency AC source and spark gap: (Figure 9.53)
spark gaphigh voltage
low frequency
HIGH voltage!
HIGH frequency!
RFC
RFC
Figure 9.53: System level diagram of Tesla coil with spark gap drive.
The purpose of the highvoltage, lowfrequency AC power source is to “charge” the pri
mary tank circuit. When the spark gap fires, its low impedance acts to complete the capac
itor/primary coil tank circuit, allowing it to oscillate at its resonant frequency. The “RFC”
inductors are “Radio Frequency Chokes,” which act as high impedances to prevent the AC
source from interfering with the oscillating tank circuit.
The secondary side of the Tesla coil transformer is also a tank circuit, relying on the para
sitic (stray) capacitance existing between the discharge terminal and earth ground to comple
ment the secondary winding’s inductance. For optimum operation, this secondary tank circuit
is tuned to the same resonant frequency as the primary circuit, with energy exchanged not only
between capacitors and inductors during resonant oscillation, but also backandforth between
262 CHAPTER 9. TRANSFORMERS
primary and secondary windings. The visual results are spectacular: (Figure 9.54)
Figure 9.54: High voltage high frequency discharge from Tesla coil.
Tesla Coils find application primarily as novelty devices, showing up in high school science
fairs, basement workshops, and the occasional low budget sciencefiction movie.
It should be noted that Tesla coils can be extremely dangerous devices. Burns caused by
radiofrequency (“RF”) current, like all electrical burns, can be very deep, unlike skin burns
caused by contact with hot objects or flames. Although the highfrequency discharge of a Tesla
coil has the curious property of being beyond the “shock perception” frequency of the human
nervous system, this does not mean Tesla coils cannot hurt or even kill you! I strongly ad
vise seeking the assistance of an experienced Tesla coil experimenter if you would embark on
building one yourself.
9.7.6 Saturable reactors
So far, we’ve explored the transformer as a device for converting different levels of voltage,
current, and even impedance from one circuit to another. Now we’ll take a look at it as a
completely different kind of device: one that allows a small electrical signal to exert control over
a much larger quantity of electrical power. In this mode, a transformer acts as an amplifier.
9.7. SPECIAL TRANSFORMERS AND APPLICATIONS 263
The device I’m referring to is called a saturablecore reactor, or simply saturable reactor.
Actually, it is not really a transformer at all, but rather a special kind of inductor whose in
ductance can be varied by the application of a DC current through a second winding wound
around the same iron core. Like the ferroresonant transformer, the saturable reactor relies on
the principle of magnetic saturation. When a material such as iron is completely saturated
(that is, all its magnetic domains are lined up with the applied magnetizing force), additional
increases in current through the magnetizing winding will not result in further increases of
magnetic flux.
Now, inductance is the measure of how well an inductor opposes changes in current by
developing a voltage in an opposing direction. The ability of an inductor to generate this
opposing voltage is directly connected with the change in magnetic flux inside the inductor
resulting from the change in current, and the number of winding turns in the inductor. If an
inductor has a saturated core, no further magnetic flux will result from further increases in
current, and so there will be no voltage induced in opposition to the change in current. In
other words, an inductor loses its inductance (ability to oppose changes in current) when its
core becomes magnetically saturated.
If an inductor’s inductance changes, then its reactance (and impedance) to AC current
changes as well. In a circuit with a constant voltage source, this will result in a change in
current: (Figure 9.55)
load
L
I
Figure 9.55: If L changes in inductance, ZL will correspondingly change, thus changing the
circuit current.
A saturable reactor capitalizes on this effect by forcing the core into a state of saturation
with a strong magnetic field generated by current through another winding. The reactor’s
“power” winding is the one carrying the AC load current, and the “control” winding is one
carrying a DC current strong enough to drive the core into saturation: (Figure 9.56)
The strangelooking transformer symbol shown in the above schematic represents a saturable
core reactor, the upper winding being the DC control winding and the lower being the “power”
winding through which the controlled AC current goes. Increased DC control current produces
more magnetic flux in the reactor core, driving it closer to a condition of saturation, thus de
creasing the power winding’s inductance, decreasing its impedance, and increasing current to
the load. Thus, the DC control current is able to exert control over the AC current delivered to
the load.
The circuit shown would work, but it would not work very well. The first problem is the
natural transformer action of the saturable reactor: AC current through the power winding
will induce a voltage in the control winding, which may cause trouble for the DC power source.
264 CHAPTER 9. TRANSFORMERS
load
saturable reactor
I
Figure 9.56: DC, via the control winding, saturates the core. Thus, modulating the power
winding inductance, impedance, and current.
Also, saturable reactors tend to regulate AC power only in one direction: in one half of the AC
cycle, the mmf’s from both windings add; in the other half, they subtract. Thus, the core will
have more flux in it during one half of the AC cycle than the other, and will saturate first in
that cycle half, passing load current more easily in one direction than the other. Fortunately,
both problems can be overcome with a little ingenuity: (Figure 9.57)
load
I
Figure 9.57: Out of phase DC control windings allow symmetrical of control AC.
Notice the placement of the phasing dots on the two reactors: the power windings are “in
phase” while the control windings are “out of phase.” If both reactors are identical, any volt
age induced in the control windings by load current through the power windings will cancel
out to zero at the battery terminals, thus eliminating the first problem mentioned. Further
more, since the DC control current through both reactors produces magnetic fluxes in different
directions through the reactor cores, one reactor will saturate more in one cycle of the AC
9.7. SPECIAL TRANSFORMERS AND APPLICATIONS 265
power while the other reactor will saturate more in the other, thus equalizing the control ac
tion through each halfcycle so that the AC power is “throttled” symmetrically. This phasing
of control windings can be accomplished with two separate reactors as shown, or in a single
reactor design with intelligent layout of the windings and core.
Saturable reactor technology has even been miniaturized to the circuitboard level in com
pact packages more generally known as magnetic amplifiers. I personally find this to be fasci
nating: the effect of amplification (one electrical signal controlling another), normally requiring
the use of physically fragile vacuum tubes or electrically “fragile” semiconductor devices, can
be realized in a device both physically and electrically rugged. Magnetic amplifiers do have
disadvantages over their more fragile counterparts, namely size, weight, nonlinearity, and
bandwidth (frequency response), but their utter simplicity still commands a certain degree of
appreciation, if not practical application.
Saturablecore reactors are less commonly known as “saturablecore inductors” or trans
ductors.
9.7.7 ScottT transformer
Nikola Tesla’s original polyphase power system was based on simple to build 2phase com
ponents. However, as transmission distances increased, the more transmission line efficient
3phase system became more prominent. Both 2φ and 3φ components coexisted for a number
of years. The ScottT transformer connection allowed 2φ and 3φ components like motors and
alternators to be interconnected. Yamamoto and Yamaguchi:
In 1896, General Electric built a 35.5 km (22 mi) threephase transmission line
operated at 11 kV to transmit power to Buffalo, New York, from the Niagara Falls
Project. The twophase generated power was changed to threephase by the use of
ScottT transformations. [1]
Y3Y2
Y1 R3
R2 R1
R4
86.6%
tap
2phase1 = V∠0°
T1
T2
3phase23 = V∠0°
3phase31 = V∠120°
3phase12 = V∠240°
2phase2 = V∠90°
50% tap
Figure 9.58: ScottT transformer converts 2φ to 3φ, or vice versa.
266 CHAPTER 9. TRANSFORMERS
The ScottT transformer set, Figure 9.58, consists of a center tapped transformer T1 and an
86.6% tapped transformer T2 on the 3φ side of the circuit. The primaries of both transformers
are connected to the 2φ voltages. One end of the T2 86.6% secondary winding is a 3φ output,
the other end is connected to the T1 secondary center tap. Both ends of the T1 secondary are
the other two 3φ connections.
Application of 2φ Niagara generator power produced a 3φ output for the more efficient 3φ
transmission line. More common these days is the application of 3φ power to produce a 2φ
output for driving an old 2φ motor.
In Figure 9.59, we use vectors in both polar and complex notation to prove that the ScottT
converts a pair of 2φ voltages to 3φ. First, one of the 3φ voltages is identical to a 2φ voltage
due to the 1:1 transformer T1 ratio, VP12= V2P1. The T1 center tapped secondary produces
opposite polarities of 0.5V2P1 on the secondary ends. This 6 0o is vectorially subtracted from T2
secondary voltage due to the KVL equations V31, V23. The T2 secondary voltage is 0.866V2P2
due to the 86.6% tap. Keep in mind that this 2nd phase of the 2φ is 6 90o. This 0.866V2P2 is
added at V31, subtracted at V23 in the KVL equations.
Y1Y2
Y3
R3
R2 R1
R4
86.6%
T1
T2
Given two 90° phased voltages:
V2P2 =Vsin(θ+90°)=Vcos(θ)=V∠90°=V(0+j1)
V2P1 =Vsin(θ+0°)=V∠0°=V(1+j0)
V2P2
V2P1
V12=V2P1 =Vsin(θ+0°)=V∠0°=V(1+j0)
Derive the three phase voltages V12 , V23 , V31 :
 +



+
+
+
+
+
−
+
−
+−
V23 =(0.5)V∠0°0.866V∠90°=V(0.5(1+j0)0.866(0+j1))=V(0.5+j0.866)=V∠−120°=V∠240°
V31 = (0.5)V∠0°+0.866V∠90°=V(0.5(1+j0)+0.866(0+j1))=V(0.5+j0.866)=V∠120°
V12 = V2P1 = V∠0°
+
−
+−
V12
V31
V23
A
B
C
D
(3) KVL: V23 = VDB  VBA = 0
(2) KVL: V31 VCB +VBD = 0
(1) KVL: V12 +VAC = 0
(3) KVL: V23 = VDB  VBA
(2) KVL: V31 = VCB +VBD
(1) KVL: V12 = VAC
VDB = 0.866V2P2 = 0.866V∠90° = 0.866V(0+j1)
VCB = VBA = 0.5V2P1 = 0.5V∠0° = 0.5V(1+j0)
50%
Figure 9.59: ScottT transformer 2φ to 3φ conversion equations.
We show “DC” polarities all over this AC only circuit, to keep track of the Kirchhoff voltage
loop polarities. Subtracting 6 0o is equivalent to adding 6 180o. The bottom line is when we add
86.6% of 6 90o to 50% of 6 180o we get 6 120o. Subtracting 86.6% of 6 90o from 50% of 6 180o
yields 6 120o or 6 240o.
In Figure 9.60 we graphically show the 2φ vectors at (a). At (b) the vectors are scaled by
transformers T1 and T2 to 0.5 and 0.866 respectively. At (c) 16 120o = 0.56 0o + 0.8666 90o, and
16 240o = 0.56 0o  0.8666 90o. The three output phases are 16 120o and 16 240o from (c), along
with input 16 0o (a).
9.7. SPECIAL TRANSFORMERS AND APPLICATIONS 267
0.866V∠90°
0.5∠0°
0.866V∠90°
1∠90°
1∠0°
1∠120°
1∠240°
1∠0°, 1∠90° yields 1∠−120° ,1∠240°
a b c
0.5∠0°
Figure 9.60: Graphical explanation of equations in Figure 9.59.
9.7.8 Linear Variable Differential Transformer
A linear variable differential transformer (LVDT) has an AC driven primary wound between
two secondaries on a cylindrical air core form. (Figure 9.61) A movable ferromagnetic slug con
verts displacement to a variable voltage by changing the coupling between the driven primary
and secondary windings. The LVDT is a displacement or distance measuring transducer. Units
are available for measuring displacement over a distance of a fraction of a millimeter to a half
a meter. LVDT’s are rugged and dirt resistant compared to linear optical encoders.
V1
V2
V3
V1
V13
V3
up center down
Figure 9.61: LVDT: linear variable differential transformer.
The excitation voltage is in the range of 0.5 to 10 VAC at a frequency of 1 to 200 Khz. A
ferrite core is suitable at these frequencies. It is extended outside the body by an nonmagnetic
rod. As the core is moved toward the top winding, the voltage across this coil increases due to
increased coupling, while the voltage on the bottom coil decreases. If the core is moved toward
the bottom winding, the voltage on this coil increases as the voltage decreases across the top
coil. Theoretically, a centered slug yields equal voltages across both coils. In practice leakage
inductance prevents the null from dropping all the way to 0 V.
With a centered slug, the seriesopposing wired secondaries cancel yielding V13 = 0. Moving
the slug up increases V13. Note that it is inphase with with V1, the top winding, and 180o out
of phase with V3, bottom winding.
Moving the slug down from the center position increases V13. However, it is 180o out of
phase with with V1, the top winding, and inphase with V3, bottom winding. Moving the slug
from top to bottom shows a minimum at the center point, with a 180o phase reversal in passing
the center.
268 CHAPTER 9. TRANSFORMERS
• REVIEW:
• Transformers can be used to transform impedance as well as voltage and current. When
this is done to improve power transfer to a load, it is called impedance matching.
• A Potential Transformer (PT) is a special instrument transformer designed to provide a
precise voltage stepdown ratio for voltmeters measuring high power system voltages.
• A Current Transformer (CT) is another special instrument transformer designed to step
down the current through a power line to a safe level for an ammeter to measure.
• An aircore transformer is one lacking a ferromagnetic core.
• A Tesla Coil is a resonant, aircore, stepup transformer designed to produce very high
AC voltages at high frequency.
• A saturable reactor is a special type of inductor, the inductance of which can be controlled
by the DC current through a second winding around the same core. With enough DC cur
rent, the magnetic core can be saturated, decreasing the inductance of the power winding
in a controlled fashion.
• A ScottT transformer converts 3φ power to 2φ power and vice versa.
• A linear variable differential transformer, also known as an LVDT, is a distance measur
ing device. It has a movable ferromagnetic core to vary the coupling between the excited
primary and a pair of secondaries.
9.8 Practical considerations
9.8.1 Power capacity
As has already been observed, transformers must be well designed in order to achieve ac
ceptable power coupling, tight voltage regulation, and low exciting current distortion. Also,
transformers must be designed to carry the expected values of primary and secondary winding
current without any trouble. This means the winding conductors must be made of the proper
gauge wire to avoid any heating problems. An ideal transformer would have perfect coupling
(no leakage inductance), perfect voltage regulation, perfectly sinusoidal exciting current, no
hysteresis or eddy current losses, and wire thick enough to handle any amount of current. Un
fortunately, the ideal transformer would have to be infinitely large and heavy to meet these
design goals. Thus, in the business of practical transformer design, compromises must be
made.
Additionally, winding conductor insulation is a concern where high voltages are encoun
tered, as they often are in stepup and stepdown power distribution transformers. Not only
do the windings have to be well insulated from the iron core, but each winding has to be suffi
ciently insulated from the other in order to maintain electrical isolation between windings.
Respecting these limitations, transformers are rated for certain levels of primary and sec
ondary winding voltage and current, though the current rating is usually derived from a volt
amp (VA) rating assigned to the transformer. For example, take a stepdown transformer with
9.8. PRACTICAL CONSIDERATIONS 269
a primary voltage rating of 120 volts, a secondary voltage rating of 48 volts, and a VA rating of
1 kVA (1000 VA). The maximum winding currents can be determined as such:
1000 VA
120 V
= 8.333 A (maximum primary winding current)
1000 VA
48 V
= 20.833 A (maximum secondary winding current)
Sometimes windings will bear current ratings in amps, but this is typically seen on small
transformers. Large transformers are almost always rated in terms of winding voltage and VA
or kVA.
9.8.2 Energy losses
When transformers transfer power, they do so with a minimum of loss. As it was stated earlier,
modern power transformer designs typically exceed 95% efficiency. It is good to know where
some of this lost power goes, however, and what causes it to be lost.
There is, of course, power lost due to resistance of the wire windings. Unless supercon
ducting wires are used, there will always be power dissipated in the form of heat through the
resistance of currentcarrying conductors. Because transformers require such long lengths of
wire, this loss can be a significant factor. Increasing the gauge of the winding wire is one way
to minimize this loss, but only with substantial increases in cost, size, and weight.
Resistive losses aside, the bulk of transformer power loss is due to magnetic effects in the
core. Perhaps the most significant of these “core losses” is eddycurrent loss, which is resistive
power dissipation due to the passage of induced currents through the iron of the core. Because
iron is a conductor of electricity as well as being an excellent “conductor” of magnetic flux,
there will be currents induced in the iron just as there are currents induced in the secondary
windings from the alternating magnetic field. These induced currents – as described by the
perpendicularity clause of Faraday’s Law – tend to circulate through the crosssection of the
core perpendicularly to the primary winding turns. Their circular motion gives them their
unusual name: like eddies in a stream of water that circulate rather than move in straight
lines.
Iron is a fair conductor of electricity, but not as good as the copper or aluminum from which
wire windings are typically made. Consequently, these “eddy currents” must overcome sig
nificant electrical resistance as they circulate through the core. In overcoming the resistance
offered by the iron, they dissipate power in the form of heat. Hence, we have a source of
inefficiency in the transformer that is difficult to eliminate.
This phenomenon is so pronounced that it is often exploited as a means of heating ferrous
(ironcontaining) materials. The photograph of (Figure 9.62) shows an “induction heating” unit
raising the temperature of a large pipe section. Loops of wire covered by hightemperature
insulation encircle the pipe’s circumference, inducing eddy currents within the pipe wall by
electromagnetic induction. In order to maximize the eddy current effect, highfrequency alter
nating current is used rather than power line frequency (60 Hz). The box units at the right of
the picture produce the highfrequency AC and control the amount of current in the wires to
stabilize the pipe temperature at a predetermined “setpoint.”
270 CHAPTER 9. TRANSFORMERS
Figure 9.62: Induction heating: Primary insulated winding induces current into lossy iron pipe
(secondary).
The main strategy in mitigating these wasteful eddy currents in transformer cores is to
form the iron core in sheets, each sheet covered with an insulating varnish so that the core
is divided up into thin slices. The result is very little width in the core for eddy currents to
circulate in: (Figure 9.63)
solid iron core
laminated iron core
"eddy"
current
Figure 9.63: Dividing the iron core into thin insulated laminations minimizes eddy current
loss.
Laminated cores like the one shown here are standard in almost all lowfrequency trans
formers. Recall from the photograph of the transformer cut in half that the iron core was
composed of many thin sheets rather than one solid piece. Eddy current losses increase with
frequency, so transformers designed to run on higherfrequency power (such as 400 Hz, used in
many military and aircraft applications) must use thinner laminations to keep the losses down
to a respectable minimum. This has the undesirable effect of increasing the manufacturing
cost of the transformer.
9.8. PRACTICAL CONSIDERATIONS 271
Another, similar technique for minimizing eddy current losses which works better for high
frequency applications is to make the core out of iron powder instead of thin iron sheets. Like
the lamination sheets, these granules of iron are individually coated in an electrically insulat
ing material, which makes the core nonconductive except for within the width of each granule.
Powdered iron cores are often found in transformers handling radiofrequency currents.
Another “core loss” is that of magnetic hysteresis. All ferromagnetic materials tend to re
tain some degree of magnetization after exposure to an external magnetic field. This tendency
to stay magnetized is called “hysteresis,” and it takes a certain investment in energy to over
come this opposition to change every time the magnetic field produced by the primary winding
changes polarity (twice per AC cycle). This type of loss can be mitigated through good core
material selection (choosing a core alloy with low hysteresis, as evidenced by a “thin” B/H hys
teresis curve), and designing the core for minimum flux density (large crosssectional area).
Transformer energy losses tend to worsen with increasing frequency. The skin effect within
winding conductors reduces the available crosssectional area for electron flow, thereby increas
ing effective resistance as the frequency goes up and creating more power lost through resistive
dissipation. Magnetic core losses are also exaggerated with higher frequencies, eddy currents
and hysteresis effects becoming more severe. For this reason, transformers of significant size
are designed to operate efficiently in a limited range of frequencies. In most power distribution
systems where the line frequency is very stable, one would think excessive frequency would
never pose a problem. Unfortunately it does, in the form of harmonics created by nonlinear
loads.
As we’ve seen in earlier chapters, nonsinusoidal waveforms are equivalent to additive series
of multiple sinusoidal waveforms at different amplitudes and frequencies. In power systems,
these other frequencies are wholenumber multiples of the fundamental (line) frequency, mean
ing that they will always be higher, not lower, than the design frequency of the transformer.
In significant measure, they can cause severe transformer overheating. Power transformers
can be engineered to handle certain levels of power system harmonics, and this capability is
sometimes denoted with a “K factor” rating.
9.8.3 Stray capacitance and inductance
Aside from power ratings and power losses, transformers often harbor other undesirable lim
itations which circuit designers must be made aware of. Like their simpler counterparts – in
ductors – transformers exhibit capacitance due to the insulation dielectric between conductors:
from winding to winding, turn to turn (in a single winding), and winding to core. Usually this
capacitance is of no concern in a power application, but small signal applications (especially
those of high frequency) may not tolerate this quirk well. Also, the effect of having capacitance
along with the windings’ designed inductance gives transformers the ability to resonate at a
particular frequency, definitely a design concern in signal applications where the applied fre
quency may reach this point (usually the resonant frequency of a power transformer is well
beyond the frequency of the AC power it was designed to operate on).
Flux containment (making sure a transformer’s magnetic flux doesn’t escape so as to inter
fere with another device, and making sure other devices’ magnetic flux is shielded from the
transformer core) is another concern shared both by inductors and transformers.
Closely related to the issue of flux containment is leakage inductance. We’ve already seen
the detrimental effects of leakage inductance on voltage regulation with SPICE simulations
272 CHAPTER 9. TRANSFORMERS
early in this chapter. Because leakage inductance is equivalent to an inductance connected in
series with the transformer’s winding, it manifests itself as a series impedance with the load.
Thus, the more current drawn by the load, the less voltage available at the secondary winding
terminals. Usually, good voltage regulation is desired in transformer design, but there are
exceptional applications. As was stated before, discharge lighting circuits require a stepup
transformer with “loose” (poor) voltage regulation to ensure reduced voltage after the estab
lishment of an arc through the lamp. One way to meet this design criterion is to engineer the
transformer with flux leakage paths for magnetic flux to bypass the secondary winding(s). The
resulting leakage flux will produce leakage inductance, which will in turn produce the poor
regulation needed for discharge lighting.
9.8.4 Core saturation
Transformers are also constrained in their performance by the magnetic flux limitations of the
core. For ferromagnetic core transformers, we must be mindful of the saturation limits of the
core. Remember that ferromagnetic materials cannot support infinite magnetic flux densities:
they tend to “saturate” at a certain level (dictated by the material and core dimensions), mean
ing that further increases in magnetic field force (mmf) do not result in proportional increases
in magnetic field flux (Φ).
When a transformer’s primary winding is overloaded from excessive applied voltage, the
core flux may reach saturation levels during peak moments of the AC sinewave cycle. If this
happens, the voltage induced in the secondary winding will no longer match the waveshape
as the voltage powering the primary coil. In other words, the overloaded transformer will dis
tort the waveshape from primary to secondary windings, creating harmonics in the secondary
winding’s output. As we discussed before, harmonic content in AC power systems typically
causes problems.
Special transformers known as peaking transformers exploit this principle to produce brief
voltage pulses near the peaks of the source voltage waveform. The core is designed to saturate
quickly and sharply, at voltage levels well below peak. This results in a severely cropped
sinewave flux waveform, and secondary voltage pulses only when the flux is changing (below
saturation levels): (Figure 9.64)
Φ = magnetic fluxep = primary voltage es = secondary voltage
ep
es
Φ
Figure 9.64: Voltage and flux waveforms for a peaking transformer.
9.8. PRACTICAL CONSIDERATIONS 273
Another cause of abnormal transformer core saturation is operation at frequencies lower
than normal. For example, if a power transformer designed to operate at 60 Hz is forced
to operate at 50 Hz instead, the flux must reach greater peak levels than before in order to
produce the same opposing voltage needed to balance against the source voltage. This is true
even if the source voltage is the same as before. (Figure 9.65)
e = voltage
Φ = magnetic flux
e Φ
e
Φ
60 Hz
50 Hz
Figure 9.65: Magnetic flux is higher in a transformer core driven by 50 Hz as compared to 60
Hz for the same voltage.
Since instantaneous winding voltage is proportional to the instantaneous magnetic flux’s
rate of change in a transformer, a voltage waveform reaching the same peak value, but taking
a longer amount of time to complete each halfcycle, demands that the flux maintain the same
rate of change as before, but for longer periods of time. Thus, if the flux has to climb at the same
rate as before, but for longer periods of time, it will climb to a greater peak value. (Figure 9.66)
Mathematically, this is another example of calculus in action. Because the voltage is pro
portional to the flux’s rateofchange, we say that the voltage waveform is the derivative of
the flux waveform, “derivative” being that calculus operation defining one mathematical func
tion (waveform) in terms of the rateofchange of another. If we take the opposite perspective,
though, and relate the original waveform to its derivative, we may call the original waveform
the integral of the derivative waveform. In this case, the voltage waveform is the derivative of
the flux waveform, and the flux waveform is the integral of the voltage waveform.
The integral of any mathematical function is proportional to the area accumulated under
neath the curve of that function. Since each halfcycle of the 50 Hz waveform accumulates
more area between it and the zero line of the graph than the 60 Hz waveform will – and we
know that the magnetic flux is the integral of the voltage – the flux will attain higher values
in Figure 9.66.
Yet another cause of transformer saturation is the presence of DC current in the primary
winding. Any amount of DC voltage dropped across the primary winding of a transformer will
274 CHAPTER 9. TRANSFORMERS
e
Φ
e
Φ
60 Hz
50 Hz
less area
more area
less height
more height
Figure 9.66: Flux changing at the same rate rises to a higher level at 50 Hz than at 60 Hz.
cause additional magnetic flux in the core. This additional flux “bias” or “offset” will push the
alternating flux waveform closer to saturation in one halfcycle than the other. (Figure 9.67)
e Φ
60 Hz
saturation limit
saturation limit
flux
centerline
Figure 9.67: DC in primary, shifts the waveform peaks toward the upper saturation limit.
For most transformers, core saturation is a very undesirable effect, and it is avoided through
good design: engineering the windings and core so that magnetic flux densities remain well be
low the saturation levels. This ensures that the relationship between mmf and Φ is more linear
throughout the flux cycle, which is good because it makes for less distortion in the magnetiza
tion current waveform. Also, engineering the core for low flux densities provides a safe margin
between the normal flux peaks and the core saturation limits to accommodate occasional, ab
normal conditions such as frequency variation and DC offset.
9.8. PRACTICAL CONSIDERATIONS 275
9.8.5 Inrush current
When a transformer is initially connected to a source of AC voltage, there may be a substan
tial surge of current through the primary winding called inrush current. (Figure 9.72) This is
analogous to the inrush current exhibited by an electric motor that is started up by sudden con
nection to a power source, although transformer inrush is caused by a different phenomenon.
We know that the rate of change of instantaneous flux in a transformer core is proportional
to the instantaneous voltage drop across the primary winding. Or, as stated before, the voltage
waveform is the derivative of the flux waveform, and the flux waveform is the integral of the
voltage waveform. In a continuouslyoperating transformer, these two waveforms are phase
shifted by 90o. (Figure 9.68) Since flux (Φ) is proportional to the magnetomotive force (mmf)
in the core, and the mmf is proportional to winding current, the current waveform will be
inphase with the flux waveform, and both will be lagging the voltage waveform by 90o:
e = voltage Φ = magnetic flux i = coil current
e Φ
i
Figure 9.68: Continuous steadystate operation: Magnetic flux, like current, lags applied volt
age by 90o.
Let us suppose that the primary winding of a transformer is suddenly connected to an AC
voltage source at the exact moment in time when the instantaneous voltage is at its positive
peak value. In order for the transformer to create an opposing voltage drop to balance against
this applied source voltage, a magnetic flux of rapidly increasing value must be generated.
The result is that winding current increases rapidly, but actually no more rapidly than under
normal conditions: (Figure 9.69)
Both core flux and coil current start from zero and build up to the same peak values expe
rienced during continuous operation. Thus, there is no “surge” or “inrush” or current in this
scenario. (Figure 9.69)
Alternatively, let us consider what happens if the transformer’s connection to the AC voltage
source occurs at the exact moment in time when the instantaneous voltage is at zero. During
continuous operation (when the transformer has been powered for quite some time), this is the
point in time where both flux and winding current are at their negative peaks, experiencing
zero rateofchange (dΦ/dt = 0 and di/dt = 0). As the voltage builds to its positive peak, the
flux and current waveforms build to their maximum positive ratesofchange, and on upward
to their positive peaks as the voltage descends to a level of zero:
A significant difference exists, however, between continuousmode operation and the sud
den starting condition assumed in this scenario: during continuous operation, the flux and
current levels were at their negative peaks when voltage was at its zero point; in a trans
former that has been sitting idle, however, both magnetic flux and winding current should
276 CHAPTER 9. TRANSFORMERS
e = voltage
Φ = magnetic flux
i = coil current
e Φ
i
Instant in time when transformer
is connected to AC voltage source.
Figure 9.69: Connecting transformer to line at AC volt peak: Flux increases rapidly from zero,
same as steadystate operation.
e = voltage
Φ = magnetic flux
i = coil current
e Φ
i
Instant in time when voltage is zero,
during continuous operation.
Figure 9.70: Starting at e=0 V is not the same as running continuously in Figure 9.3 These
expected waveforms are incorrect– Φ and i should start at zero.
9.8. PRACTICAL CONSIDERATIONS 277
start at zero. When the magnetic flux increases in response to a rising voltage, it will increase
from zero upward, not from a previously negative (magnetized) condition as we would normally
have in a transformer that’s been powered for awhile. Thus, in a transformer that’s just “start
ing,” the flux will reach approximately twice its normal peak magnitude as it “integrates” the
area under the voltage waveform’s first halfcycle: (Figure 9.71)
e
Φ
Instant in time when voltage is zero,
from a "cold start" condition.
flux peak approximately
twice normal height!
Figure 9.71: Starting at e=0 V, Φ starts at initial condition Φ=0, increasing to twice the normal
value, assuming it doesn’t saturate the core.
In an ideal transformer, the magnetizing current would rise to approximately twice its nor
mal peak value as well, generating the necessary mmf to create this higherthannormal flux.
However, most transformers aren’t designed with enough of a margin between normal flux
peaks and the saturation limits to avoid saturating in a condition like this, and so the core
will almost certainly saturate during this first halfcycle of voltage. During saturation, dispro
portionate amounts of mmf are needed to generate magnetic flux. This means that winding
current, which creates the mmf to cause flux in the core, will disproportionately rise to a value
easily exceeding twice its normal peak: (Figure 9.72)
This is the mechanism causing inrush current in a transformer’s primary winding when
connected to an AC voltage source. As you can see, the magnitude of the inrush current
strongly depends on the exact time that electrical connection to the source is made. If the
transformer happens to have some residual magnetism in its core at the moment of connection
to the source, the inrush could be even more severe. Because of this, transformer overcurrent
protection devices are usually of the “slowacting” variety, so as to tolerate current surges such
as this without opening the circuit.
9.8.6 Heat and Noise
In addition to unwanted electrical effects, transformers may also exhibit undesirable physical
effects, the most notable being the production of heat and noise. Noise is primarily a nuisance
effect, but heat is a potentially serious problem because winding insulation will be damaged if
allowed to overheat. Heating may be minimized by good design, ensuring that the core does
278 CHAPTER 9. TRANSFORMERS
e
Φ
Instant in time when voltage is zero,
from a "cold start" condition.
flux peak approximately
twice normal height!
i
current peak much
greater than normal!
Figure 9.72: Starting at e=0 V, Current also increases to twice the normal value for an unsat
urated core, or considerably higher in the (designed for) case of saturation.
not approach saturation levels, that eddy currents are minimized, and that the windings are
not overloaded or operated too close to maximum ampacity.
Large power transformers have their core and windings submerged in an oil bath to transfer
heat and muffle noise, and also to displace moisture which would otherwise compromise the
integrity of the winding insulation. Heatdissipating “radiator” tubes on the outside of the
transformer case provide a convective oil flow path to transfer heat from the transformer’s core
to ambient air: (Figure 9.73)
Oilless, or “dry,” transformers are often rated in terms of maximum operating temperature
“rise” (temperature increase beyond ambient) according to a letterclass system: A, B, F, or H.
These letter codes are arranged in order of lowest heat tolerance to highest:
• Class A: No more than 55o Celsius winding temperature rise, at 40o Celsius (maximum)
ambient air temperature.
• Class B: No more than 80o Celsius winding temperature rise, at 40o Celsius (maxi
mum)ambient air temperature.
• Class F: No more than 115o Celsius winding temperature rise, at 40o Celsius (maxi
mum)ambient air temperature.
• Class H: No more than 150o Celsius winding temperature rise, at 40o Celsius (maxi
mum)ambient air temperature.
Audible noise is an effect primarily originating from the phenomenon of magnetostriction:
the slight change of length exhibited by a ferromagnetic object when magnetized. The familiar
“hum” heard around large power transformers is the sound of the iron core expanding and
9.8. PRACTICAL CONSIDERATIONS 279
Core
Oil
Primary
terminals
Secondary
terminals
Radiator
tube
flow
Heat Heat
Radiator
tube
Figure 9.73: Large power transformers are submerged in heat dissipating insulating oil.
280 CHAPTER 9. TRANSFORMERS
contracting at 120 Hz (twice the system frequency, which is 60 Hz in the United States) –
one cycle of core contraction and expansion for every peak of the magnetic flux waveform –
plus noise created by mechanical forces between primary and secondary windings. Again,
maintaining low magnetic flux levels in the core is the key to minimizing this effect, which
explains why ferroresonant transformers – which must operate in saturation for a large portion
of the current waveform – operate both hot and noisy.
Another noiseproducing phenomenon in power transformers is the physical reaction force
between primary and secondary windings when heavily loaded. If the secondary winding is
opencircuited, there will be no current through it, and consequently no magnetomotive force
(mmf) produced by it. However, when the secondary is “loaded” (current supplied to a load), the
winding generates an mmf, which becomes counteracted by a “reflected” mmf in the primary
winding to prevent core flux levels from changing. These opposing mmf’s generated between
primary and secondary windings as a result of secondary (load) current produce a repulsive,
physical force between the windings which will tend to make them vibrate. Transformer de
signers have to consider these physical forces in the construction of the winding coils, to ensure
there is adequate mechanical support to handle the stresses. Under heavy load (high current)
conditions, though, these stresses may be great enough to cause audible noise to emanate from
the transformer.
• REVIEW:
• Power transformers are limited in the amount of power they can transfer from primary
to secondary winding(s). Large units are typically rated in VA (voltamps) or kVA (kilo
voltamps).
• Resistance in transformer windings contributes to inefficiency, as current will dissipate
heat, wasting energy.
• Magnetic effects in a transformer’s iron core also contribute to inefficiency. Among the
effects are eddy currents (circulating induction currents in the iron core) and hysteresis
(power lost due to overcoming the tendency of iron to magnetize in a particular direction).
• Increased frequency results in increased power losses within a power transformer. The
presence of harmonics in a power system is a source of frequencies significantly higher
than normal, which may cause overheating in large transformers.
• Both transformers and inductors harbor certain unavoidable amounts of capacitance due
to wire insulation (dielectric) separating winding turns from the iron core and from each
other. This capacitance can be significant enough to give the transformer a natural reso
nant frequency, which can be problematic in signal applications.
• Leakage inductance is caused by magnetic flux not being 100% coupled between windings
in a transformer. Any flux not involved with transferring energy from one winding to
another will store and release energy, which is how (self) inductance works. Leakage
inductance tends to worsen a transformer’s voltage regulation (secondary voltage “sags”
more for a given amount of load current).
• Magnetic saturation of a transformer core may be caused by excessive primary voltage,
operation at too low of a frequency, and/or by the presence of a DC current in any of
9.9. CONTRIBUTORS 281
the windings. Saturation may be minimized or avoided by conservative design, which
provides an adequate margin of safety between peak magnetic flux density values and
the saturation limits of the core.
• Transformers often experience significant inrush currents when initially connected to an
AC voltage source. Inrush current is most severe when connection to the AC source is
made at the moment instantaneous source voltage is zero.
• Noise is a common phenomenon exhibited by transformers – especially power transform
ers – and is primarily caused by magnetostriction of the core. Physical forces causing
winding vibration may also generate noise under conditions of heavy (high current) sec
ondary winding load.
9.9 Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
Bart Anderson (January 2004): Corrected conceptual errors regarding Tesla coil operation
and safety.
Jason Starck (June 2000): HTML document formatting, which led to a much better
looking second edition.
Bibliography
[1] Mitsuyoshi Yamamoto, Mitsugi Yamaguchi, “Electric Power In
Japan, Rapid Electrification a Century Ago”, EDN, (4/11/2002).
http://www.ieee.org/organizations/pes/public/2005/mar/peshistory.html
282 CHAPTER 9. TRANSFORMERS
Chapter 10
POLYPHASE AC CIRCUITS
Contents
10.1 Singlephase power systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
10.2 Threephase power systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
10.3 Phase rotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
10.4 Polyphase motor design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
10.5 Threephase Y and Delta configurations . . . . . . . . . . . . . . . . . . . . 306
10.6 Threephase transformer circuits . . . . . . . . . . . . . . . . . . . . . . . . 313
10.7 Harmonics in polyphase power systems . . . . . . . . . . . . . . . . . . . . 318
10.8 Harmonic phase sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
10.9 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345
10.1 Singlephase power systems
load load
#1 #2
Figure 10.1: Single phase power system schematic diagram shows little about the wiring of a
practical power circuit.
Depicted above (Figure 10.1) is a very simple AC circuit. If the load resistor’s power dis
sipation were substantial, we might call this a “power circuit” or “power system” instead of
regarding it as just a regular circuit. The distinction between a “power circuit” and a “regular
circuit” may seem arbitrary, but the practical concerns are definitely not.
283
284 CHAPTER 10. POLYPHASE AC CIRCUITS
One such concern is the size and cost of wiring necessary to deliver power from the AC
source to the load. Normally, we do not give much thought to this type of concern if we’re
merely analyzing a circuit for the sake of learning about the laws of electricity. However, in the
real world it can be a major concern. If we give the source in the above circuit a voltage value
and also give power dissipation values to the two load resistors, we can determine the wiring
needs for this particular circuit: (Figure 10.2)
load load
#1 #2
P = 10 kW P = 10 kW
120 V
Figure 10.2: As a practical matter, the wiring for the 20 kW loads at 120 Vac is rather substan
tial (167 A).
I = P
E
I = 10 kW
120 V
(for each load resistor)
Itotal = Iload#1 + Iload#2
Itotal = 166.67 A
83.33 AI =
Itotal = (83.33 A) + (83.33 A)
Ptotal = (10 kW) + (10 kW)
Ptotal = 20 kW
83.33 amps for each load resistor in Figure 10.2 adds up to 166.66 amps total circuit current.
This is no small amount of current, and would necessitate copper wire conductors of at least
1/0 gage. Such wire is well over 1/4 inch (6 mm) in diameter, weighing over 300 pounds per
thousand feet. Bear in mind that copper is not cheap either! It would be in our best interest
to find ways to minimize such costs if we were designing a power system with long conductor
lengths.
One way to do this would be to increase the voltage of the power source and use loads built
to dissipate 10 kW each at this higher voltage. The loads, of course, would have to have greater
resistance values to dissipate the same power as before (10 kW each) at a greater voltage than
before. The advantage would be less current required, permitting the use of smaller, lighter,
and cheaper wire: (Figure 10.3)
10.1. SINGLEPHASE POWER SYSTEMS 285
load load
#1 #2240 V
P = 10 kW P = 10 kW
Figure 10.3: Same 10 kW loads at 240 Vac requires less substantial wiring than at 120 Vac (83
A).
I = P
E
I = 10 kW
(for each load resistor)
Itotal = Iload#1 + Iload#2
I =
Ptotal = (10 kW) + (10 kW)
Ptotal = 20 kW
240 V
41.67 A
Itotal = (41.67 A) + (41.67 A)
Itotal = 83.33 A
Now our total circuit current is 83.33 amps, half of what it was before. We can now use
number 4 gage wire, which weighs less than half of what 1/0 gage wire does per unit length.
This is a considerable reduction in system cost with no degradation in performance. This is why
power distribution system designers elect to transmit electric power using very high voltages
(many thousands of volts): to capitalize on the savings realized by the use of smaller, lighter,
cheaper wire.
However, this solution is not without disadvantages. Another practical concern with power
circuits is the danger of electric shock from high voltages. Again, this is not usually the sort
of thing we concentrate on while learning about the laws of electricity, but it is a very valid
concern in the real world, especially when large amounts of power are being dealt with. The
gain in efficiency realized by stepping up the circuit voltage presents us with increased danger
of electric shock. Power distribution companies tackle this problem by stringing their power
lines along high poles or towers, and insulating the lines from the supporting structures with
large, porcelain insulators.
At the point of use (the electric power customer), there is still the issue of what voltage
to use for powering loads. High voltage gives greater system efficiency by means of reduced
conductor current, but it might not always be practical to keep power wiring out of reach at
the point of use the way it can be elevated out of reach in distribution systems. This tradeoff
between efficiency and danger is one that European power system designers have decided to
286 CHAPTER 10. POLYPHASE AC CIRCUITS
risk, all their households and appliances operating at a nominal voltage of 240 volts instead of
120 volts as it is in North America. That is why tourists from America visiting Europe must
carry small stepdown transformers for their portable appliances, to step the 240 VAC (volts
AC) power down to a more suitable 120 VAC.
Is there any way to realize the advantages of both increased efficiency and reduced safety
hazard at the same time? One solution would be to install stepdown transformers at the end
point of power use, just as the American tourist must do while in Europe. However, this would
be expensive and inconvenient for anything but very small loads (where the transformers can
be built cheaply) or very large loads (where the expense of thick copper wires would exceed the
expense of a transformer).
An alternative solution would be to use a higher voltage supply to provide power to two
lower voltage loads in series. This approach combines the efficiency of a highvoltage system
with the safety of a lowvoltage system: (Figure 10.4)
load
load
#1
#2
+

240 V
83.33 A
83.33 A
240 V
10 kW
10 kW
120 V
120 V
+

+

+

Figure 10.4: Series connected 120 Vac loads, driven by 240 Vac source at 83.3 A total current.
Notice the polarity markings (+ and ) for each voltage shown, as well as the unidirectional
arrows for current. For the most part, I’ve avoided labeling “polarities” in the AC circuits
we’ve been analyzing, even though the notation is valid to provide a frame of reference for
phase. In later sections of this chapter, phase relationships will become very important, so I’m
introducing this notation early on in the chapter for your familiarity.
The current through each load is the same as it was in the simple 120 volt circuit, but the
currents are not additive because the loads are in series rather than parallel. The voltage
across each load is only 120 volts, not 240, so the safety factor is better. Mind you, we still have
a full 240 volts across the power system wires, but each load is operating at a reduced voltage.
If anyone is going to get shocked, the odds are that it will be from coming into contact with
the conductors of a particular load rather than from contact across the main wires of a power
system.
There’s only one disadvantage to this design: the consequences of one load failing open, or
being turned off (assuming each load has a series on/off switch to interrupt current) are not
good. Being a series circuit, if either load were to open, current would stop in the other load as
well. For this reason, we need to modify the design a bit: (Figure 10.5)
10.1. SINGLEPHASE POWER SYSTEMS 287
load
load
#1
#2
+

+

"hot"
"hot"
"neutral"
120 V
∠ 0o
120 V
∠ 0o
83.33 A
83.33 A
+

+

240 V
+

120 V
∠ 0o
120 V
∠ 0o
0 A
Figure 10.5: Addition of neutral conductor allows loads to be individually driven.
Etotal = (120 V ∠ 0o) + (120 V ∠ 0o)
Etotal = 240 V ∠ 0o
I = P
E
I = 10 kW
120 V
I = 83.33 A (for each load resistor)
Ptotal = (10 kW) + (10 kW)
Ptotal = 20 kW
Instead of a single 240 volt power supply, we use two 120 volt supplies (in phase with
each other!) in series to produce 240 volts, then run a third wire to the connection point
between the loads to handle the eventuality of one load opening. This is called a splitphase
power system. Three smaller wires are still cheaper than the two wires needed with the simple
parallel design, so we’re still ahead on efficiency. The astute observer will note that the neutral
wire only has to carry the difference of current between the two loads back to the source. In
the above case, with perfectly “balanced” loads consuming equal amounts of power, the neutral
wire carries zero current.
Notice how the neutral wire is connected to earth ground at the power supply end. This is a
common feature in power systems containing “neutral” wires, since grounding the neutral wire
ensures the least possible voltage at any given time between any “hot” wire and earth ground.
An essential component to a splitphase power system is the dual AC voltage source. Fortu
nately, designing and building one is not difficult. Since most AC systems receive their power
from a stepdown transformer anyway (stepping voltage down from high distribution levels
to a userlevel voltage like 120 or 240), that transformer can be built with a centertapped
secondary winding: (Figure 10.6)
If the AC power comes directly from a generator (alternator), the coils can be similarly
centertapped for the same effect. The extra expense to include a centertap connection in a
288 CHAPTER 10. POLYPHASE AC CIRCUITS
Stepdown transformer with
centertapped secondary winding
2.4 kV
120 V
120 V
240 V
+

+

+

+

Figure 10.6: American 120/240 Vac power is derived from a center tapped utility transformer.
transformer or alternator winding is minimal.
Here is where the (+) and () polarity markings really become important. This notation is
often used to reference the phasings of multiple AC voltage sources, so it is clear whether they
are aiding (“boosting”) each other or opposing (“bucking”) each other. If not for these polarity
markings, phase relations between multiple AC sources might be very confusing. Note that
the splitphase sources in the schematic (each one 120 volts 6 0o), with polarity marks (+) to
() just like seriesaiding batteries can alternatively be represented as such: (Figure 10.7)
+

+

+

"hot"
"hot"
120 V
∠ 0o
120 V
∠ 180o
240 V
∠ 0o
Figure 10.7: Split phase 120/240 Vac source is equivalent to two series aiding 120 Vac sources.
To mathematically calculate voltage between “hot” wires, we must subtract voltages, be
cause their polarity marks show them to be opposed to each other:
120 + j0 V
240 + j0 V
120 ∠ 0o
120 ∠ 180o
240 ∠ 0o
Polar Rectangular
  (120 + j0) V
If we mark the two sources’ common connection point (the neutral wire) with the same
polarity mark (), we must express their relative phase shifts as being 180o apart. Otherwise,
we’d be denoting two voltage sources in direct opposition with each other, which would give
0 volts between the two “hot” conductors. Why am I taking the time to elaborate on polarity
marks and phase angles? It will make more sense in the next section!
10.2. THREEPHASE POWER SYSTEMS 289
Power systems in American households and light industry are most often of the splitphase
variety, providing socalled 120/240 VAC power. The term “splitphase” merely refers to the
splitvoltage supply in such a system. In a more general sense, this kind of AC power supply
is called single phase because both voltage waveforms are in phase, or in step, with each other.
The term “single phase” is a counterpoint to another kind of power system called “polyphase”
which we are about to investigate in detail. Apologies for the long introduction leading up to
the titletopic of this chapter. The advantages of polyphase power systems are more obvious if
one first has a good understanding of single phase systems.
• REVIEW:
• Single phase power systems are defined by having an AC source with only one voltage
waveform.
• A splitphase power system is one with multiple (inphase) AC voltage sources connected
in series, delivering power to loads at more than one voltage, with more than two wires.
They are used primarily to achieve balance between system efficiency (low conductor
currents) and safety (low load voltages).
• Splitphase AC sources can be easily created by centertapping the coil windings of trans
formers or alternators.
10.2 Threephase power systems
Splitphase power systems achieve their high conductor efficiency and low safety risk by split
ting up the total voltage into lesser parts and powering multiple loads at those lesser voltages,
while drawing currents at levels typical of a fullvoltage system. This technique, by the way,
works just as well for DC power systems as it does for singlephase AC systems. Such sys
tems are usually referred to as threewire systems rather than splitphase because “phase” is a
concept restricted to AC.
But we know from our experience with vectors and complex numbers that AC voltages don’t
always add up as we think they would if they are out of phase with each other. This principle,
applied to power systems, can be put to use to make power systems with even greater conductor
efficiencies and lower shock hazard than with splitphase.
Suppose that we had two sources of AC voltage connected in series just like the splitphase
system we saw before, except that each voltage source was 120o out of phase with the other:
(Figure 10.8)
Since each voltage source is 120 volts, and each load resistor is connected directly in parallel
with its respective source, the voltage across each load must be 120 volts as well. Given load
currents of 83.33 amps, each load must still be dissipating 10 kilowatts of power. However,
voltage between the two “hot” wires is not 240 volts (120 6 0o  120 6 180o) because the phase
difference between the two sources is not 180o. Instead, the voltage is:
Etotal = (120 V ∠ 0o)  (120 V ∠ 120o)
Etotal = 207.85 V ∠ 30o
290 CHAPTER 10. POLYPHASE AC CIRCUITS
load
load
#1
#2
+

+

"hot"
"hot"
"neutral"
+

120 V
∠ 0o
120 V
∠ 120o
83.33 A
83.33 A
+

+

120 V
∠ 0o
120 V
∠ 120o
∠ 30o
∠ 0o
∠ 120o
207.85 V
Figure 10.8: Pair of 120 Vac sources phased 120o, similar to splitphase.
Nominally, we say that the voltage between “hot” conductors is 208 volts (rounding up), and
thus the power system voltage is designated as 120/208.
If we calculate the current through the “neutral” conductor, we find that it is not zero, even
with balanced load resistances. Kirchhoff ’s Current Law tells us that the currents entering
and exiting the node between the two loads must be zero: (Figure 10.9)
load
load
#1
#2
"hot"
"hot"
"neutral"
Node
83.33 A ∠ 0o
83.33 A ∠ 120o
+

+

120 V ∠ 0o
120 V ∠ 120o
Ineutral
Figure 10.9: Neutral wire carries a current in the case of a pair of 120o phased sources.
Ineutral = Iload#1 + Iload#2
Iload#1  Iload#2  Ineutral = 0

Ineutral = Iload#1  Iload#2
Ineutral =  (83.33 A ∠ 0o)  (83.33 A ∠ 1200)
Ineutral = 83.33 A ∠ 240o or 83.33 A ∠ 120o
10.2. THREEPHASE POWER SYSTEMS 291
So, we find that the “neutral” wire is carrying a full 83.33 amps, just like each “hot” wire.
Note that we are still conveying 20 kW of total power to the two loads, with each load’s
“hot” wire carrying 83.33 amps as before. With the same amount of current through each “hot”
wire, we must use the same gage copper conductors, so we haven’t reduced system cost over
the splitphase 120/240 system. However, we have realized a gain in safety, because the overall
voltage between the two “hot” conductors is 32 volts lower than it was in the splitphase system
(208 volts instead of 240 volts).
The fact that the neutral wire is carrying 83.33 amps of current raises an interesting pos
sibility: since its carrying current anyway, why not use that third wire as another “hot” con
ductor, powering another load resistor with a third 120 volt source having a phase angle of
240o? That way, we could transmit more power (another 10 kW) without having to add any
more conductors. Let’s see how this might look: (Figure 10.10)
load
load
#1
#2
+

+

+

load #3
 +
83.33 A ∠ 0o
83.33 A ∠ 240o
83.33 A ∠ 120o
120 V
120 V
120 V
∠ 0o
∠ 120o
∠ 240o
208 V
∠ 30o
120 V
120 V
120 V
10 kW
10 kW
10 kW
Figure 10.10: With a third load phased 120o to the other two, the currents are the same as for
two loads.
A full mathematical analysis of all the voltages and currents in this circuit would necessi
tate the use of a network theorem, the easiest being the Superposition Theorem. I’ll spare you
the long, drawnout calculations because you should be able to intuitively understand that the
three voltage sources at three different phase angles will deliver 120 volts each to a balanced
triad of load resistors. For proof of this, we can use SPICE to do the math for us: (Figure 10.11,
SPICE listing: 120/208 polyphase power system)
Sure enough, we get 120 volts across each load resistor, with (approximately) 208 volts
between any two “hot” conductors and conductor currents equal to 83.33 amps. (Figure 10.12)
At that current and voltage, each load will be dissipating 10 kW of power. Notice that this
circuit has no “neutral” conductor to ensure stable voltage to all loads if one should open.
What we have here is a situation similar to our splitphase power circuit with no “neutral”
conductor: if one load should happen to fail open, the voltage drops across the remaining
load(s) will change. To ensure load voltage stability in the event of another load opening, we
need a neutral wire to connect the source node and load node together:
So long as the loads remain balanced (equal resistance, equal currents), the neutral wire
will not have to carry any current at all. It is there just in case one or more load resistors
should fail open (or be shut off through a disconnecting switch).
This circuit we’ve been analyzing with three voltage sources is called a polyphase circuit.
292 CHAPTER 10. POLYPHASE AC CIRCUITS
+

+

 +
1
2
3
0
3
1
2
4
120 V
120 V
120 V
∠ 0o
∠ 240o
∠ 120o
R1
R3
R2
1.44 Ω
1.44 Ω
1.44 Ω
Figure 10.11: SPICE circuit: Three 3Φ loads phased at 120o.
120/208 polyphase power system
v1 1 0 ac 120 0 sin
v2 2 0 ac 120 120 sin
v3 3 0 ac 120 240 sin
r1 1 4 1.44
r2 2 4 1.44
r3 3 4 1.44
.ac lin 1 60 60
.print ac v(1,4) v(2,4) v(3,4)
.print ac v(1,2) v(2,3) v(3,1)
.print ac i(v1) i(v2) i(v3)
.end
VOLTAGE ACROSS EACH LOAD
freq v(1,4) v(2,4) v(3,4)
6.000E+01 1.200E+02 1.200E+02 1.200E+02
VOLTAGE BETWEEN ‘‘HOT’’ CONDUCTORS
freq v(1,2) v(2,3) v(3,1)
6.000E+01 2.078E+02 2.078E+02 2.078E+02
CURRENT THROUGH EACH VOLTAGE SOURCE
freq i(v1) i(v2) i(v3)
6.000E+01 8.333E+01 8.333E+01 8.333E+01
10.2. THREEPHASE POWER SYSTEMS 293
load
load
#1
#2
+

+

load #3
 +
83.33 A ∠ 0o
83.33 A ∠ 240o
83.33 A ∠ 120o
120 V
120 V
120 V
∠ 0o
∠ 120o
∠ 240o
120 V
120 V
120 V
10 kW
10 kW
10 kW
"neutral"
"hot"
"hot"
"hot"
0 A
Figure 10.12: SPICE circuit annotated with simulation results: Three 3Φ loads phased at
120o.
The prefix “poly” simply means “more than one,” as in “polytheism” (belief in more than one de
ity), “polygon” (a geometrical shape made of multiple line segments: for example, pentagon and
hexagon), and “polyatomic” (a substance composed of multiple types of atoms). Since the volt
age sources are all at different phase angles (in this case, three different phase angles), this is
a “polyphase” circuit. More specifically, it is a threephase circuit, the kind used predominantly
in large power distribution systems.
Let’s survey the advantages of a threephase power system over a singlephase system of
equivalent load voltage and power capacity. A singlephase system with three loads connected
directly in parallel would have a very high total current (83.33 times 3, or 250 amps. (Fig
ure 10.13)
load load
#1 #2
load
#3120V
250 A 10 kW 10 kW 10 kW
Figure 10.13: For comparison, three 10 Kw loads on a 120 Vac system draw 250 A.
This would necessitate 3/0 gage copper wire (very large!), at about 510 pounds per thousand
feet, and with a considerable price tag attached. If the distance from source to load was 1000
feet, we would need over a halfton of copper wire to do the job. On the other hand, we could
build a splitphase system with two 15 kW, 120 volt loads. (Figure 10.14)
Our current is half of what it was with the simple parallel circuit, which is a great improve
ment. We could get away with using number 2 gage copper wire at a total mass of about 600
294 CHAPTER 10. POLYPHASE AC CIRCUITS
load
load
#1
#2
240 V
+

+

"hot"
"hot"
"neutral"
0 A
+

125 A ∠ 0o
125 A ∠ 180o
120 V
∠ 0o
120 V
∠ 180o
120 V
120 V
15 kW
15 kW
∠ 0o
Figure 10.14: Split phase system draws half the current of 125 A at 240 Vac compared to 120
Vac system.
pounds, figuring about 200 pounds per thousand feet with three runs of 1000 feet each between
source and loads. However, we also have to consider the increased safety hazard of having 240
volts present in the system, even though each load only receives 120 volts. Overall, there is
greater potential for dangerous electric shock to occur.
When we contrast these two examples against our threephase system (Figure 10.12), the
advantages are quite clear. First, the conductor currents are quite a bit less (83.33 amps versus
125 or 250 amps), permitting the use of much thinner and lighter wire. We can use number
4 gage wire at about 125 pounds per thousand feet, which will total 500 pounds (four runs
of 1000 feet each) for our example circuit. This represents a significant cost savings over the
splitphase system, with the additional benefit that the maximum voltage in the system is
lower (208 versus 240).
One question remains to be answered: how in the world do we get three AC voltage sources
whose phase angles are exactly 120o apart? Obviously we can’t centertap a transformer or
alternator winding like we did in the splitphase system, since that can only give us voltage
waveforms that are either in phase or 180o out of phase. Perhaps we could figure out some
way to use capacitors and inductors to create phase shifts of 120o, but then those phase shifts
would depend on the phase angles of our load impedances as well (substituting a capacitive or
inductive load for a resistive load would change everything!).
The best way to get the phase shifts we’re looking for is to generate it at the source: con
struct the AC generator (alternator) providing the power in such a way that the rotating mag
netic field passes by three sets of wire windings, each set spaced 120o apart around the circum
ference of the machine as in Figure 10.15.
Together, the six “pole” windings of a threephase alternator are connected to comprise three
winding pairs, each pair producing AC voltage with a phase angle 120o shifted from either of
the other two winding pairs. The interconnections between pairs of windings (as shown for the
singlephase alternator: the jumper wire between windings 1a and 1b) have been omitted from
the threephase alternator drawing for simplicity.
In our example circuit, we showed the three voltage sources connected together in a “Y”
configuration (sometimes called the “star” configuration), with one lead of each source tied to
10.2. THREEPHASE POWER SYSTEMS 295
N
S
N
Swinding
1a
winding
1b winding1a
winding
1b
winding
2a
winding
2b
winding
3a
winding
3b
Singlephase alternator (a)
Threephase alternator (b)
Figure 10.15: (a) Singlephase alternator, (b) Threephase alternator.
a common point (the node where we attached the “neutral” conductor). The common way to
depict this connection scheme is to draw the windings in the shape of a “Y” like Figure 10.16.
+


+

+
120 V 120 V
120 V
∠ 0o ∠ 120o
∠ 240o
Figure 10.16: Alternator ”Y” configuration.
The “Y” configuration is not the only option open to us, but it is probably the easiest to
understand at first. More to come on this subject later in the chapter.
• REVIEW:
• A singlephase power system is one where there is only one AC voltage source (one source
voltage waveform).
• A splitphase power system is one where there are two voltage sources, 180o phaseshifted
from each other, powering a two seriesconnected loads. The advantage of this is the
ability to have lower conductor currents while maintaining low load voltages for safety
reasons.
• A polyphase power system uses multiple voltage sources at different phase angles from
each other (many “phases” of voltage waveforms at work). A polyphase power system
can deliver more power at less voltage with smallergage conductors than single or split
phase systems.
296 CHAPTER 10. POLYPHASE AC CIRCUITS
• The phaseshifted voltage sources necessary for a polyphase power system are created in
alternators with multiple sets of wire windings. These winding sets are spaced around
the circumference of the rotor’s rotation at the desired angle(s).
10.3 Phase rotation
Let’s take the threephase alternator design laid out earlier (Figure 10.17) and watch what
happens as the magnet rotates.
N
S
winding
1a
winding
1b
winding
2a
winding
2b
winding
3a
winding
3b
Figure 10.17: Threephase alternator
The phase angle shift of 120o is a function of the actual rotational angle shift of the three
pairs of windings (Figure 10.18). If the magnet is rotating clockwise, winding 3 will generate its
peak instantaneous voltage exactly 120o (of alternator shaft rotation) after winding 2, which
will hits its peak 120o after winding 1. The magnet passes by each pole pair at different
positions in the rotational movement of the shaft. Where we decide to place the windings will
dictate the amount of phase shift between the windings’ AC voltage waveforms. If we make
winding 1 our “reference” voltage source for phase angle (0o), then winding 2 will have a phase
angle of 120o (120o lagging, or 240o leading) and winding 3 an angle of 240o (or 120o leading).
This sequence of phase shifts has a definite order. For clockwise rotation of the shaft, the
order is 123 (winding 1 peaks first, them winding 2, then winding 3). This order keeps re
peating itself as long as we continue to rotate the alternator’s shaft. (Figure 10.18)
However, if we reverse the rotation of the alternator’s shaft (turn it counterclockwise), the
magnet will pass by the pole pairs in the opposite sequence. Instead of 123, we’ll have 321.
Now, winding 2’s waveform will be leading 120o ahead of 1 instead of lagging, and 3 will be
another 120o ahead of 2. (Figure 10.19)
The order of voltage waveform sequences in a polyphase system is called phase rotation or
phase sequence. If we’re using a polyphase voltage source to power resistive loads, phase rota
tion will make no difference at all. Whether 123 or 321, the voltage and current magnitudes
will all be the same. There are some applications of threephase power, as we will see shortly,
that depend on having phase rotation being one way or the other. Since voltmeters and amme
ters would be useless in telling us what the phase rotation of an operating power system is, we
10.3. PHASE ROTATION 297
1 2 3
1 2 3   1  2  3  1  2  3
phase sequence:
TIME
Figure 10.18: Clockwise rotation phase sequence: 123.
123
3  12 
phase sequence:
TIME
3  2  1  3  2  1
Figure 10.19: Counterclockwise rotation phase sequence: 321.
298 CHAPTER 10. POLYPHASE AC CIRCUITS
need to have some other kind of instrument capable of doing the job.
One ingenious circuit design uses a capacitor to introduce a phase shift between voltage
and current, which is then used to detect the sequence by way of comparison between the
brightness of two indicator lamps in Figure 10.20.
to phase
#1
to phase
#2
to phase
#3
C
Figure 10.20: Phase sequence detector compares brightness of two lamps.
The two lamps are of equal filament resistance and wattage. The capacitor is sized to have
approximately the same amount of reactance at system frequency as each lamp’s resistance.
If the capacitor were to be replaced by a resistor of equal value to the lamps’ resistance, the
two lamps would glow at equal brightness, the circuit being balanced. However, the capacitor
introduces a phase shift between voltage and current in the third leg of the circuit equal to
90o. This phase shift, greater than 0o but less than 120o, skews the voltage and current values
across the two lamps according to their phase shifts relative to phase 3. The following SPICE
analysis demonstrates what will happen: (Figure 10.21), ”phase rotation detector – sequence
= v1v2v3”
120 V
120 V
+

+

120 V
 +
1
2
3
0
3
1
2
4
2650 Ω
2650 Ω
R1
R2
C1
1 µF
∠ 0o
∠ 240o
∠ 120o
Figure 10.21: SPICE circuit for phase sequence detector.
The resulting phase shift from the capacitor causes the voltage across phase 1 lamp (be
tween nodes 1 and 4) to fall to 48.1 volts and the voltage across phase 2 lamp (between nodes
10.3. PHASE ROTATION 299
phase rotation detector  sequence = v1v2v3
v1 1 0 ac 120 0 sin
v2 2 0 ac 120 120 sin
v3 3 0 ac 120 240 sin
r1 1 4 2650
r2 2 4 2650
c1 3 4 1u
.ac lin 1 60 60
.print ac v(1,4) v(2,4) v(3,4)
.end
freq v(1,4) v(2,4) v(3,4)
6.000E+01 4.810E+01 1.795E+02 1.610E+02
2 and 4) to rise to 179.5 volts, making the first lamp dim and the second lamp bright. Just the
opposite will happen if the phase sequence is reversed: ”phase rotation detector – sequence =
v3v2v1 ”
phase rotation detector  sequence = v3v2v1
v1 1 0 ac 120 240 sin
v2 2 0 ac 120 120 sin
v3 3 0 ac 120 0 sin
r1 1 4 2650
r2 2 4 2650
c1 3 4 1u
.ac lin 1 60 60
.print ac v(1,4) v(2,4) v(3,4)
.end
freq v(1,4) v(2,4) v(3,4)
6.000E+01 1.795E+02 4.810E+01 1.610E+02
Here,(”phase rotation detector – sequence = v3v2v1”) the first lamp receives 179.5 volts
while the second receives only 48.1 volts.
We’ve investigated how phase rotation is produced (the order in which pole pairs get passed
by the alternator’s rotating magnet) and how it can be changed by reversing the alternator’s
shaft rotation. However, reversal of the alternator’s shaft rotation is not usually an option
open to an enduser of electrical power supplied by a nationwide grid (“the” alternator actually
being the combined total of all alternators in all power plants feeding the grid). There is a
much easier way to reverse phase sequence than reversing alternator rotation: just exchange
any two of the three “hot” wires going to a threephase load.
This trick makes more sense if we take another look at a running phase sequence of a
threephase voltage source:
123 rotation: 123123123123123 . . .
321 rotation: 321321321321321 . . .
What is commonly designated as a “123” phase rotation could just as well be called “231”
or “312,” going from left to right in the number string above. Likewise, the opposite rotation
300 CHAPTER 10. POLYPHASE AC CIRCUITS
(321) could just as easily be called “213” or “132.”
Starting out with a phase rotation of 321, we can try all the possibilities for swapping any
two of the wires at a time and see what happens to the resulting sequence in Figure 10.22.
Original 123
phase rotation End result
1
2
3 3
2
1
(wires 1 and 2 swapped)
1
2
3 2
3
1 (wires 2 and 3 swapped)
phase rotation = 213
phase rotation = 132
1
2
3 1
3
2
(wires 1 and 3 swapped)
phase rotation = 321
Figure 10.22: All possibilities of swapping any two wires.
No matter which pair of “hot” wires out of the three we choose to swap, the phase rotation
ends up being reversed (123 gets changed to 213, 132 or 321, all equivalent).
• REVIEW:
• Phase rotation, or phase sequence, is the order in which the voltage waveforms of a
polyphase AC source reach their respective peaks. For a threephase system, there are
only two possible phase sequences: 123 and 321, corresponding to the two possible
directions of alternator rotation.
• Phase rotation has no impact on resistive loads, but it will have impact on unbalanced
reactive loads, as shown in the operation of a phase rotation detector circuit.
• Phase rotation can be reversed by swapping any two of the three “hot” leads supplying
threephase power to a threephase load.
10.4 Polyphase motor design
Perhaps the most important benefit of polyphase AC power over singlephase is the design and
operation of AC motors. As we studied in the first chapter of this book, some types of AC motors
are virtually identical in construction to their alternator (generator) counterparts, consisting
of stationary wire windings and a rotating magnet assembly. (Other AC motor designs are not
quite this simple, but we will leave those details to another lesson).
10.4. POLYPHASE MOTOR DESIGN 301
N S
+
II
N
S
N
S
NS
+
I I
Step #1 Step #2
Step #3 Step #4
S SNN
NS NS
Figure 10.23: Clockwise AC motor operation.
If the rotating magnet is able to keep up with the frequency of the alternating current
energizing the electromagnet windings (coils), it will continue to be pulled around clockwise.
(Figure 10.23) However, clockwise is not the only valid direction for this motor’s shaft to spin.
It could just as easily be powered in a counterclockwise direction by the same AC voltage
waveform a in Figure 10.24.
N S
+
II
N
S
N
S
NS
+
I I
Step #1 Step #2
Step #3 Step #4
S SNN
NS NS
Figure 10.24: Counterclockwise AC motor operation.
302 CHAPTER 10. POLYPHASE AC CIRCUITS
Notice that with the exact same sequence of polarity cycles (voltage, current, and magnetic
poles produced by the coils), the magnetic rotor can spin in either direction. This is a common
trait of all singlephase AC “induction” and “synchronous” motors: they have no normal or “cor
rect” direction of rotation. The natural question should arise at this point: how can the motor
get started in the intended direction if it can run either way just as well? The answer is that
these motors need a little help getting started. Once helped to spin in a particular direction.
they will continue to spin that way as long as AC power is maintained to the windings.
Where that “help” comes from for a singlephase AC motor to get going in one direction can
vary. Usually, it comes from an additional set of windings positioned differently from the main
set, and energized with an AC voltage that is out of phase with the main power. (Figure 10.25)
N
S
winding
1a
winding
1b
winding
2a
winding
2b
winding 2’s voltage waveform is 90 degrees
out of phase with winding 1’s voltage waveform
winding 2’s voltage waveform is 90 degrees
out of phase with winding 1’s voltage waveform
Figure 10.25: Unidirectionalstarting AC twophase motor.
These supplementary coils are typically connected in series with a capacitor to introduce a
phase shift in current between the two sets of windings. (Figure 10.26)
That phase shift creates magnetic fields from coils 2a and 2b that are equally out of step
with the fields from coils 1a and 1b. The result is a set of magnetic fields with a definite phase
rotation. It is this phase rotation that pulls the rotating magnet around in a definite direction.
Polyphase AC motors require no such trickery to spin in a definite direction. Because their
supply voltage waveforms already have a definite rotation sequence, so do the respective mag
netic fields generated by the motor’s stationary windings. In fact, the combination of all three
phase winding sets working together creates what is often called a rotating magnetic field. It
was this concept of a rotating magnetic field that inspired Nikola Tesla to design the world’s
first polyphase electrical systems (simply to make simpler, more efficient motors). The line
current and safety advantages of polyphase power over single phase power were discovered
later.
What can be a confusing concept is made much clearer through analogy. Have you ever
seen a row of blinking light bulbs such as the kind used in Christmas decorations? Some
strings appear to “move” in a definite direction as the bulbs alternately glow and darken in
sequence. Other strings just blink on and off with no apparent motion. What makes the
difference between the two types of bulb strings? Answer: phase shift!
Examine a string of lights where every other bulb is lit at any given time as in (Figure 10.27)
10.4. POLYPHASE MOTOR DESIGN 303
1a
1b
2a
2b
out of phase with each other
these two branch currents are
C
I I
Figure 10.26: Capacitor phase shift adds second phase.
Figure 10.27: Phase sequence 1212: lamps appear to move.
304 CHAPTER 10. POLYPHASE AC CIRCUITS
When all of the “1” bulbs are lit, the “2” bulbs are dark, and vice versa. With this blinking
sequence, there is no definite “motion” to the bulbs’ light. Your eyes could follow a “motion”
from left to right just as easily as from right to left. Technically, the “1” and “2” bulb blinking
sequences are 180o out of phase (exactly opposite each other). This is analogous to the single
phase AC motor, which can run just as easily in either direction, but which cannot start on its
own because its magnetic field alternation lacks a definite “rotation.”
Now let’s examine a string of lights where there are three sets of bulbs to be sequenced in
stead of just two, and these three sets are equally out of phase with each other in Figure 10.28.
1 2 3 1 2 3 1 2 3 1 2 3
1 2 3 1 2 3 1 2 3 1 2 3
1 2 3 1 2 3 1 2 3 1 2 3
1 2 3 1 2 3 1 2 3 1 2 3
phase sequence = 123
all "1" bulbs lit
all "2" bulbs lit
all "3" bulbs lit
all "1" bulbs lit
Time
bulbs appear to be "moving" from left to right
Figure 10.28: Phase sequence: 123: bulbs appear to move left to right.
If the lighting sequence is 123 (the sequence shown in (Figure 10.28)), the bulbs will
appear to “move” from left to right. Now imagine this blinking string of bulbs arranged into a
circle as in Figure 10.29.
Now the lights in Figure 10.29 appear to be “moving” in a clockwise direction because they
are arranged around a circle instead of a straight line. It should come as no surprise that the
appearance of motion will reverse if the phase sequence of the bulbs is reversed.
The blinking pattern will either appear to move clockwise or counterclockwise depending
on the phase sequence. This is analogous to a threephase ACmotor with three sets of windings
energized by voltage sources of three different phase shifts in Figure 10.30.
With phase shifts of less than 180o we get true rotation of the magnetic field. With single
phase motors, the rotating magnetic field necessary for selfstarting must to be created by way
of capacitive phase shift. With polyphase motors, the necessary phase shifts are there already.
Plus, the direction of shaft rotation for polyphase motors is very easily reversed: just swap any
two “hot” wires going to the motor, and it will run in the opposite direction!
• REVIEW:
10.4. POLYPHASE MOTOR DESIGN 305
1 1
2 3
23
all "1" bulbs lit
all "2" bulbs lit
all "3" bulbs lit
The bulbs appear to
"move" in a clockwise
direction
1
32
1
3 2
1
32
1
3 2
Figure 10.29: Circular arrangement; bulbs appear to rotate clockwise.
N
S
winding
1a
winding
1b
winding
2a
winding
2b
winding
3a
winding
3b
Figure 10.30: Threephase AC motor: A phase sequence of 123 spins the magnet clockwise,
321 spins the magnet counterclockwise.
306 CHAPTER 10. POLYPHASE AC CIRCUITS
• AC “induction” and “synchronous” motors work by having a rotating magnet follow the
alternating magnetic fields produced by stationary wire windings.
• Singlephase AC motors of this type need help to get started spinning in a particular
direction.
• By introducing a phase shift of less than 180o to the magnetic fields in such a motor, a
definite direction of shaft rotation can be established.
• Singlephase induction motors often use an auxiliary winding connected in series with a
capacitor to create the necessary phase shift.
• Polyphase motors don’t need such measures; their direction of rotation is fixed by the
phase sequence of the voltage they’re powered by.
• Swapping any two “hot” wires on a polyphase AC motor will reverse its phase sequence,
thus reversing its shaft rotation.
10.5 Threephase Y and Delta configurations
Initially we explored the idea of threephase power systems by connecting three voltage sources
together in what is commonly known as the “Y” (or “star”) configuration. This configuration of
voltage sources is characterized by a common connection point joining one side of each source.
(Figure 10.31)
+

+

 +
120 V
120 V
120 V
∠ 0o
∠ 120o
∠ 240o
Figure 10.31: Threephase “Y” connection has three voltage sources connected to a common
point.
If we draw a circuit showing each voltage source to be a coil of wire (alternator or trans
former winding) and do some slight rearranging, the “Y” configuration becomes more obvious
in Figure 10.32.
The three conductors leading away from the voltage sources (windings) toward a load are
typically called lines, while the windings themselves are typically called phases. In a Y
connected system, there may or may not (Figure 10.33) be a neutral wire attached at the
junction point in the middle, although it certainly helps alleviate potential problems should
one element of a threephase load fail open, as discussed earlier.
10.5. THREEPHASE Y AND DELTA CONFIGURATIONS 307
+


+

+
120 V 120 V
120 V
"line"
"line"
"line"
"neutral"∠ 0
o ∠ 120o
∠ 240o
Figure 10.32: Threephase, fourwire “Y” connection uses a ”common” fourth wire.
+


+

+
120 V 120 V
120 V
"line"
"line"
"line"
∠ 0o ∠ 120
o
∠ 240o
(no "neutral" wire)
3phase, 3wire "Y" connection
Figure 10.33: Threephase, threewire “Y” connection does not use the neutral wire.
308 CHAPTER 10. POLYPHASE AC CIRCUITS
When we measure voltage and current in threephase systems, we need to be specific as to
where we’re measuring. Line voltage refers to the amount of voltage measured between any
two line conductors in a balanced threephase system. With the above circuit, the line voltage
is roughly 208 volts. Phase voltage refers to the voltage measured across any one component
(source winding or load impedance) in a balanced threephase source or load. For the circuit
shown above, the phase voltage is 120 volts. The terms line current and phase current follow
the same logic: the former referring to current through any one line conductor, and the latter
to current through any one component.
Yconnected sources and loads always have line voltages greater than phase voltages, and
line currents equal to phase currents. If the Yconnected source or load is balanced, the line
voltage will be equal to the phase voltage times the square root of 3:
Eline = Ephase3
Iline = Iphase
For "Y" circuits:
However, the “Y” configuration is not the only valid one for connecting threephase voltage
source or load elements together. Another configuration is known as the “Delta,” for its geo
metric resemblance to the Greek letter of the same name (∆). Take close notice of the polarity
for each winding in Figure 10.34.
"line"
"line"
"line"
+ 
+
 +

120 V120 V
120 V ∠ 0o
∠ 240o ∠ 120o
Figure 10.34: Threephase, threewire ∆ connection has no common.
At first glance it seems as though three voltage sources like this would create a shortcircuit,
electrons flowing around the triangle with nothing but the internal impedance of the windings
to hold them back. Due to the phase angles of these three voltage sources, however, this is not
the case.
One quick check of this is to use Kirchhoff ’s Voltage Law to see if the three voltages around
the loop add up to zero. If they do, then there will be no voltage available to push current
around and around that loop, and consequently there will be no circulating current. Starting
with the top winding and progressing counterclockwise, our KVL expression looks something
like this:
10.5. THREEPHASE Y AND DELTA CONFIGURATIONS 309
(120 V ∠ 0o) + (120 V ∠ 240o) + (120 V ∠ 120o)
Does it all equal 0?
Yes!
Indeed, if we add these three vector quantities together, they do add up to zero. Another way
to verify the fact that these three voltage sources can be connected together in a loop without
resulting in circulating currents is to open up the loop at one junction point and calculate
voltage across the break: (Figure 10.35)
+ 
+
 +

120 V120 V
120 V ∠ 0o
∠ 240o ∠ 120o
Ebreak should equal 0 V
Figure 10.35: Voltage across open ∆ should be zero.
Starting with the right winding (120 V 6 120o) and progressing counterclockwise, our KVL
equation looks like this:
(120 V ∠ 120o) + (120 ∠ 0o) + (120 V ∠ 240o) + Ebreak = 0
0 + Ebreak = 0
Ebreak = 0
Sure enough, there will be zero voltage across the break, telling us that no current will
circulate within the triangular loop of windings when that connection is made complete.
Having established that a ∆connected threephase voltage source will not burn itself to a
crisp due to circulating currents, we turn to its practical use as a source of power in threephase
circuits. Because each pair of line conductors is connected directly across a single winding in
a ∆ circuit, the line voltage will be equal to the phase voltage. Conversely, because each line
conductor attaches at a node between two windings, the line current will be the vector sum of
the two joining phase currents. Not surprisingly, the resulting equations for a ∆ configuration
are as follows:
310 CHAPTER 10. POLYPHASE AC CIRCUITS
Eline = Ephase
For ∆ ("delta") circuits:
Iline = 3 Iphase
Let’s see how this works in an example circuit: (Figure 10.36)
+ 
+
 +

120 V120 V
10 kW
10 kW 10 kW
120 V ∠ 0o
∠ 120o∠ 240o
Figure 10.36: The load on the ∆ source is wired in a ∆.
With each load resistance receiving 120 volts from its respective phase winding at the
source, the current in each phase of this circuit will be 83.33 amps:
I = P
E
I = 10 kW
120 V
I = 83.33 A (for each load resistor and source winding)
Iline = 3 Iphase
Iline = 3 (83.33 A)
Iline = 144.34 A
So each line current in this threephase power system is equal to 144.34 amps, which is
substantially more than the line currents in the Yconnected system we looked at earlier. One
might wonder if we’ve lost all the advantages of threephase power here, given the fact that we
have such greater conductor currents, necessitating thicker, more costly wire. The answer is
no. Although this circuit would require three number 1 gage copper conductors (at 1000 feet of
distance between source and load this equates to a little over 750 pounds of copper for the whole
system), it is still less than the 1000+ pounds of copper required for a singlephase system
delivering the same power (30 kW) at the same voltage (120 volts conductortoconductor).
10.5. THREEPHASE Y AND DELTA CONFIGURATIONS 311
One distinct advantage of a ∆connected system is its lack of a neutral wire. With a Y
connected system, a neutral wire was needed in case one of the phase loads were to fail open
(or be turned off), in order to keep the phase voltages at the load from changing. This is
not necessary (or even possible!) in a ∆connected circuit. With each load phase element
directly connected across a respective source phase winding, the phase voltage will be constant
regardless of open failures in the load elements.
Perhaps the greatest advantage of the ∆connected source is its fault tolerance. It is pos
sible for one of the windings in a ∆connected threephase source to fail open (Figure 10.37)
without affecting load voltage or current!
+ 
+

120 Vwinding
failed open!
120 V
120 V 120 V
120 V ∠ 0o
∠ 120o
Figure 10.37: Even with a source winding failure, the line voltage is still 120 V, and load phase
voltage is still 120 V. The only difference is extra current in the remaining functional source
windings.
The only consequence of a source winding failing open for a ∆connected source is increased
phase current in the remaining windings. Compare this fault tolerance with a Yconnected
system suffering an open source winding in Figure 10.38.
+

+

120 V120 V
winding
failed open!
208 V
104 V 104 V∠ 0o ∠ 120o
Figure 10.38: Open “Y” source winding halves the voltage on two loads of a ∆ connected load.
With a ∆connected load, two of the resistances suffer reduced voltage while one remains
at the original line voltage, 208. A Yconnected load suffers an even worse fate (Figure 10.39)
with the same winding failure in a Yconnected source
In this case, two load resistances suffer reduced voltage while the third loses supply voltage
completely! For this reason, ∆connected sources are preferred for reliability. However, if dual
voltages are needed (e.g. 120/208) or preferred for lower line currents, Yconnected systems are
312 CHAPTER 10. POLYPHASE AC CIRCUITS
+

+

120 V120 V
winding
failed open!
104 V 104 V
0 V
∠ 0o ∠ 120o
Figure 10.39: Open source winding of a ”YY” system halves the voltage on two loads, and
looses one load entirely.
the configuration of choice.
• REVIEW:
• The conductors connected to the three points of a threephase source or load are called
lines.
• The three components comprising a threephase source or load are called phases.
• Line voltage is the voltage measured between any two lines in a threephase circuit.
• Phase voltage is the voltage measured across a single component in a threephase source
or load.
• Line current is the current through any one line between a threephase source and load.
• Phase current is the current through any one component comprising a threephase source
or load.
• In balanced “Y” circuits, line voltage is equal to phase voltage times the square root of 3,
while line current is equal to phase current.
•
Eline = Ephase3
Iline = Iphase
For "Y" circuits:
• In balanced ∆ circuits, line voltage is equal to phase voltage, while line current is equal
to phase current times the square root of 3.
•
Eline = Ephase
For ∆ ("delta") circuits:
Iline = 3 Iphase
10.6. THREEPHASE TRANSFORMER CIRCUITS 313
• ∆connected threephase voltage sources give greater reliability in the event of winding
failure than Yconnected sources. However, Yconnected sources can deliver the same
amount of power with less line current than ∆connected sources.
10.6 Threephase transformer circuits
Since threephase is used so often for power distribution systems, it makes sense that we
would need threephase transformers to be able to step voltages up or down. This is only
partially true, as regular singlephase transformers can be ganged together to transform power
between two threephase systems in a variety of configurations, eliminating the requirement
for a special threephase transformer. However, special threephase transformers are built for
those tasks, and are able to perform with less material requirement, less size, and less weight
than their modular counterparts.
A threephase transformer is made of three sets of primary and secondary windings, each
set wound around one leg of an iron core assembly. Essentially it looks like three singlephase
transformers sharing a joined core as in Figure 10.40.
Threephase transformer core
Figure 10.40: Three phase transformer core has three sets of windings.
Those sets of primary and secondary windings will be connected in either ∆ or Y configu
rations to form a complete unit. The various combinations of ways that these windings can be
connected together in will be the focus of this section.
Whether the winding sets share a common core assembly or each winding pair is a separate
transformer, the winding connection options are the same:
• Primary  Secondary
• Y  Y
• Y  ∆
• ∆  Y
• ∆  ∆
The reasons for choosing a Y or ∆ configuration for transformer winding connections are
the same as for any other threephase application: Y connections provide the opportunity for
multiple voltages, while ∆ connections enjoy a higher level of reliability (if one winding fails
open, the other two can still maintain full line voltages to the load).
314 CHAPTER 10. POLYPHASE AC CIRCUITS
Probably the most important aspect of connecting three sets of primary and secondary wind
ings together to form a threephase transformer bank is paying attention to proper winding
phasing (the dots used to denote “polarity” of windings). Remember the proper phase relation
ships between the phase windings of ∆ and Y: (Figure 10.41)
+

+


+
+ 
+


+
(Y)
∠ 0o
∠ 0o
∠ 120o
∠ 120o∠ 240o ∠ 240o
(∆)
Figure 10.41: (Y) The center point of the “Y” must tie either all the “” or all the “+” winding
points together. (∆) The winding polarities must stack together in a complementary manner (
+ to ).
Getting this phasing correct when the windings aren’t shown in regular Y or ∆ configura
tion can be tricky. Let me illustrate, starting with Figure 10.42.
T1 T2 T3
A1
B1
C1
A2
B2
C2
Figure 10.42: Inputs A1, A2, A3 may be wired either “∆” or “Y”, as may outputs B1, B2, B3.
Three individual transformers are to be connected together to transform power from one
threephase system to another. First, I’ll show the wiring connections for a YY configuration:
Figure 10.43
Note in Figure 10.43 how all the winding ends marked with dots are connected to their
respective phases A, B, and C, while the nondot ends are connected together to form the cen
ters of each “Y”. Having both primary and secondary winding sets connected in “Y” formations
allows for the use of neutral conductors (N1 and N2) in each power system.
Now, we’ll take a look at a Y∆ configuration: (Figure 10.44)
Note how the secondary windings (bottom set, Figure 10.44) are connected in a chain, the
“dot” side of one winding connected to the “nondot” side of the next, forming the ∆ loop. At
10.6. THREEPHASE TRANSFORMER CIRCUITS 315
A1
B1
C1
N1
N2
A2
B2
C2
T1 T2 T3
Y  Y
Figure 10.43: Phase wiring for “YY” transformer.
T1 T2 T3
A1
B1
C1
N1
A2
B2
C2
Y  ∆
Figure 10.44: Phase wiring for “Y∆” transformer.
316 CHAPTER 10. POLYPHASE AC CIRCUITS
every connection point between pairs of windings, a connection is made to a line of the second
power system (A, B, and C).
Now, let’s examine a ∆Y system in Figure 10.45.
A1
B1
C1
N2
A2
B2
C2
T1 T2 T3
∆  Y
Figure 10.45: Phase wiring for “∆Y” transformer.
Such a configuration (Figure 10.45) would allow for the provision of multiple voltages (line
toline or linetoneutral) in the second power system, from a source power system having no
neutral.
And finally, we turn to the ∆∆ configuration: (Figure 10.46)
When there is no need for a neutral conductor in the secondary power system, ∆∆ connec
tion schemes (Figure 10.46) are preferred because of the inherent reliability of the ∆ configu
ration.
Considering that a ∆ configuration can operate satisfactorily missing one winding, some
power system designers choose to create a threephase transformer bank with only two trans
formers, representing a ∆∆ configuration with a missing winding in both the primary and
secondary sides: (Figure 10.47)
This configuration is called “V” or “Open∆.” Of course, each of the two transformers have
to be oversized to handle the same amount of power as three in a standard ∆ configuration,
but the overall size, weight, and cost advantages are often worth it. Bear in mind, however,
that with one winding set missing from the ∆ shape, this system no longer provides the fault
tolerance of a normal ∆∆ system. If one of the two transformers were to fail, the load voltage
and current would definitely be affected.
The following photograph (Figure 10.48) shows a bank of stepup transformers at the Grand
Coulee hydroelectric dam in Washington state. Several transformers (green in color) may be
seen from this vantage point, and they are grouped in threes: three transformers per hydro
electric generator, wired together in some form of threephase configuration. The photograph
doesn’t reveal the primary winding connections, but it appears the secondaries are connected
in a Y configuration, being that there is only one large highvoltage insulator protruding from
10.6. THREEPHASE TRANSFORMER CIRCUITS 317
T1 T2 T3
A1
B1
C1
A2
B2
C2
∆  ∆
Figure 10.46: Phase wiring for “∆∆” transformer.
A1
B1
C1
A2
B2
C2
T1 T2
"Open ∆"
Figure 10.47: “V” or “open∆” provides 2φ power with only two transformers.
318 CHAPTER 10. POLYPHASE AC CIRCUITS
each transformer. This suggests the other side of each transformer’s secondary winding is at
or near ground potential, which could only be true in a Y system. The building to the left is the
powerhouse, where the generators and turbines are housed. On the right, the sloping concrete
wall is the downstream face of the dam:
Figure 10.48: Stepup transfromer bank at Grand Coulee hydroelectric dam, Washington state,
USA.
10.7 Harmonics in polyphase power systems
In the chapter on mixedfrequency signals, we explored the concept of harmonics in AC sys
tems: frequencies that are integer multiples of the fundamental source frequency. With AC
power systems where the source voltage waveform coming from an AC generator (alternator)
is supposed to be a singlefrequency sine wave, undistorted, there should be no harmonic con
tent . . . ideally.
This would be true were it not for nonlinear components. Nonlinear components draw
current disproportionately with respect to the source voltage, causing nonsinusoidal current
waveforms. Examples of nonlinear components include gasdischarge lamps, semiconductor
powercontrol devices (diodes, transistors, SCRs, TRIACs), transformers (primary winding
magnetization current is usually nonsinusoidal due to the B/H saturation curve of the core),
and electric motors (again, when magnetic fields within the motor’s core operate near satu
ration levels). Even incandescent lamps generate slightly nonsinusoidal currents, as the fil
ament resistance changes throughout the cycle due to rapid fluctuations in temperature. As
we learned in the mixedfrequency chapter, any distortion of an otherwise sinewave shaped
waveform constitutes the presence of harmonic frequencies.
When the nonsinusoidal waveform in question is symmetrical above and below its average
centerline, the harmonic frequencies will be odd integer multiples of the fundamental source
frequency only, with no even integer multiples. (Figure 10.49) Most nonlinear loads produce
10.7. HARMONICS IN POLYPHASE POWER SYSTEMS 319
current waveforms like this, and so evennumbered harmonics (2nd, 4th, 6th, 8th, 10th, 12th,
etc.) are absent or only minimally present in most AC power systems.
Pure sine wave =
1st harmonic only
Figure 10.49: Examples of symmetrical waveforms – odd harmonics only.
Examples of nonsymmetrical waveforms with even harmonics present are shown for refer
ence in Figure 10.50.
Figure 10.50: Examples of nonsymmetrical waveforms – even harmonics present.
Even though half of the possible harmonic frequencies are eliminated by the typically sym
metrical distortion of nonlinear loads, the odd harmonics can still cause problems. Some of
these problems are general to all power systems, singlephase or otherwise. Transformer over
heating due to eddy current losses, for example, can occur in any AC power system where
there is significant harmonic content. However, there are some problems caused by harmonic
currents that are specific to polyphase power systems, and it is these problems to which this
section is specifically devoted.
It is helpful to be able to simulate nonlinear loads in SPICE so as to avoid a lot of complex
mathematics and obtain a more intuitive understanding of harmonic effects. First, we’ll begin
our simulation with a very simple AC circuit: a single sinewave voltage source with a purely
linear load and all associated resistances: (Figure 10.51)
The Rsource and Rline resistances in this circuit do more than just mimic the real world: they
also provide convenient shunt resistances for measuring currents in the SPICE simulation: by
reading voltage across a 1 Ω resistance, you obtain a direct indication of current through it,
since E = IR.
A SPICE simulation of this circuit (SPICE listing: “linear load simulation”) with Fourier
analysis on the voltage measured across Rline should show us the harmonic content of this
circuit’s line current. Being completely linear in nature, we should expect no harmonics other
than the 1st (fundamental) of 60 Hz, assuming a 60 Hz source. See SPICE output “Fourier
components of transient response v(2,3)” and Figure 10.52.
A .plot command appears in the SPICE netlist, and normally this would result in a sine
wave graph output. In this case, however, I’ve purposely omitted the waveform display for
320 CHAPTER 10. POLYPHASE AC CIRCUITS
Vsource
Rsource
Rline
Rload
120 V
1 Ω
1 Ω
1 kΩ
0
1
2 3
0
Figure 10.51: SPICE circuit with single sinewave source.
linear load simulation
vsource 1 0 sin(0 120 60 0 0)
rsource 1 2 1
rline 2 3 1
rload 3 0 1k
.options itl5=0
.tran 0.5m 30m 0 1u
.plot tran v(2,3)
.four 60 v(2,3)
.end
Fourier components of transient response v(2,3)
dc component = 4.028E12
harmonic frequency Fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 1.198E01 1.000000 72.000 0.000
2 1.200E+02 5.793E12 0.000000 51.122 123.122
3 1.800E+02 7.407E12 0.000000 34.624 37.376
4 2.400E+02 9.056E12 0.000000 4.267 76.267
5 3.000E+02 1.651E11 0.000000 83.461 11.461
6 3.600E+02 3.931E11 0.000000 36.399 108.399
7 4.200E+02 2.338E11 0.000000 41.343 30.657
8 4.800E+02 4.716E11 0.000000 53.324 125.324
9 5.400E+02 3.453E11 0.000000 21.691 93.691
total harmonic distortion = 0.000000 percent
10.7. HARMONICS IN POLYPHASE POWER SYSTEMS 321
Figure 10.52: Frequency domain plot of single frequency component. See SPICE listing: “linear
load simulation”.
brevity’s sake – the .plot command is in the netlist simply to satisfy a quirk of SPICE’s
Fourier transform function.
No discrete Fourier transform is perfect, and so we see very small harmonic currents indi
cated (in the picoamp range!) for all frequencies up to the 9th harmonic (in the table ), which
is as far as SPICE goes in performing Fourier analysis. We show 0.1198 amps (1.198E01)
for the “Fourier component” of the 1st harmonic, or the fundamental frequency, which is our
expected load current: about 120 mA, given a source voltage of 120 volts and a load resistance
of 1 kΩ.
Next, I’d like to simulate a nonlinear load so as to generate harmonic currents. This can be
done in two fundamentally different ways. One way is to design a load using nonlinear compo
nents such as diodes or other semiconductor devices which are easy to simulate with SPICE.
Another is to add some AC current sources in parallel with the load resistor. The latter method
is often preferred by engineers for simulating harmonics, since current sources of known value
lend themselves better to mathematical network analysis than components with highly com
plex response characteristics. Since we’re letting SPICE do all the math work, the complexity
of a semiconductor component would cause no trouble for us, but since current sources can be
finetuned to produce any arbitrary amount of current (a convenient feature), I’ll choose the
latter approach shown in Figure 10.53 and SPICE listing: “Nonlinear load simulation”.
In this circuit, we have a current source of 50 mA magnitude and a frequency of 180 Hz,
which is three times the source frequency of 60 Hz. Connected in parallel with the 1 kΩ load
resistor, its current will add with the resistor’s to make a nonsinusoidal total line current. I’ll
show the waveform plot in Figure 10.54 just so you can see the effects of this 3rdharmonic
current on the total current, which would ordinarily be a plain sine wave.
In the Fourier analysis, (See Figure 10.55 and “Fourier components of transient response
v(2,3)”) the mixed frequencies are unmixed and presented separately. Here we see the same
0.1198 amps of 60 Hz (fundamental) current as we did in the first simulation, but appearing in
the 3rd harmonic row we see 49.9 mA: our 50 mA, 180 Hz current source at work. Why don’t
322 CHAPTER 10. POLYPHASE AC CIRCUITS
Vsource
Rsource
Rline
Rload
120 V
1 Ω
1 Ω
1 kΩ
0
1
2 3
0
3
0
60 Hz
50 mA
180 Hz
Figure 10.53: SPICE circuit: 60 Hz source with 3rd harmonic added.
Nonlinear load simulation
vsource 1 0 sin(0 120 60 0 0)
rsource 1 2 1
rline 2 3 1
rload 3 0 1k
i3har 3 0 sin(0 50m 180 0 0)
.options itl5=0
.tran 0.5m 30m 0 1u
.plot tran v(2,3)
.four 60 v(2,3)
.end
Figure 10.54: SPICE timedomain plot showing sum of 60 Hz source and 3rd harmonic of 180
Hz.
10.7. HARMONICS IN POLYPHASE POWER SYSTEMS 323
Fourier components of transient response v(2,3)
dc component = 1.349E11
harmonic frequency Fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 1.198E01 1.000000 72.000 0.000
2 1.200E+02 1.609E11 0.000000 67.570 139.570
3 1.800E+02 4.990E02 0.416667 144.000 216.000
4 2.400E+02 1.074E10 0.000000 169.546 97.546
5 3.000E+02 3.871E11 0.000000 169.582 241.582
6 3.600E+02 5.736E11 0.000000 140.845 212.845
7 4.200E+02 8.407E11 0.000000 177.071 249.071
8 4.800E+02 1.329E10 0.000000 156.772 228.772
9 5.400E+02 2.619E10 0.000000 160.498 232.498
total harmonic distortion = 41.666663 percent
Figure 10.55: SPICE Fourier plot showing 60 Hz source and 3rd harmonic of 180 Hz.
324 CHAPTER 10. POLYPHASE AC CIRCUITS
we see the entire 50 mA through the line? Because that current source is connected across the
1 kΩ load resistor, so some of its current is shunted through the load and never goes through
the line back to the source. It’s an inevitable consequence of this type of simulation, where one
part of the load is “normal” (a resistor) and the other part is imitated by a current source.
If we were to add more current sources to the “load,” we would see further distortion of
the line current waveform from the ideal sinewave shape, and each of those harmonic cur
rents would appear in the Fourier analysis breakdown. See Figure 10.56 and SPICE listing:
“Nonlinear load simulation”.
Vsource
Rsource
Rline
Rload
120 V
1 Ω
1 Ω
1 kΩ
0
1
2 3
0
3
0
60 Hz
50 mA
180 Hz
0
3
50 mA
300 Hz
50 mA
3
0
420 Hz
3
0
50 mA
540 Hz
Nonlinear load: 1st, 3rd, 5th, 7th, and 9th
harmonics present
Figure 10.56: Nonlinear load: 1st, 3rd, 5th, 7th, and 9th harmonics present.
Nonlinear load simulation
vsource 1 0 sin(0 120 60 0 0)
rsource 1 2 1
rline 2 3 1
rload 3 0 1k
i3har 3 0 sin(0 50m 180 0 0)
i5har 3 0 sin(0 50m 300 0 0)
i7har 3 0 sin(0 50m 420 0 0)
i9har 3 0 sin(0 50m 540 0 0)
.options itl5=0
.tran 0.5m 30m 0 1u
.plot tran v(2,3)
.four 60 v(2,3)
.end
As you can see from the Fourier analysis, (Figure 10.57) every harmonic current source is
equally represented in the line current, at 49.9 mA each. So far, this is just a singlephase
power system simulation. Things get more interesting when we make it a threephase simula
tion. Two Fourier analyses will be performed: one for the voltage across a line resistor, and one
for the voltage across the neutral resistor. As before, reading voltages across fixed resistances
of 1 Ω each gives direct indications of current through those resistors. See Figure 10.58 and
SPICE listing “YY source/load 4wire system with harmonics”.
10.7. HARMONICS IN POLYPHASE POWER SYSTEMS 325
Fourier components of transient response v(2,3)
dc component = 6.299E11
harmonic frequency Fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 1.198E01 1.000000 72.000 0.000
2 1.200E+02 1.900E09 0.000000 93.908 21.908
3 1.800E+02 4.990E02 0.416667 144.000 216.000
4 2.400E+02 5.469E09 0.000000 116.873 44.873
5 3.000E+02 4.990E02 0.416667 0.000 72.000
6 3.600E+02 6.271E09 0.000000 85.062 157.062
7 4.200E+02 4.990E02 0.416666 144.000 72.000
8 4.800E+02 2.742E09 0.000000 38.781 33.219
9 5.400E+02 4.990E02 0.416666 72.000 144.000
total harmonic distortion = 83.333296 percent
Figure 10.57: Fourier analysis: “Fourier components of transient response v(2,3)”.
326 CHAPTER 10. POLYPHASE AC CIRCUITS
YY source/load 4wire system with harmonics
*
* phase1 voltage source and r (120 v / 0 deg)
vsource1 1 0 sin(0 120 60 0 0)
rsource1 1 2 1
*
* phase2 voltage source and r (120 v / 120 deg)
vsource2 3 0 sin(0 120 60 5.55555m 0)
rsource2 3 4 1
*
* phase3 voltage source and r (120 v / 240 deg)
vsource3 5 0 sin(0 120 60 11.1111m 0)
rsource3 5 6 1
*
* line and neutral wire resistances
rline1 2 8 1
rline2 4 9 1
rline3 6 10 1
rneutral 0 7 1
*
* phase 1 of load
rload1 8 7 1k
i3har1 8 7 sin(0 50m 180 0 0)
i5har1 8 7 sin(0 50m 300 0 0)
i7har1 8 7 sin(0 50m 420 0 0)
i9har1 8 7 sin(0 50m 540 0 0)
*
* phase 2 of load
rload2 9 7 1k
i3har2 9 7 sin(0 50m 180 5.55555m 0)
i5har2 9 7 sin(0 50m 300 5.55555m 0)
i7har2 9 7 sin(0 50m 420 5.55555m 0)
i9har2 9 7 sin(0 50m 540 5.55555m 0)
*
* phase 3 of load
rload3 10 7 1k
i3har3 10 7 sin(0 50m 180 11.1111m 0)
i5har3 10 7 sin(0 50m 300 11.1111m 0)
i7har3 10 7 sin(0 50m 420 11.1111m 0)
i9har3 10 7 sin(0 50m 540 11.1111m 0)
*
* analysis stuff
.options itl5=0
.tran 0.5m 100m 12m 1u
.plot tran v(2,8)
.four 60 v(2,8)
.plot tran v(0,7)
.four 60 v(0,7)
.end
10.7. HARMONICS IN POLYPHASE POWER SYSTEMS 327
Rload1 kΩ
50 mA
180 Hz
50 mA
300 Hz
50 mA
420 Hz
50 mA
540 Hz
Rload1 kΩ
50 mA
180 Hz
50 mA
300 Hz
50 mA
420 Hz
50 mA
540 Hz
Rload1 kΩ
50 mA
180 Hz
50 mA
300 Hz
50 mA
420 Hz
50 mA
540 Hz
Vsource
Rsource
120 V
1 Ω
60 Hz
Rline
1 Ω
Vsource
Rsource
120 V
1 Ω
60 Hz
Vsource
Rsource
120 V
1 Ω
60 Hz
Rline
1 Ω
1 Ω
Rline
1 Ω
Rneutral
0
1
2
3
4
5
6
120o0o
240o
7
8 9
10
Figure 10.58: SPICE circuit: analysis of “line current” and “neutral current”, YY source/load
4wire system with harmonics.
Fourier analysis of line current:
Fourier components of transient response v(2,8)
dc component = 6.404E12
harmonic frequency Fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 1.198E01 1.000000 0.000 0.000
2 1.200E+02 2.218E10 0.000000 172.985 172.985
3 1.800E+02 4.975E02 0.415423 0.000 0.000
4 2.400E+02 4.236E10 0.000000 166.990 166.990
5 3.000E+02 4.990E02 0.416667 0.000 0.000
6 3.600E+02 1.877E10 0.000000 147.146 147.146
7 4.200E+02 4.990E02 0.416666 0.000 0.000
8 4.800E+02 2.784E10 0.000000 148.811 148.811
9 5.400E+02 4.975E02 0.415422 0.000 0.000
total harmonic distortion = 83.209009 percent
Fourier analysis of neutral current:
This is a balanced YY power system, each phase identical to the singlephase AC system
simulated earlier. Consequently, it should come as no surprise that the Fourier analysis for line
current in one phase of the 3phase system is nearly identical to the Fourier analysis for line
current in the singlephase system: a fundamental (60 Hz) line current of 0.1198 amps, and
odd harmonic currents of approximately 50 mA each. See Figure 10.59 and Fourier analysis:
“Fourier components of transient response v(2,8)”
What should be surprising here is the analysis for the neutral conductor’s current, as de
termined by the voltage drop across the Rneutral resistor between SPICE nodes 0 and 7. (Fig
ure 10.60) In a balanced 3phase Y load, we would expect the neutral current to be zero. Each
328 CHAPTER 10. POLYPHASE AC CIRCUITS
Figure 10.59: Fourier analysis of line current in balanced YY system
Fourier components of transient response v(0,7)
dc component = 1.819E10
harmonic frequency Fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 4.337E07 1.000000 60.018 0.000
2 1.200E+02 1.869E10 0.000431 91.206 31.188
3 1.800E+02 1.493E01 344147.7638 180.000 240.018
4 2.400E+02 1.257E09 0.002898 21.103 81.121
5 3.000E+02 9.023E07 2.080596 119.981 59.963
6 3.600E+02 3.396E10 0.000783 15.882 44.136
7 4.200E+02 1.264E06 2.913955 59.993 0.025
8 4.800E+02 5.975E10 0.001378 35.584 24.434
9 5.400E+02 1.493E01 344147.4889 179.999 240.017
10.7. HARMONICS IN POLYPHASE POWER SYSTEMS 329
Figure 10.60: Fourier analysis of neutral current shows other than no harmonics! Compare to
line current in Figure 10.59
phase current – which by itself would go through the neutral wire back to the supplying phase
on the source Y – should cancel each other in regard to the neutral conductor because they’re
all the same magnitude and all shifted 120o apart. In a system with no harmonic currents,
this is what happens, leaving zero current through the neutral conductor. However, we cannot
say the same for harmonic currents in the same system.
Note that the fundamental frequency (60 Hz, or the 1st harmonic) current is virtually ab
sent from the neutral conductor. Our Fourier analysis shows only 0.4337 µA of 1st harmonic
when reading voltage across Rneutral. The same may be said about the 5th and 7th harmonics,
both of those currents having negligible magnitude. In contrast, the 3rd and 9th harmonics
are strongly represented within the neutral conductor, with 149.3 mA (1.493E01 volts across
1 Ω) each! This is very nearly 150 mA, or three times the current sources’ values, individually.
With three sources per harmonic frequency in the load, it appears our 3rd and 9th harmonic
currents in each phase are adding to form the neutral current. See Fourier analysis: “Fourier
components of transient response v(0,7) ”
This is exactly what’s happening, though it might not be apparent why this is so. The key
to understanding this is made clear in a timedomain graph of phase currents. Examine this
plot of balanced phase currents over time, with a phase sequence of 123. (Figure 10.61)
With the three fundamental waveforms equally shifted across the time axis of the graph,
it is easy to see how they would cancel each other to give a resultant current of zero in the
neutral conductor. Let’s consider, though, what a 3rd harmonic waveform for phase 1 would
look like superimposed on the graph in Figure 10.62.
Observe how this harmonic waveform has the same phase relationship to the 2nd and 3rd
fundamental waveforms as it does with the 1st: in each positive halfcycle of any of the funda
mental waveforms, you will find exactly two positive halfcycles and one negative halfcycle of
the harmonic waveform. What this means is that the 3rdharmonic waveforms of three 120o
phaseshifted fundamentalfrequency waveforms are actually in phase with each other. The
phase shift figure of 120o generally assumed in threephase AC systems applies only to the
330 CHAPTER 10. POLYPHASE AC CIRCUITS
1 2 3
1 2 3   1  2  3  1  2  3
phase sequence:
TIME
Figure 10.61: Phase sequence 123123123 of equally spaced waves.
1 2 3
TIME
Figure 10.62: Third harmonic waveform for phase1 superimposed on threephase fundamen
tal waveforms.
10.7. HARMONICS IN POLYPHASE POWER SYSTEMS 331
fundamental frequencies, not to their harmonic multiples!
If we were to plot all three 3rdharmonic waveforms on the same graph, we would see them
precisely overlap and appear as a single, unified waveform (shown in bold in (Figure 10.63)
1 2 3
TIME
Figure 10.63: Third harmonics for phases 1, 2, 3 all coincide when superimposed on the funda
mental threephase waveforms.
For the more mathematically inclined, this principle may be expressed symbolically. Sup
pose that A represents one waveform and B another, both at the same frequency, but shifted
120o from each other in terms of phase. Let’s call the 3rd harmonic of each waveform A’ and
B’, respectively. The phase shift between A’ and B’ is not 120o (that is the phase shift between
A and B), but 3 times that, because the A’ and B’ waveforms alternate three times as fast as
A and B. The shift between waveforms is only accurately expressed in terms of phase angle
when the same angular velocity is assumed. When relating waveforms of different frequency,
the most accurate way to represent phase shift is in terms of time; and the timeshift between
A’ and B’ is equivalent to 120o at a frequency three times lower, or 360o at the frequency of A’
and B’. A phase shift of 360o is the same as a phase shift of 0o, which is to say no phase shift
at all. Thus, A’ and B’ must be in phase with each other:
Fundamental A B C0o 120o 240o
A’ B’ C’
3 x 0o 3 x 120o 3 x 240o
(0o) (360o = 0o) (720o = 0o)
Phase sequence = ABC
3rd harmonic
This characteristic of the 3rd harmonic in a threephase system also holds true for any in
teger multiples of the 3rd harmonic. So, not only are the 3rd harmonic waveforms of each fun
damental waveform in phase with each other, but so are the 6th harmonics, the 9th harmonics,
the 12th harmonics, the 15th harmonics, the 18th harmonics, the 21st harmonics, and so on.
Since only odd harmonics appear in systems where waveform distortion is symmetrical about
the centerline – and most nonlinear loads create symmetrical distortion – evennumbered mul
tiples of the 3rd harmonic (6th, 12th, 18th, etc.) are generally not significant, leaving only the
oddnumbered multiples (3rd, 9th, 15th, 21st, etc.) to significantly contribute to neutral cur
rents.
In polyphase power systems with some number of phases other than three, this effect occurs
332 CHAPTER 10. POLYPHASE AC CIRCUITS
with harmonics of the same multiple. For instance, the harmonic currents that add in the neu
tral conductor of a starconnected 4phase system where the phase shift between fundamental
waveforms is 90o would be the 4th, 8th, 12th, 16th, 20th, and so on.
Due to their abundance and significance in threephase power systems, the 3rd harmonic
and its multiples have their own special name: triplen harmonics. All triplen harmonics add
with each other in the neutral conductor of a 4wire Yconnected load. In power systems con
taining substantial nonlinear loading, the triplen harmonic currents may be of great enough
magnitude to cause neutral conductors to overheat. This is very problematic, as other safety
concerns prohibit neutral conductors from having overcurrent protection, and thus there is no
provision for automatic interruption of these high currents.
The following illustration shows how triplen harmonic currents created at the load add
within the neutral conductor. The symbol “ω” is used to represent angular velocity, and is
mathematically equivalent to 2pif. So, “ω” represents the fundamental frequency, “3ω ” repre
sents the 3rd harmonic, “5ω” represents the 5th harmonic, and so on: (Figure 10.64)
Source Load
ω
3ω
5ω
7ω
9ω
ω
3ω
5ω
7ω
9ω
ω
3ω
5ω
7ω
9ω
ω 3ω 5ω 7ω 9ω
ω 3ω 5ω 7ω 9ω
ω 3ω 5ω 7ω 9ω
3ω 9ω
3ω 9ω
3ω 9ω
line
line
line
neutral
Figure 10.64: “YY”Triplen source/load: Harmonic currents add in neutral conductor.
In an effort to mitigate these additive triplen currents, one might be tempted to remove the
neutral wire entirely. If there is no neutral wire in which triplen currents can flow together,
then they won’t, right? Unfortunately, doing so just causes a different problem: the load’s “Y”
centerpoint will no longer be at the same potential as the source’s, meaning that each phase
of the load will receive a different voltage than what is produced by the source. We’ll rerun
the last SPICE simulation without the 1 Ω Rneutral resistor and see what happens:
Fourier analysis of line current:
Fourier analysis of voltage between the two “Y” centerpoints:
Fourier analysis of load phase voltage:
Strange things are happening, indeed. First, we see that the triplen harmonic currents (3rd
and 9th) all but disappear in the lines connecting load to source. The 5th and 7th harmonic
currents are present at their normal levels (approximately 50 mA), but the 3rd and 9th har
monic currents are of negligible magnitude. Second, we see that there is substantial harmonic
10.7. HARMONICS IN POLYPHASE POWER SYSTEMS 333
YY source/load (no neutral) with harmonics
*
* phase1 voltage source and r (120 v / 0 deg)
vsource1 1 0 sin(0 120 60 0 0)
rsource1 1 2 1
*
* phase2 voltage source and r (120 v / 120 deg)
vsource2 3 0 sin(0 120 60 5.55555m 0)
rsource2 3 4 1
*
* phase3 voltage source and r (120 v / 240 deg)
vsource3 5 0 sin(0 120 60 11.1111m 0)
rsource3 5 6 1
*
* line resistances
rline1 2 8 1
rline2 4 9 1
rline3 6 10 1
*
* phase 1 of load
rload1 8 7 1k
i3har1 8 7 sin(0 50m 180 0 0)
i5har1 8 7 sin(0 50m 300 0 0)
i7har1 8 7 sin(0 50m 420 0 0)
i9har1 8 7 sin(0 50m 540 0 0)
*
* phase 2 of load
rload2 9 7 1k
i3har2 9 7 sin(0 50m 180 5.55555m 0)
i5har2 9 7 sin(0 50m 300 5.55555m 0)
i7har2 9 7 sin(0 50m 420 5.55555m 0)
i9har2 9 7 sin(0 50m 540 5.55555m 0)
*
* phase 3 of load
rload3 10 7 1k
i3har3 10 7 sin(0 50m 180 11.1111m 0)
i5har3 10 7 sin(0 50m 300 11.1111m 0)
i7har3 10 7 sin(0 50m 420 11.1111m 0)
i9har3 10 7 sin(0 50m 540 11.1111m 0)
*
* analysis stuff
.options itl5=0
.tran 0.5m 100m 12m 1u
.plot tran v(2,8)
.four 60 v(2,8)
.plot tran v(0,7)
.four 60 v(0,7)
.plot tran v(8,7)
.four 60 v(8,7)
.end
334 CHAPTER 10. POLYPHASE AC CIRCUITS
Fourier components of transient response v(2,8)
dc component = 5.423E11
harmonic frequency Fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 1.198E01 1.000000 0.000 0.000
2 1.200E+02 2.388E10 0.000000 158.016 158.016
3 1.800E+02 3.136E07 0.000003 90.009 90.009
4 2.400E+02 5.963E11 0.000000 111.510 111.510
5 3.000E+02 4.990E02 0.416665 0.000 0.000
6 3.600E+02 8.606E11 0.000000 124.565 124.565
7 4.200E+02 4.990E02 0.416668 0.000 0.000
8 4.800E+02 8.126E11 0.000000 159.638 159.638
9 5.400E+02 9.406E07 0.000008 90.005 90.005
total harmonic distortion = 58.925539 percent
Fourier components of transient response v(0,7)
dc component = 6.093E08
harmonic frequency Fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 1.453E04 1.000000 60.018 0.000
2 1.200E+02 6.263E08 0.000431 91.206 31.188
3 1.800E+02 5.000E+01 344147.7879 180.000 240.018
4 2.400E+02 4.210E07 0.002898 21.103 81.121
5 3.000E+02 3.023E04 2.080596 119.981 59.963
6 3.600E+02 1.138E07 0.000783 15.882 44.136
7 4.200E+02 4.234E04 2.913955 59.993 0.025
8 4.800E+02 2.001E07 0.001378 35.584 24.434
9 5.400E+02 5.000E+01 344147.4728 179.999 240.017
total harmonic distortion = ************ percent
Fourier components of transient response v(8,7)
dc component = 6.070E08
harmonic frequency Fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 1.198E+02 1.000000 0.000 0.000
2 1.200E+02 6.231E08 0.000000 90.473 90.473
3 1.800E+02 5.000E+01 0.417500 180.000 180.000
4 2.400E+02 4.278E07 0.000000 19.747 19.747
5 3.000E+02 9.995E02 0.000835 179.850 179.850
6 3.600E+02 1.023E07 0.000000 13.485 13.485
7 4.200E+02 9.959E02 0.000832 179.790 179.789
8 4.800E+02 1.991E07 0.000000 35.462 35.462
9 5.400E+02 5.000E+01 0.417499 179.999 179.999
total harmonic distortion = 59.043467 percent
10.7. HARMONICS IN POLYPHASE POWER SYSTEMS 335
voltage between the two “Y” centerpoints, between which the neutral conductor used to con
nect. According to SPICE, there is 50 volts of both 3rd and 9th harmonic frequency between
these two points, which is definitely not normal in a linear (no harmonics), balanced Y system.
Finally, the voltage as measured across one of the load’s phases (between nodes 8 and 7 in the
SPICE analysis) likewise shows strong triplen harmonic voltages of 50 volts each.
Figure 10.65 is a graphical summary of the aforementioned effects.
Source Load
ω
3ω
5ω
7ω
9ω
ω
3ω
5ω
7ω
9ω
ω
3ω
5ω
7ω
9ω
ω 5ω 7ω
ω 5ω 7ω
ω 5ω 7ω
line
line
line
3ω 9ω
V
V 3ω 9ω
Figure 10.65: Threewire “YY” (no neutral) system: Triplen voltages appear between “Y” cen
ters. Triplen voltages appear across load phases. Nontriplen currents appear in line conduc
tors.
In summary, removal of the neutral conductor leads to a “hot” centerpoint on the load “Y”,
and also to harmonic load phase voltages of equal magnitude, all comprised of triplen frequen
cies. In the previous simulation where we had a 4wire, Yconnected system, the undesirable
effect from harmonics was excessive neutral current, but at least each phase of the load re
ceived voltage nearly free of harmonics.
Since removing the neutral wire didn’t seem to work in eliminating the problems caused
by harmonics, perhaps switching to a ∆ configuration will. Let’s try a ∆ source instead of
a Y, keeping the load in its present Y configuration, and see what happens. The measured
parameters will be line current (voltage across Rline, nodes 0 and 8), load phase voltage (nodes
8 and 7), and source phase current (voltage across Rsource, nodes 1 and 2). (Figure 10.66)
Note: the following paragraph is for those curious readers who follow every detail of my
SPICE netlists. If you just want to find out what happens in the circuit, skip this paragraph!
When simulating circuits having AC sources of differing frequency and differing phase, the
only way to do it in SPICE is to set up the sources with a delay time or phase offset specified
in seconds. Thus, the 0o source has these five specifying figures: “(0 207.846 60 0 0)”, which
means 0 volts DC offset, 207.846 volts peak amplitude (120 times the square root of three, to
ensure the load phase voltages remain at 120 volts each), 60 Hz, 0 time delay, and 0 damping
factor. The 120o phaseshifted source has these figures: “(0 207.846 60 5.55555m 0)”, all the
same as the first except for the time delay factor of 5.55555 milliseconds, or 1/3 of the full
336 CHAPTER 10. POLYPHASE AC CIRCUITS
DeltaY source/load with harmonics
*
* phase1 voltage source and r (120 v / 0 deg)
vsource1 1 0 sin(0 207.846 60 0 0)
rsource1 1 2 1
*
* phase2 voltage source and r (120 v / 120 deg)
vsource2 3 2 sin(0 207.846 60 5.55555m 0)
rsource2 3 4 1
*
* phase3 voltage source and r (120 v / 240 deg)
vsource3 5 4 sin(0 207.846 60 11.1111m 0)
rsource3 5 0 1
*
* line resistances
rline1 0 8 1
rline2 2 9 1
rline3 4 10 1
*
* phase 1 of load
rload1 8 7 1k
i3har1 8 7 sin(0 50m 180 9.72222m 0)
i5har1 8 7 sin(0 50m 300 9.72222m 0)
i7har1 8 7 sin(0 50m 420 9.72222m 0)
i9har1 8 7 sin(0 50m 540 9.72222m 0)
*
* phase 2 of load
rload2 9 7 1k
i3har2 9 7 sin(0 50m 180 15.2777m 0)
i5har2 9 7 sin(0 50m 300 15.2777m 0)
i7har2 9 7 sin(0 50m 420 15.2777m 0)
i9har2 9 7 sin(0 50m 540 15.2777m 0)
*
* phase 3 of load
rload3 10 7 1k
i3har3 10 7 sin(0 50m 180 4.16666m 0)
i5har3 10 7 sin(0 50m 300 4.16666m 0)
i7har3 10 7 sin(0 50m 420 4.16666m 0)
i9har3 10 7 sin(0 50m 540 4.16666m 0)
*
* analysis stuff
.options itl5=0
.tran 0.5m 100m 16m 1u
.plot tran v(0,8) v(8,7) v(1,2)
.four 60 v(0,8) v(8,7) v(1,2)
.end
10.7. HARMONICS IN POLYPHASE POWER SYSTEMS 337
Rload1 kΩ
50 mA
180 Hz
50 mA
300 Hz
50 mA
420 Hz
50 mA
540 Hz
Rload1 kΩ
50 mA
180 Hz
50 mA
300 Hz
50 mA
420 Hz
50 mA
540 Hz
Rload1 kΩ
50 mA
180 Hz
50 mA
300 Hz
50 mA
420 Hz
50 mA
540 Hz
Vsource Rsource
120 V
1 Ω
60 Hz
Rline
1 Ω
Rsource
120 V
1 Ω
60 Hz
120 V
1 Ω
60 Hz
Rline
1 Ω
Rline
1 Ω
0 1 2
3
4
5 120
o
0o
240o
7
8 9
10
Vsource
Vsource
Rsource
0
4
4
4
2
Figure 10.66: DeltaY source/load with harmonics
period of 16.6667 milliseconds for a 60 Hz waveform. The 240o source must be timedelayed
twice that amount, equivalent to a fraction of 240/360 of 16.6667 milliseconds, or 11.1111
milliseconds. This is for the ∆connected source. The Yconnected load, on the other hand,
requires a different set of timedelay figures for its harmonic current sources, because the
phase voltages in a Y load are not in phase with the phase voltages of a ∆ source. If ∆ source
voltages VAC , VBA, and VCB are referenced at 0o, 120o, and 240o, respectively, then “Y” load
voltages VA, VB , and VC will have phase angles of 30o, 90o, and 210o, respectively. This is an
intrinsic property of all ∆Y circuits and not a quirk of SPICE. Therefore, when I specified the
delay times for the harmonic sources, I had to set them at 15.2777 milliseconds (30o, or +330o),
4.16666 milliseconds (90o), and 9.72222 milliseconds (210o). One final note: when delaying AC
sources in SPICE, they don’t “turn on” until their delay time has elapsed, which means any
mathematical analysis up to that point in time will be in error. Consequently, I set the .tran
transient analysis line to hold off analysis until 16 milliseconds after start, which gives all
sources in the netlist time to engage before any analysis takes place.
The result of this analysis is almost as disappointing as the last. (Figure 10.67) Line cur
rents remain unchanged (the only substantial harmonic content being the 5th and 7th harmon
ics), and load phase voltages remain unchanged as well, with a full 50 volts of triplen harmonic
(3rd and 9th) frequencies across each load component. Source phase current is a fraction of the
line current, which should come as no surprise. Both 5th and 7th harmonics are represented
there, with negligible triplen harmonics:
Fourier analysis of line current:
Fourier analysis of load phase voltage:
Fourier analysis of source phase current:
Really, the only advantage of the ∆Y configuration from the standpoint of harmonics is
that there is no longer a centerpoint at the load posing a shock hazard. Otherwise, the load
components receive the same harmonicallyrich voltages and the lines see the same currents
as in a threewire Y system.
338 CHAPTER 10. POLYPHASE AC CIRCUITS
Fourier components of transient response v(0,8)
dc component = 6.850E11
harmonic frequency Fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 1.198E01 1.000000 150.000 0.000
2 1.200E+02 2.491E11 0.000000 159.723 9.722
3 1.800E+02 1.506E06 0.000013 0.005 149.996
4 2.400E+02 2.033E11 0.000000 52.772 97.228
5 3.000E+02 4.994E02 0.416682 30.002 119.998
6 3.600E+02 1.234E11 0.000000 57.802 92.198
7 4.200E+02 4.993E02 0.416644 29.998 179.998
8 4.800E+02 8.024E11 0.000000 174.200 324.200
9 5.400E+02 4.518E06 0.000038 179.995 329.995
total harmonic distortion = 58.925038 percent
Fourier components of transient response v(8,7)
dc component = 1.259E08
harmonic frequency Fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 1.198E+02 1.000000 150.000 0.000
2 1.200E+02 1.941E07 0.000000 49.693 100.307
3 1.800E+02 5.000E+01 0.417222 89.998 239.998
4 2.400E+02 1.519E07 0.000000 66.397 83.603
5 3.000E+02 6.466E02 0.000540 151.112 301.112
6 3.600E+02 2.433E07 0.000000 68.162 81.838
7 4.200E+02 6.931E02 0.000578 148.548 1.453
8 4.800E+02 2.398E07 0.000000 174.897 324.897
9 5.400E+02 5.000E+01 0.417221 90.006 59.995
total harmonic distortion = 59.004109 percent
Fourier components of transient response v(1,2)
dc component = 3.564E11
harmonic frequency Fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 6.906E02 1.000000 0.181 0.000
2 1.200E+02 1.525E11 0.000000 156.674 156.493
3 1.800E+02 1.422E06 0.000021 179.996 179.815
4 2.400E+02 2.949E11 0.000000 110.570 110.390
5 3.000E+02 2.883E02 0.417440 179.996 179.815
6 3.600E+02 2.324E11 0.000000 91.926 91.745
7 4.200E+02 2.883E02 0.417398 179.994 179.813
8 4.800E+02 4.140E11 0.000000 39.875 39.694
9 5.400E+02 4.267E06 0.000062 0.006 0.186
total harmonic distortion = 59.031969 percent
10.7. HARMONICS IN POLYPHASE POWER SYSTEMS 339
Source Load
ω
3ω
5ω
7ω
9ω
ω
3ω
5ω
7ω
9ω
ω
3ω
5ω
7ω
9ω
ω 5ω 7ω
ω 5ω 7ω
ω 5ω 7ω
line
line
line
3ω 9ω V
ω 5ω 7ωω 5ω 7ω
ω 5ω 7ω
Figure 10.67: “∆Y” source/load: Triplen voltages appear across load phases. Nontriplen cur
rents appear in line conductors and in source phase windings.
If we were to reconfigure the system into a ∆∆ arrangement, (Figure 10.68) that should
guarantee that each load component receives nonharmonic voltage, since each load phase
would be directly connected in parallel with each source phase. The complete lack of any
neutral wires or “center points” in a ∆∆ system prevents strange voltages or additive cur
rents from occurring. It would seem to be the ideal solution. Let’s simulate and observe,
analyzing line current, load phase voltage, and source phase current. See SPICE listing:
“DeltaDelta source/load with harmonics”, “Fourier analysis: Fourier components of transient
response v(0,6)”, and “Fourier components of transient response v(2,1)”.
Fourier analysis of line current:
Fourier analysis of load phase voltage:
Fourier analysis of source phase current:
As predicted earlier, the load phase voltage is almost a pure sinewave, with negligible
harmonic content, thanks to the direct connection with the source phases in a ∆∆ system.
But what happened to the triplen harmonics? The 3rd and 9th harmonic frequencies don’t
appear in any substantial amount in the line current, nor in the load phase voltage, nor in the
source phase current! We know that triplen currents exist, because the 3rd and 9th harmonic
current sources are intentionally placed in the phases of the load, but where did those currents
go?
Remember that the triplen harmonics of 120o phaseshifted fundamental frequencies are
in phase with each other. Note the directions that the arrows of the current sources within
the load phases are pointing, and think about what would happen if the 3rd and 9th harmonic
sources were DC sources instead. What we would have is current circulating within the loop
formed by the ∆connected phases. This is where the triplen harmonic currents have gone: they
stay within the ∆ of the load, never reaching the line conductors or the windings of the source.
These results may be graphically summarized as such in Figure 10.69.
This is a major benefit of the ∆∆ system configuration: triplen harmonic currents remain
340 CHAPTER 10. POLYPHASE AC CIRCUITS
Rload
1 kΩ
50 mA
180 Hz
50 mA
300 Hz
50 mA
420 Hz
50 mA
540 Hz
Rload1 kΩ
50 mA
180 Hz
50 mA
300 Hz
50 mA
420 Hz
50 mA
540 Hz
Rload1 kΩ
50 mA
180 Hz
50 mA
300 Hz
50 mA
420 Hz
50 mA
540 Hz
Vsource Rsource
120 V
1 Ω
60 Hz
Rline
1 Ω
Rsource
120 V
1 Ω
60 Hz
120 V
1 Ω
60 Hz
Rline
1 Ω
Rline
1 Ω
0 1 2
3
4
5 120
o
0o
240o
Vsource
Vsource
Rsource
0
4
4
4
2 6 7
8
Figure 10.68: DeltaDelta source/load with harmonics.
Source Load
ω
3ω
5ω
7ω
9ω
ω 5ω 7ω
ω 5ω 7ω
ω 5ω 7ω
line
line
line
ω 5ω 7ωω 5ω 7ω
ω 5ω 7ω
9ω3ω
9ω3ω
9ω3ω
ω 5ω 7ω3ω 9ω
ω
5ω
7ω
3ω
9ωVω
Figure 10.69: ∆∆ source/load: Load phases receive undistorted sinewave voltages. Triplen
currents are confined to circulate within load phases. Nontriplen currents apprear in line
conductors and in source phase windings.
10.7. HARMONICS IN POLYPHASE POWER SYSTEMS 341
DeltaDelta source/load with harmonics
*
* phase1 voltage source and r (120 v / 0 deg)
vsource1 1 0 sin(0 120 60 0 0)
rsource1 1 2 1
*
* phase2 voltage source and r (120 v / 120 deg)
vsource2 3 2 sin(0 120 60 5.55555m 0)
rsource2 3 4 1
*
* phase3 voltage source and r (120 v / 240 deg)
vsource3 5 4 sin(0 120 60 11.1111m 0)
rsource3 5 0 1
*
* line resistances
rline1 0 6 1
rline2 2 7 1
rline3 4 8 1
*
* phase 1 of load
rload1 7 6 1k
i3har1 7 6 sin(0 50m 180 0 0)
i5har1 7 6 sin(0 50m 300 0 0)
i7har1 7 6 sin(0 50m 420 0 0)
i9har1 7 6 sin(0 50m 540 0 0)
*
* phase 2 of load
rload2 8 7 1k
i3har2 8 7 sin(0 50m 180 5.55555m 0)
i5har2 8 7 sin(0 50m 300 5.55555m 0)
i7har2 8 7 sin(0 50m 420 5.55555m 0)
i9har2 8 7 sin(0 50m 540 5.55555m 0)
*
* phase 3 of load
rload3 6 8 1k
i3har3 6 8 sin(0 50m 180 11.1111m 0)
i5har3 6 8 sin(0 50m 300 11.1111m 0)
i7har3 6 8 sin(0 50m 420 11.1111m 0)
i9har3 6 8 sin(0 50m 540 11.1111m 0)
*
* analysis stuff
.options itl5=0
.tran 0.5m 100m 16m 1u
.plot tran v(0,6) v(7,6) v(2,1) i(3har1)
.four 60 v(0,6) v(7,6) v(2,1)
.end
342 CHAPTER 10. POLYPHASE AC CIRCUITS
Fourier components of transient response v(0,6)
dc component = 6.007E11
harmonic frequency Fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 2.070E01 1.000000 150.000 0.000
2 1.200E+02 5.480E11 0.000000 156.666 6.666
3 1.800E+02 6.257E07 0.000003 89.990 60.010
4 2.400E+02 4.911E11 0.000000 8.187 141.813
5 3.000E+02 8.626E02 0.416664 149.999 300.000
6 3.600E+02 1.089E10 0.000000 31.997 181.997
7 4.200E+02 8.626E02 0.416669 150.001 0.001
8 4.800E+02 1.578E10 0.000000 63.940 213.940
9 5.400E+02 1.877E06 0.000009 89.987 60.013
total harmonic distortion = 58.925538 percent
Fourier components of transient response v(7,6)
dc component = 5.680E10
harmonic frequency Fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 1.195E+02 1.000000 0.000 0.000
2 1.200E+02 1.039E09 0.000000 144.749 144.749
3 1.800E+02 1.251E06 0.000000 89.974 89.974
4 2.400E+02 4.215E10 0.000000 36.127 36.127
5 3.000E+02 1.992E01 0.001667 180.000 180.000
6 3.600E+02 2.499E09 0.000000 4.760 4.760
7 4.200E+02 1.992E01 0.001667 180.000 180.000
8 4.800E+02 2.951E09 0.000000 151.385 151.385
9 5.400E+02 3.752E06 0.000000 89.905 89.905
total harmonic distortion = 0.235702 percent
Fourier components of transient response v(2,1)
dc component = 1.923E12
harmonic frequency Fourier normalized phase normalized
no (hz) component component (deg) phase (deg)
1 6.000E+01 1.194E01 1.000000 179.940 0.000
2 1.200E+02 2.569E11 0.000000 133.491 46.449
3 1.800E+02 3.129E07 0.000003 89.985 89.955
4 2.400E+02 2.657E11 0.000000 23.368 156.571
5 3.000E+02 4.980E02 0.416918 180.000 359.939
6 3.600E+02 4.595E11 0.000000 22.475 202.415
7 4.200E+02 4.980E02 0.416921 180.000 359.939
8 4.800E+02 7.385E11 0.000000 63.759 243.699
9 5.400E+02 9.385E07 0.000008 89.991 89.949
total harmonic distortion = 58.961298 percent
10.8. HARMONIC PHASE SEQUENCES 343
confined in whatever set of components create them, and do not “spread” to other parts of the
system.
• REVIEW:
• Nonlinear components are those that draw a nonsinusoidal (nonsinewave) current wave
form when energized by a sinusoidal (sinewave) voltage. Since any distortion of an
originally pure sinewave constitutes harmonic frequencies, we can say that nonlinear
components generate harmonic currents.
• When the sinewave distortion is symmetrical above and below the average centerline of
the waveform, the only harmonics present will be oddnumbered, not evennumbered.
• The 3rd harmonic, and integer multiples of it (6th, 9th, 12th, 15th) are known as triplen
harmonics. They are in phase with each other, despite the fact that their respective
fundamental waveforms are 120o out of phase with each other.
• In a 4wire YY system, triplen harmonic currents add within the neutral conductor.
• Triplen harmonic currents in a ∆connected set of components circulate within the loop
formed by the ∆.
10.8 Harmonic phase sequences
In the last section, we saw how the 3rd harmonic and all of its integer multiples (collectively
called triplen harmonics) generated by 120o phaseshifted fundamental waveforms are actually
in phase with each other. In a 60 Hz threephase power system, where phases A, B, and C
are 120o apart, the thirdharmonic multiples of those frequencies (180 Hz) fall perfectly into
phase with each other. This can be thought of in graphical terms, (Figure 10.70) and/or in
mathematical terms:
TIME
A B C
Figure 10.70: Harmonic currents of Phases A, B, C all coincide, that is, no rotation.
344 CHAPTER 10. POLYPHASE AC CIRCUITS
Fundamental A B C0o 120o 240o
A’ B’ C’
3 x 0o 3 x 120o 3 x 240o
(0o) (360o = 0o) (720o = 0o)
Phase sequence = ABC
3rd harmonic
If we extend the mathematical table to include higher oddnumbered harmonics, we will
notice an interesting pattern develop with regard to the rotation or sequence of the harmonic
frequencies:
Fundamental A B C0o 120o 240o
A’ B’ C’
3 x 0o 3 x 120o 3 x 240o
(0o) (360o = 0o) (720o = 0o)
ABC
no
rotation
5th harmonic
A’’ B’’ C’’
5 x 0o 5 x 120o 5 x 240o
(0o) (600o = 720o  120o)(120o)
(1200o = 1440o  240o)
(240o)
CBA
7th harmonic
A’’’ B’’’ C’’’
7 x 0o
(0o)
7 x 120o 7 x 240o
(840o = 720o + 120o)
(120o)
(1680o = 1440o + 240o)
(240o)
ABC
A’’’’ B’’’’ C’’’’
9 x 0o
(0o)
9 x 120o
(1080o = 0o)
9 x 240o
(2160o = 0o)
no
rotation
3rd harmonic
9th harmonic
Harmonics such as the 7th, which “rotate” with the same sequence as the fundamental, are
called positive sequence. Harmonics such as the 5th, which “rotate” in the opposite sequence
as the fundamental, are called negative sequence. Triplen harmonics (3rd and 9th shown in
this table) which don’t “rotate” at all because they’re in phase with each other, are called zero
sequence.
This pattern of positivezeronegativepositive continues indefinitely for all oddnumbered
harmonics, lending itself to expression in a table like this:
10.9. CONTRIBUTORS 345
+

0
1st
3rd
5th
7th
9th
11th
13th
15th
17th
19th
21st
23rd
Rotation sequences according
to harmonic number
Rotates with fundamental
Rotates against fundamental
Does not rotate
Sequence especially matters when we’re dealing with AC motors, since the mechanical ro
tation of the rotor depends on the torque produced by the sequential “rotation” of the applied
3phase power. Positivesequence frequencies work to push the rotor in the proper direction,
whereas negativesequence frequencies actually work against the direction of the rotor’s rota
tion. Zerosequence frequencies neither contribute to nor detract from the rotor’s torque. An
excess of negativesequence harmonics (5th, 11th, 17th, and/or 23rd) in the power supplied to
a threephase AC motor will result in a degradation of performance and possible overheating.
Since the higherorder harmonics tend to be attenuated more by system inductances and mag
netic core losses, and generally originate with less amplitude anyway, the primary harmonic
of concern is the 5th, which is 300 Hz in 60 Hz power systems and 250 Hz in 50 Hz power
systems.
10.9 Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
Ed Beroset (May 6, 2002): Suggested better ways to illustrate the meaning of the prefix
“poly”.
Jason Starck (June 2000): HTML document formatting, which led to a much better
looking second edition.
346 CHAPTER 10. POLYPHASE AC CIRCUITS
Chapter 11
POWER FACTOR
Contents
11.1 Power in resistive and reactive AC circuits . . . . . . . . . . . . . . . . . . 347
11.2 True, Reactive, and Apparent power . . . . . . . . . . . . . . . . . . . . . . . 352
11.3 Calculating power factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
11.4 Practical power factor correction . . . . . . . . . . . . . . . . . . . . . . . . 360
11.5 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365
11.1 Power in resistive and reactive AC circuits
Consider a circuit for a singlephase AC power system, where a 120 volt, 60 Hz AC voltage
source is delivering power to a resistive load: (Figure 11.1)
120 V
60 Hz R 60 Ω
Figure 11.1: Ac source drives a purely resistive load.
347
348 CHAPTER 11. POWER FACTOR
ZR = 60 + j0 Ω or 60 Ω ∠ 0o
I = E
Z
I = 120 V
60 Ω
I = 2 A
In this example, the current to the load would be 2 amps, RMS. The power dissipated at
the load would be 240 watts. Because this load is purely resistive (no reactance), the current
is in phase with the voltage, and calculations look similar to that in an equivalent DC circuit.
If we were to plot the voltage, current, and power waveforms for this circuit, it would look like
Figure 11.2.
Time
+

e =
i =
p =
Figure 11.2: Current is in phase with voltage in a resistive circuit.
Note that the waveform for power is always positive, never negative for this resistive circuit.
This means that power is always being dissipated by the resistive load, and never returned to
the source as it is with reactive loads. If the source were a mechanical generator, it would take
240 watts worth of mechanical energy (about 1/3 horsepower) to turn the shaft.
Also note that the waveform for power is not at the same frequency as the voltage or cur
rent! Rather, its frequency is double that of either the voltage or current waveforms. This
different frequency prohibits our expression of power in an AC circuit using the same complex
(rectangular or polar) notation as used for voltage, current, and impedance, because this form
of mathematical symbolism implies unchanging phase relationships. When frequencies are
not the same, phase relationships constantly change.
As strange as it may seem, the best way to proceed with AC power calculations is to use
scalar notation, and to handle any relevant phase relationships with trigonometry.
For comparison, let’s consider a simple AC circuit with a purely reactive load in Figure 11.3.
11.1. POWER IN RESISTIVE AND REACTIVE AC CIRCUITS 349
120 V
60 Hz L 160 mH
Figure 11.3: AC circuit with a purely reactive (inductive) load.
I = E
Z
I = 120 V
XL = 60.319 Ω
ZL = 0 + j60.319 Ω or 60.319 Ω ∠ 90o
60.319 Ω
I = 1.989 A
Time
+

e =
i =
p =
Figure 11.4: Power is not dissipated in a purely reactive load. Though it is alternately absorbed
from and returned to the source.
Note that the power alternates equally between cycles of positive and negative. (Fig
ure 11.4) This means that power is being alternately absorbed from and returned to the source.
If the source were a mechanical generator, it would take (practically) no net mechanical energy
to turn the shaft, because no power would be used by the load. The generator shaft would be
easy to spin, and the inductor would not become warm as a resistor would.
Now, let’s consider an AC circuit with a load consisting of both inductance and resistance
in Figure 11.5.
350 CHAPTER 11. POWER FACTOR
120 V
60 Hz
Lload
Rload
Load
160 mH
60 Ω
Figure 11.5: AC circuit with both reactance and resistance.
I = E
Z
I = 120 V
XL = 60.319 Ω
ZL = 0 + j60.319 Ω or 60.319 Ω ∠ 90o
ZR = 60 + j0 Ω or 60 Ω ∠ 0o
Ztotal = 60 + j60.319 Ω or 85.078 Ω ∠ 45.152o
85.078 Ω
I = 1.410 A
At a frequency of 60 Hz, the 160 millihenrys of inductance gives us 60.319 Ω of inductive
reactance. This reactance combines with the 60 Ω of resistance to form a total load impedance
of 60 + j60.319 Ω, or 85.078 Ω 6 45.152o. If we’re not concerned with phase angles (which we’re
not at this point), we may calculate current in the circuit by taking the polar magnitude of
the voltage source (120 volts) and dividing it by the polar magnitude of the impedance (85.078
Ω). With a power supply voltage of 120 volts RMS, our load current is 1.410 amps. This is the
figure an RMS ammeter would indicate if connected in series with the resistor and inductor.
We already know that reactive components dissipate zero power, as they equally absorb
power from, and return power to, the rest of the circuit. Therefore, any inductive reactance in
this load will likewise dissipate zero power. The only thing left to dissipate power here is the
11.1. POWER IN RESISTIVE AND REACTIVE AC CIRCUITS 351
resistive portion of the load impedance. If we look at the waveform plot of voltage, current, and
total power for this circuit, we see how this combination works in Figure 11.6.
Time
+

e =
i =
p =
Figure 11.6: A combined resistive/reactive circuit dissipates more power than it returns to the
source. The reactance dissipates no power; though, the resistor does.
As with any reactive circuit, the power alternates between positive and negative instan
taneous values over time. In a purely reactive circuit that alternation between positive and
negative power is equally divided, resulting in a net power dissipation of zero. However, in
circuits with mixed resistance and reactance like this one, the power waveform will still alter
nate between positive and negative, but the amount of positive power will exceed the amount
of negative power. In other words, the combined inductive/resistive load will consume more
power than it returns back to the source.
Looking at the waveform plot for power, it should be evident that the wave spends more
time on the positive side of the center line than on the negative, indicating that there is more
power absorbed by the load than there is returned to the circuit. What little returning of power
that occurs is due to the reactance; the imbalance of positive versus negative power is due to
the resistance as it dissipates energy outside of the circuit (usually in the form of heat). If the
source were a mechanical generator, the amount of mechanical energy needed to turn the shaft
would be the amount of power averaged between the positive and negative power cycles.
Mathematically representing power in an AC circuit is a challenge, because the power wave
isn’t at the same frequency as voltage or current. Furthermore, the phase angle for power
means something quite different from the phase angle for either voltage or current. Whereas
the angle for voltage or current represents a relative shift in timing between two waves, the
phase angle for power represents a ratio between power dissipated and power returned. Be
cause of this way in which AC power differs from AC voltage or current, it is actually easier to
arrive at figures for power by calculating with scalar quantities of voltage, current, resistance,
and reactance than it is to try to derive it from vector, or complex quantities of voltage, current,
and impedance that we’ve worked with so far.
• REVIEW:
• In a purely resistive circuit, all circuit power is dissipated by the resistor(s). Voltage and
current are in phase with each other.
• In a purely reactive circuit, no circuit power is dissipated by the load(s). Rather, power
is alternately absorbed from and returned to the AC source. Voltage and current are 90o
out of phase with each other.
352 CHAPTER 11. POWER FACTOR
• In a circuit consisting of resistance and reactance mixed, there will be more power dissi
pated by the load(s) than returned, but some power will definitely be dissipated and some
will merely be absorbed and returned. Voltage and current in such a circuit will be out of
phase by a value somewhere between 0o and 90o.
11.2 True, Reactive, and Apparent power
We know that reactive loads such as inductors and capacitors dissipate zero power, yet the
fact that they drop voltage and draw current gives the deceptive impression that they actually
do dissipate power. This “phantom power” is called reactive power, and it is measured in a
unit called VoltAmpsReactive (VAR), rather than watts. The mathematical symbol for reac
tive power is (unfortunately) the capital letter Q. The actual amount of power being used, or
dissipated, in a circuit is called true power, and it is measured in watts (symbolized by the cap
ital letter P, as always). The combination of reactive power and true power is called apparent
power, and it is the product of a circuit’s voltage and current, without reference to phase angle.
Apparent power is measured in the unit of VoltAmps (VA) and is symbolized by the capital
letter S.
As a rule, true power is a function of a circuit’s dissipative elements, usually resistances
(R). Reactive power is a function of a circuit’s reactance (X). Apparent power is a function of a
circuit’s total impedance (Z). Since we’re dealing with scalar quantities for power calculation,
any complex starting quantities such as voltage, current, and impedance must be represented
by their polar magnitudes, not by real or imaginary rectangular components. For instance, if
I’m calculating true power from current and resistance, I must use the polar magnitude for
current, and not merely the “real” or “imaginary” portion of the current. If I’m calculating
apparent power from voltage and impedance, both of these formerly complex quantities must
be reduced to their polar magnitudes for the scalar arithmetic.
There are several power equations relating the three types of power to resistance, reactance,
and impedance (all using scalar quantities):
11.2. TRUE, REACTIVE, AND APPARENT POWER 353
P = true power P = I2R P = E
2
R
Q = reactive power E
2
X
Measured in units of Watts
Measured in units of VoltAmpsReactive (VAR)
S = apparent power
Q =Q = I2X
S = I2Z E
2
S =
Z
S = IE
Measured in units of VoltAmps (VA)
Please note that there are two equations each for the calculation of true and reactive power.
There are three equations available for the calculation of apparent power, P=IE being useful
only for that purpose. Examine the following circuits and see how these three types of power
interrelate for: a purely resistive load in Figure 11.7, a purely reactive load in Figure 11.8, and
a resistive/reactive load in Figure 11.9.
Resistive load only:
120 V
60 Hz Rnoreactance
I = 2 A
60 Ω
P = true power = I2R = 240 W
Q = reactive power = I2X = 0 VAR
S = apparent power = I2Z = 240 VA
Figure 11.7: True power, reactive power, and apparent power for a purely resistive load.
Reactive load only:
Resistive/reactive load:
These three types of power – true, reactive, and apparent – relate to one another in trigono
metric form. We call this the power triangle: (Figure 11.10).
Using the laws of trigonometry, we can solve for the length of any side (amount of any type
of power), given the lengths of the other two sides, or the length of one side and an angle.
354 CHAPTER 11. POWER FACTOR
120 V
60 Hz L
160 mH
I = 1.989 A
XL = 60.319 Ω
no
resistance
P = true power = I2R = 0 W
Q = reactive power = I2X = 238.73 VAR
S = apparent power = I2Z = 238.73 VA
Figure 11.8: True power, reactive power, and apparent power for a purely reactive load.
120 V
60 Hz
160 mHLload
Rload
LoadI = 1.410 A
XL = 60.319 Ω
60 Ω
P = true power = I2R = 119.365 W
Q = reactive power = I2X = 119.998 VAR
S = apparent power = I2Z = 169.256 VA
Figure 11.9: True power, reactive power, and apparent power for a resistive/reactive load.
11.3. CALCULATING POWER FACTOR 355
True power (P)
Reactive power (Q)
Apparent power (S)
phase angle
measured in Watts
measured in VA
measured in VAR
The "Power Triangle"
Impedance
Figure 11.10: Power triangle relating appearant power to true power and reactive power.
• REVIEW:
• Power dissipated by a load is referred to as true power. True power is symbolized by the
letter P and is measured in the unit of Watts (W).
• Power merely absorbed and returned in load due to its reactive properties is referred to
as reactive power. Reactive power is symbolized by the letter Q and is measured in the
unit of VoltAmpsReactive (VAR).
• Total power in an AC circuit, both dissipated and absorbed/returned is referred to as
apparent power. Apparent power is symbolized by the letter S and is measured in the
unit of VoltAmps (VA).
• These three types of power are trigonometrically related to one another. In a right trian
gle, P = adjacent length, Q = opposite length, and S = hypotenuse length. The opposite
angle is equal to the circuit’s impedance (Z) phase angle.
11.3 Calculating power factor
As was mentioned before, the angle of this “power triangle” graphically indicates the ratio
between the amount of dissipated (or consumed) power and the amount of absorbed/returned
power. It also happens to be the same angle as that of the circuit’s impedance in polar form.
When expressed as a fraction, this ratio between true power and apparent power is called the
power factor for this circuit. Because true power and apparent power form the adjacent and
356 CHAPTER 11. POWER FACTOR
hypotenuse sides of a right triangle, respectively, the power factor ratio is also equal to the
cosine of that phase angle. Using values from the last example circuit:
Power factor = True power
Apparent power
Power factor = 119.365 W
169.256 VA
Power factor = 0.705
cos 45.152o = 0.705
It should be noted that power factor, like all ratio measurements, is a unitless quantity.
For the purely resistive circuit, the power factor is 1 (perfect), because the reactive power
equals zero. Here, the power triangle would look like a horizontal line, because the opposite
(reactive power) side would have zero length.
For the purely inductive circuit, the power factor is zero, because true power equals zero.
Here, the power triangle would look like a vertical line, because the adjacent (true power) side
would have zero length.
The same could be said for a purely capacitive circuit. If there are no dissipative (resistive)
components in the circuit, then the true power must be equal to zero, making any power in
the circuit purely reactive. The power triangle for a purely capacitive circuit would again be a
vertical line (pointing down instead of up as it was for the purely inductive circuit).
Power factor can be an important aspect to consider in an AC circuit, because any power
factor less than 1 means that the circuit’s wiring has to carry more current than what would
be necessary with zero reactance in the circuit to deliver the same amount of (true) power to
the resistive load. If our last example circuit had been purely resistive, we would have been
able to deliver a full 169.256 watts to the load with the same 1.410 amps of current, rather
than the mere 119.365 watts that it is presently dissipating with that same current quantity.
The poor power factor makes for an inefficient power delivery system.
Poor power factor can be corrected, paradoxically, by adding another load to the circuit
drawing an equal and opposite amount of reactive power, to cancel out the effects of the load’s
inductive reactance. Inductive reactance can only be canceled by capacitive reactance, so we
have to add a capacitor in parallel to our example circuit as the additional load. The effect of
these two opposing reactances in parallel is to bring the circuit’s total impedance equal to its
total resistance (to make the impedance phase angle equal, or at least closer, to zero).
Since we know that the (uncorrected) reactive power is 119.998 VAR (inductive), we need to
calculate the correct capacitor size to produce the same quantity of (capacitive) reactive power.
Since this capacitor will be directly in parallel with the source (of known voltage), we’ll use the
power formula which starts from voltage and reactance:
11.3. CALCULATING POWER FACTOR 357
E2
X
. . . solving for X . . .
X = E
2
Q =
Q
X = (120 V)
2
119.998 VAR
X = 120.002 Ω
XC = 2pifC
1
. . . solving for C . . .
C =
2pifXC
1
C =
2pi(60 Hz)(120.002 Ω)
1
C = 22.105 µF
Let’s use a rounded capacitor value of 22 µF and see what happens to our circuit: (Fig
ure 11.11)
120 V
60 Hz
160 mHLload
Rload
Load
C
995.257
mA
Iload = 1.41 AIC =
Itotal = 994.716 mA
22 µF
60 Ω
XL = 60.319 Ω
V1
V2
0
1
2
3
Figure 11.11: Parallel capacitor corrects lagging power factor of inductive load. V2 and node
numbers: 0, 1, 2, and 3 are SPICE related, and may be ignored for the moment.
358 CHAPTER 11. POWER FACTOR
Ztotal = ZC // (ZL  ZR)
Ztotal = (120.57 Ω ∠ 90o) // (60.319 Ω ∠ 90o  60 Ω ∠ 0o)
Ztotal = 120.64  j573.58m Ω or 120.64 Ω ∠ 0.2724o
P = true power = I2R = 119.365 W
S = apparent power = I2Z = 119.366 VA
The power factor for the circuit, overall, has been substantially improved. The main current
has been decreased from 1.41 amps to 994.7 milliamps, while the power dissipated at the load
resistor remains unchanged at 119.365 watts. The power factor is much closer to being 1:
Power factor = True power
Apparent power
Power factor = 119.365 W
119.366 VA
Power factor = 0.9999887
Impedance (polar) angle = 0.272o
cos 0.272o = 0.9999887
Since the impedance angle is still a positive number, we know that the circuit, overall,
is still more inductive than it is capacitive. If our power factor correction efforts had been
perfectly ontarget, we would have arrived at an impedance angle of exactly zero, or purely
resistive. If we had added too large of a capacitor in parallel, we would have ended up with
an impedance angle that was negative, indicating that the circuit was more capacitive than
inductive.
A SPICE simulation of the circuit of (Figure 11.11) shows total voltage and total current are
nearly in phase. The SPICE circuit file has a zero volt voltagesource (V2) in series with the
capacitor so that the capacitor current may be measured. The start time of 200 msec ( instead
of 0) in the transient analysis statement allows the DC conditions to stabilize before collecting
data. See SPICE listing “pf.cir power factor”.
The Nutmeg plot of the various currents with respect to the applied voltage Vtotal is shown
in (Figure 11.12). The reference is Vtotal, to which all other measurements are compared. This
is because the applied voltage, Vtotal, appears across the parallel branches of the circuit. There
is no single current common to all components. We can compare those currents to Vtotal.
Note that the total current (Itotal) is in phase with the applied voltage (Vtotal), indicating a
phase angle of near zero. This is no coincidence. Note that the lagging current, IL of the in
ductor would have caused the total current to have a lagging phase somewhere between (Itotal)
11.3. CALCULATING POWER FACTOR 359
pf.cir power factor
V1 1 0 sin(0 170 60)
C1 1 3 22uF
v2 3 0 0
L1 1 2 160mH
R1 2 0 60
# resolution stop start
.tran 1m 200m 160m
.end
Figure 11.12: Zero phase angle due to inphase Vtotal and Itotal . The lagging IL with respect to
Vtotal is corrected by a leading IC .
360 CHAPTER 11. POWER FACTOR
and IL. However, the leading capacitor current, IC , compensates for the lagging inductor cur
rent. The result is a total current phaseangle somewhere between the inductor and capacitor
currents. Moreover, that total current (Itotal) was forced to be inphase with the total applied
voltage (Vtotal), by the calculation of an appropriate capacitor value.
Since the total voltage and current are in phase, the product of these two waveforms, power,
will always be positive throughout a 60 Hz cycle, real power as in Figure 11.2. Had the phase
angle not been corrected to zero (PF=1), the product would have been negative where positive
portions of one waveform overlapped negative portions of the other as in Figure 11.6. Negative
power is fed back to the generator. It cannont be sold; though, it does waste power in the
resistance of electric lines between load and generator. The parallel capacitor corrects this
problem.
Note that reduction of line losses applies to the lines from the generator to the point where
the power factor correction capacitor is applied. In other words, there is still circulating current
between the capacitor and the inductive load. This is not normally a problem because the
power factor correction is applied close to the offending load, like an induction motor.
It should be noted that too much capacitance in an AC circuit will result in a low power
factor just as well as too much inductance. You must be careful not to overcorrect when adding
capacitance to an AC circuit. You must also be very careful to use the proper capacitors for the
job (rated adequately for power system voltages and the occasional voltage spike from lightning
strikes, for continuous AC service, and capable of handling the expected levels of current).
If a circuit is predominantly inductive, we say that its power factor is lagging (because the
current wave for the circuit lags behind the applied voltage wave). Conversely, if a circuit is
predominantly capacitive, we say that its power factor is leading. Thus, our example circuit
started out with a power factor of 0.705 lagging, and was corrected to a power factor of 0.999
lagging.
• REVIEW:
• Poor power factor in an AC circuit may be “corrected”, or reestablished at a value close
to 1, by adding a parallel reactance opposite the effect of the load’s reactance. If the load’s
reactance is inductive in nature (which is almost always will be), parallel capacitance is
what is needed to correct poor power factor.
11.4 Practical power factor correction
When the need arises to correct for poor power factor in an AC power system, you probably
won’t have the luxury of knowing the load’s exact inductance in henrys to use for your calcula
tions. You may be fortunate enough to have an instrument called a power factor meter to tell
you what the power factor is (a number between 0 and 1), and the apparent power (which can
be figured by taking a voltmeter reading in volts and multiplying by an ammeter reading in
amps). In less favorable circumstances you may have to use an oscilloscope to compare voltage
and current waveforms, measuring phase shift in degrees and calculating power factor by the
cosine of that phase shift.
Most likely, you will have access to a wattmeter for measuring true power, whose reading
you can compare against a calculation of apparent power (from multiplying total voltage and
11.4. PRACTICAL POWER FACTOR CORRECTION 361
total current measurements). From the values of true and apparent power, you can deter
mine reactive power and power factor. Let’s do an example problem to see how this works:
(Figure 11.13)
P
wattmeter
A
ammeter
Wattmeter reading = 1.5 kW
Ammeter reading = 9.615 A RMS
240 V
RMS
60 Hz
Load
Figure 11.13: Wattmeter reads true power; product of voltmeter and ammeter readings yields
appearant power.
First, we need to calculate the apparent power in kVA. We can do this by multiplying load
voltage by load current:
S = IE
S = (9.615 A)(240 V)
S = 2.308 kVA
As we can see, 2.308 kVA is a much larger figure than 1.5 kW, which tells us that the power
factor in this circuit is rather poor (substantially less than 1). Now, we figure the power factor
of this load by dividing the true power by the apparent power:
Power factor = P
S
Power factor = 1.5 kW
2.308 kVA
Power factor = 0.65
Using this value for power factor, we can draw a power triangle, and from that determine
the reactive power of this load: (Figure 11.14)
To determine the unknown (reactive power) triangle quantity, we use the Pythagorean The
orem “backwards,” given the length of the hypotenuse (apparent power) and the length of the
adjacent side (true power):
362 CHAPTER 11. POWER FACTOR
True power (P)
Reactive power (Q)
Apparent power (S)
1.5 kW
2.308 kVA
???
Figure 11.14: Reactive power may be calculated from true power and appearant power.
Reactive power = (Apparent power)2  (True power)2
Q = 1.754 kVAR
If this load is an electric motor, or most any other industrial AC load, it will have a lagging
(inductive) power factor, which means that we’ll have to correct for it with a capacitor of appro
priate size, wired in parallel. Now that we know the amount of reactive power (1.754 kVAR),
we can calculate the size of capacitor needed to counteract its effects:
11.4. PRACTICAL POWER FACTOR CORRECTION 363
E2
X
. . . solving for X . . .
X = E
2
Q =
Q
X =
XC = 2pifC
1
. . . solving for C . . .
C =
2pifXC
1
C = 1
(240)2
1.754 kVAR
X = 32.845 Ω
2pi(60 Hz)(32.845 Ω)
C = 80.761 µF
Rounding this answer off to 80 µF, we can place that size of capacitor in the circuit and
calculate the results: (Figure 11.15)
P
wattmeter
A
ammeter
240 V
RMS
60 Hz
C
80 µF
Load
Figure 11.15: Parallel capacitor corrects lagging (inductive) load.
An 80 µF capacitor will have a capacitive reactance of 33.157 Ω, giving a current of 7.238
amps, and a corresponding reactive power of 1.737 kVAR (for the capacitor only). Since the ca
pacitor’s current is 180o out of phase from the the load’s inductive contribution to current draw,
the capacitor’s reactive power will directly subtract from the load’s reactive power, resulting
in:
Inductive kVAR  Capacitive kVAR = Total kVAR
1.754 kVAR  1.737 kVAR = 16.519 VAR
This correction, of course, will not change the amount of true power consumed by the load,
but it will result in a substantial reduction of apparent power, and of the total current drawn
from the 240 Volt source: (Figure 11.16)
364 CHAPTER 11. POWER FACTOR
True power (P)
Reactive power (Q)
Apparent power (S)
1.5 kW
2.308 kVA
1.754 kVAR
(inductive)
1.737 kVAR
(capacitive)
Reactive power (Q)
16.519 VAR
True power (P)
1.5 kW
Apparent power (S)
Power triangle for uncorrected (original) circuit
Power triangle after adding capacitor
Figure 11.16: Power triangle before and after capacitor correction.
11.5. CONTRIBUTORS 365
The new apparent power can be found from the true and new reactive power values, using
the standard form of the Pythagorean Theorem:
Apparent power = (Reactive power)2 + (True power)2
Apparent power = 1.50009 kVA
This gives a corrected power factor of (1.5kW / 1.5009 kVA), or 0.99994, and a new total
current of (1.50009 kVA / 240 Volts), or 6.25 amps, a substantial improvement over the uncor
rected value of 9.615 amps! This lower total current will translate to less heat losses in the
circuit wiring, meaning greater system efficiency (less power wasted).
11.5 Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
Jason Starck (June 2000): HTML document formatting, which led to a much better
looking second edition.
366 CHAPTER 11. POWER FACTOR
Chapter 12
AC METERING CIRCUITS
Contents
12.1 AC voltmeters and ammeters . . . . . . . . . . . . . . . . . . . . . . . . . . . 367
12.2 Frequency and phase measurement . . . . . . . . . . . . . . . . . . . . . . . 374
12.3 Power measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382
12.4 Power quality measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
12.5 AC bridge circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
12.6 AC instrumentation transducers . . . . . . . . . . . . . . . . . . . . . . . . . 396
12.7 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 406
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 406
12.1 AC voltmeters and ammeters
AC electromechanical meter movements come in two basic arrangements: those based on DC
movement designs, and those engineered specifically for AC use. Permanentmagnet moving
coil (PMMC) meter movements will not work correctly if directly connected to alternating cur
rent, because the direction of needle movement will change with each halfcycle of the AC.
(Figure 12.1) Permanentmagnet meter movements, like permanentmagnet motors, are de
vices whose motion depends on the polarity of the applied voltage (or, you can think of it in
terms of the direction of the current).
In order to use a DCstyle meter movement such as the D’Arsonval design, the alternating
current must be rectified into DC. This is most easily accomplished through the use of devices
called diodes. We saw diodes used in an example circuit demonstrating the creation of har
monic frequencies from a distorted (or rectified) sine wave. Without going into elaborate detail
over how and why diodes work as they do, just remember that they each act like a oneway
valve for electrons to flow: acting as a conductor for one polarity and an insulator for another.
Oddly enough, the arrowhead in each diode symbol points against the permitted direction of
electron flow rather than with it as one might expect. Arranged in a bridge, four diodes will
367
368 CHAPTER 12. AC METERING CIRCUITS
wire coil
magnet magnet
"needle"
0
50
100
Figure 12.1: Passing AC through this D’Arsonval meter movement causes useless flutter of the
needle.
serve to steer AC through the meter movement in a constant direction throughout all portions
of the AC cycle: (Figure 12.2)
wire coil
magnet magnet
"needle"
0
50
100
 +
Bridge
rectifier
AC
source
Meter movement needle
will always be driven in
the proper direction.
Figure 12.2: Passing AC through this Rectified AC meter movement will drive it in one direc
tion.
12.1. AC VOLTMETERS AND AMMETERS 369
Another strategy for a practical AC meter movement is to redesign the movement without
the inherent polarity sensitivity of the DC types. This means avoiding the use of permanent
magnets. Probably the simplest design is to use a nonmagnetized iron vane to move the needle
against spring tension, the vane being attracted toward a stationary coil of wire energized by
the AC quantity to be measured as in Figure 12.3.
"needle"
0
50
100
wire coil
iron vane
Figure 12.3: Ironvane electromechanical meter movement.
Electrostatic attraction between two metal plates separated by an air gap is an alternative
mechanism for generating a needlemoving force proportional to applied voltage. This works
just as well for AC as it does for DC, or should I say, just as poorly! The forces involved are very
small, much smaller than the magnetic attraction between an energized coil and an iron vane,
and as such these “electrostatic” meter movements tend to be fragile and easily disturbed by
physical movement. But, for some highvoltage AC applications, the electrostatic movement is
an elegant technology. If nothing else, this technology possesses the advantage of extremely
high input impedance, meaning that no current need be drawn from the circuit under test.
Also, electrostatic meter movements are capable of measuring very high voltages without need
for range resistors or other, external apparatus.
When a sensitive meter movement needs to be reranged to function as an AC voltmeter,
seriesconnected “multiplier” resistors and/or resistive voltage dividers may be employed just
as in DC meter design: (Figure 12.4)
Capacitors may be used instead of resistors, though, to make voltmeter divider circuits.
This strategy has the advantage of being nondissipative (no true power consumed and no heat
produced): (Figure 12.5)
If the meter movement is electrostatic, and thus inherently capacitive in nature, a single
“multiplier” capacitor may be connected in series to give it a greater voltage measuring range,
just as a seriesconnected multiplier resistor gives a movingcoil (inherently resistive) meter
movement a greater voltage range: (Figure 12.6)
The Cathode Ray Tube (CRT) mentioned in the DC metering chapter is ideally suited for
measuring AC voltages, especially if the electron beam is swept sidetoside across the screen
370 CHAPTER 12. AC METERING CIRCUITS
Voltage
to be
measured
Sensitive
meter movement
Rmultiplier
AC voltmeter
Voltage
to be
measured
Sensitive
meter movement
Rmultiplier
AC voltmeter
(a) (b)
Figure 12.4: Multiplier resistor (a) or resistive divider (b) scales the range of the basic meter
movement.
Voltage
to be
measured
Sensitive
meter movement
Rmultiplier
Figure 12.5: AC voltmeter with capacitive divider.
Voltage
to be
measured
meter movement
Cmultiplier
Electrostatic
Figure 12.6: An electrostatic meter movement may use a capacitive multiplier to multiply the
scale of the basic meter movement..
12.1. AC VOLTMETERS AND AMMETERS 371
of the tube while the measured AC voltage drives the beam up and down. A graphical repre
sentation of the AC wave shape and not just a measurement of magnitude can easily be had
with such a device. However, CRT’s have the disadvantages of weight, size, significant power
consumption, and fragility (being made of evacuated glass) working against them. For these
reasons, electromechanical AC meter movements still have a place in practical usage.
With some of the advantages and disadvantages of these meter movement technologies
having been discussed already, there is another factor crucially important for the designer and
user of AC metering instruments to be aware of. This is the issue of RMS measurement. As
we already know, AC measurements are often cast in a scale of DC power equivalence, called
RMS (RootMeanSquare) for the sake of meaningful comparisons with DC and with other AC
waveforms of varying shape. None of the meter movement technologies so far discussed inher
ently measure the RMS value of an AC quantity. Meter movements relying on the motion of
a mechanical needle (“rectified” D’Arsonval, ironvane, and electrostatic) all tend to mechani
cally average the instantaneous values into an overall average value for the waveform. This
average value is not necessarily the same as RMS, although many times it is mistaken as such.
Average and RMS values rate against each other as such for these three common waveform
shapes: (Figure 12.7)
RMS = 0.707 (Peak)
AVG = 0.637 (Peak)
PP = 2 (Peak)
RMS = Peak
AVG = Peak
PP = 2 (Peak)
RMS = 0.577 (Peak)
AVG = 0.5 (Peak)
PP = 2 (Peak)
Figure 12.7: RMS, Average, and PeaktoPeak values for sine, square, and triangle waves.
Since RMS seems to be the kind of measurement most people are interested in obtaining
with an instrument, and electromechanical meter movements naturally deliver average mea
surements rather than RMS, what are AC meter designers to do? Cheat, of course! Typically
the assumption is made that the waveform shape to be measured is going to be sine (by far the
most common, especially for power systems), and then the meter movement scale is altered by
the appropriate multiplication factor. For sine waves we see that RMS is equal to 0.707 times
the peak value while Average is 0.637 times the peak, so we can divide one figure by the other
to obtain an averagetoRMS conversion factor of 1.109:
0.707
0.637
= 1.1099
In other words, the meter movement will be calibrated to indicate approximately 1.11 times
higher than it would ordinarily (naturally) indicate with no special accommodations. It must
be stressed that this “cheat” only works well when the meter is used to measure pure sine wave
sources. Note that for triangle waves, the ratio between RMS and Average is not the same as
for sine waves:
372 CHAPTER 12. AC METERING CIRCUITS
0.577
0.5
= 1.154
With square waves, the RMS and Average values are identical! An AC meter calibrated to
accurately read RMS voltage or current on a pure sine wave will not give the proper value while
indicating the magnitude of anything other than a perfect sine wave. This includes triangle
waves, square waves, or any kind of distorted sine wave. With harmonics becoming an ever
present phenomenon in large AC power systems, this matter of accurate RMS measurement is
no small matter.
The astute reader will note that I have omitted the CRT “movement” from the RMS/Average
discussion. This is because a CRT with its practically weightless electron beam “movement”
displays the Peak (or PeaktoPeak if you wish) of an AC waveform rather than Average or
RMS. Still, a similar problem arises: how do you determine the RMS value of a waveform
from it? Conversion factors between Peak and RMS only hold so long as the waveform falls
neatly into a known category of shape (sine, triangle, and square are the only examples with
Peak/RMS/Average conversion factors given here!).
One answer is to design the meter movement around the very definition of RMS: the ef
fective heating value of an AC voltage/current as it powers a resistive load. Suppose that the
AC source to be measured is connected across a resistor of known value, and the heat output
of that resistor is measured with a device like a thermocouple. This would provide a far more
direct measurement means of RMS than any conversion factor could, for it will work with ANY
waveform shape whatsoever: (Figure 12.8)
thermocouple bonded
with resistive heating
element
sensitive
meter
movement
AC voltage to
be measured
Figure 12.8: Direct reading thermal RMS voltmeter accommodates any wave shape.
While the device shown above is somewhat crude and would suffer from unique engineering
problems of its own, the concept illustrated is very sound. The resistor converts the AC voltage
or current quantity into a thermal (heat) quantity, effectively squaring the values in real
time. The system’s mass works to average these values by the principle of thermal inertia,
and then the meter scale itself is calibrated to give an indication based on the squareroot
of the thermal measurement: perfect RootMeanSquare indication all in one device! In fact,
one major instrument manufacturer has implemented this technique into its highend line of
handheld electronic multimeters for “trueRMS” capability.
Calibrating AC voltmeters and ammeters for different fullscale ranges of operation is much
12.1. AC VOLTMETERS AND AMMETERS 373
the same as with DC instruments: series “multiplier” resistors are used to give voltmeter move
ments higher range, and parallel “shunt” resistors are used to allow ammeter movements to
measure currents beyond their natural range. However, we are not limited to these techniques
as we were with DC: because we can use transformers with AC, meter ranges can be electro
magnetically rather than resistively “stepped up” or “stepped down,” sometimes far beyond
what resistors would have practically allowed for. Potential Transformers (PT’s) and Current
Transformers (CT’s) are precision instrument devices manufactured to produce very precise
ratios of transformation between primary and secondary windings. They can allow small, sim
ple AC meter movements to indicate extremely high voltages and currents in power systems
with accuracy and complete electrical isolation (something multiplier and shunt resistors could
never do): (Figure 12.9)
loadhighvoltagepower source 13.8 kV
fusefuse
V
precision
stepdown
ratio
PT
precision
ratio
stepupCT
A
05 A AC movement range
0120 V AC movement range
Figure 12.9: (CT) Current transformer scales current down. (PT) Potential transformer scales
voltage down.
Shown here is a voltage and current meter panel from a threephase AC system. The
three “donut” current transformers (CT’s) can be seen in the rear of the panel. Three AC
ammeters (rated 5 amps fullscale deflection each) on the front of the panel indicate current
through each conductor going through a CT. As this panel has been removed from service, there
are no currentcarrying conductors threaded through the center of the CT “donuts” anymore:
(Figure 12.10)
Because of the expense (and often large size) of instrument transformers, they are not used
to scale AC meters for any applications other than high voltage and high current. For scaling a
milliamp or microamp movement to a range of 120 volts or 5 amps, normal precision resistors
(multipliers and shunts) are used, just as with DC.
• REVIEW:
374 CHAPTER 12. AC METERING CIRCUITS
Figure 12.10: Toroidal current transformers scale high current levels down for application to 5
A fullscale AC ammeters.
• Polarized (DC) meter movements must use devices called diodes to be able to indicate AC
quantities.
• Electromechanical meter movements, whether electromagnetic or electrostatic, naturally
provide the average value of a measured AC quantity. These instruments may be ranged
to indicate RMS value, but only if the shape of the AC waveform is precisely known
beforehand!
• Socalled true RMS meters use different technology to provide indications representing
the actual RMS (rather than skewed average or peak) of an AC waveform.
12.2 Frequency and phase measurement
An important electrical quantity with no equivalent in DC circuits is frequency. Frequency
measurement is very important in many applications of alternating current, especially in AC
power systems designed to run efficiently at one frequency and one frequency only. If the AC is
being generated by an electromechanical alternator, the frequency will be directly proportional
to the shaft speed of the machine, and frequency could be measured simply by measuring the
speed of the shaft. If frequency needs to be measured at some distance from the alternator,
though, other means of measurement will be necessary.
One simple but crude method of frequency measurement in power systems utilizes the
principle of mechanical resonance. Every physical object possessing the property of elasticity
(springiness) has an inherent frequency at which it will prefer to vibrate. The tuning fork is
a great example of this: strike it once and it will continue to vibrate at a tone specific to its
length. Longer tuning forks have lower resonant frequencies: their tones will be lower on the
musical scale than shorter forks.
12.2. FREQUENCY AND PHASE MEASUREMENT 375
Imagine a row of progressivelysized tuning forks arranged sidebyside. They are all
mounted on a common base, and that base is vibrated at the frequency of the measured AC
voltage (or current) by means of an electromagnet. Whichever tuning fork is closest in reso
nant frequency to the frequency of that vibration will tend to shake the most (or the loudest).
If the forks’ tines were flimsy enough, we could see the relative motion of each by the length
of the blur we would see as we inspected each one from an endview perspective. Well, make a
collection of “tuning forks” out of a strip of sheet metal cut in a pattern akin to a rake, and you
have the vibrating reed frequency meter: (Figure 12.11)
to AC voltagesheet metal reeds
shaken by magnetic
field from the coil
Figure 12.11: Vibrating reed frequency meter diagram.
The user of this meter views the ends of all those unequal length reeds as they are collec
tively shaken at the frequency of the applied AC voltage to the coil. The one closest in resonant
frequency to the applied AC will vibrate the most, looking something like Figure 12.12.
Frequency Meter
120 Volts AC
6058565452 62 64 66 68
Figure 12.12: Vibrating reed frequency meter front panel.
376 CHAPTER 12. AC METERING CIRCUITS
Vibrating reed meters, obviously, are not precision instruments, but they are very simple
and therefore easy to manufacture to be rugged. They are often found on small enginedriven
generator sets for the purpose of setting engine speed so that the frequency is somewhat close
to 60 (50 in Europe) Hertz.
While reedtype meters are imprecise, their operational principle is not. In lieu of mechan
ical resonance, we may substitute electrical resonance and design a frequency meter using an
inductor and capacitor in the form of a tank circuit (parallel inductor and capacitor). See Fig
ure 12.13. One or both components are made adjustable, and a meter is placed in the circuit to
indicate maximum amplitude of voltage across the two components. The adjustment knob(s)
are calibrated to show resonant frequency for any given setting, and the frequency is read from
them after the device has been adjusted for maximum indication on the meter. Essentially, this
is a tunable filter circuit which is adjusted and then read in a manner similar to a bridge circuit
(which must be balanced for a “null” condition and then read).
Sensitive AC
meter movement
variable capacitor with
adjustment knob calibrated
in Hertz.
Figure 12.13: Resonant frequency meter “peaks” as LC resonant frequency is tuned to test
frequency.
This technique is a popular one for amateur radio operators (or at least it was before the ad
vent of inexpensive digital frequency instruments called counters), especially because it doesn’t
require direct connection to the circuit. So long as the inductor and/or capacitor can intercept
enough stray field (magnetic or electric, respectively) from the circuit under test to cause the
meter to indicate, it will work.
In frequency as in other types of electrical measurement, the most accurate means of mea
surement are usually those where an unknown quantity is compared against a known stan
dard, the basic instrument doing nothing more than indicating when the two quantities are
equal to each other. This is the basic principle behind the DC (Wheatstone) bridge circuit and
it is a sound metrological principle applied throughout the sciences. If we have access to an ac
curate frequency standard (a source of AC voltage holding very precisely to a single frequency),
then measurement of any unknown frequency by comparison should be relatively easy.
For that frequency standard, we turn our attention back to the tuning fork, or at least a
more modern variation of it called the quartz crystal. Quartz is a naturally occurring mineral
possessing a very interesting property called piezoelectricity. Piezoelectric materials produce
a voltage across their length when physically stressed, and will physically deform when an
external voltage is applied across their lengths. This deformation is very, very slight in most
cases, but it does exist.
12.2. FREQUENCY AND PHASE MEASUREMENT 377
Quartz rock is elastic (springy) within that small range of bending which an external volt
age would produce, which means that it will have a mechanical resonant frequency of its own
capable of being manifested as an electrical voltage signal. In other words, if a chip of quartz
is struck, it will “ring” with its own unique frequency determined by the length of the chip,
and that resonant oscillation will produce an equivalent voltage across multiple points of the
quartz chip which can be tapped into by wires fixed to the surface of the chip. In reciprocal
manner, the quartz chip will tend to vibrate most when it is “excited” by an applied AC voltage
at precisely the right frequency, just like the reeds on a vibratingreed frequency meter.
Chips of quartz rock can be precisely cut for desired resonant frequencies, and that chip
mounted securely inside a protective shell with wires extending for connection to an external
electric circuit. When packaged as such, the resulting device is simply called a crystal (or
sometimes “xtal”). The schematic symbol is shown in Figure 12.14.
crystal or xtal
Figure 12.14: Crystal (frequency determing element) schematic symbol.
Electrically, that quartz chip is equivalent to a series LC resonant circuit. (Figure 12.15)
The dielectric properties of quartz contribute an additional capacitive element to the equiva
lent circuit.
C
L
characteristics
of the quartz
Ccapacitance
caused by wire
connections
across quartz
Figure 12.15: Quartz crystal equivalent circuit.
378 CHAPTER 12. AC METERING CIRCUITS
The “capacitance” and “inductance” shown in series are merely electrical equivalents of the
quartz’s mechanical resonance properties: they do not exist as discrete components within the
crystal. The capacitance shown in parallel due to the wire connections across the dielectric
(insulating) quartz body is real, and it has an effect on the resonant response of the whole
system. A full discussion on crystal dynamics is not necessary here, but what needs to be
understood about crystals is this resonant circuit equivalence and how it can be exploited
within an oscillator circuit to achieve an output voltage with a stable, known frequency.
Crystals, as resonant elements, typically have much higher “Q” (quality) values than tank
circuits built from inductors and capacitors, principally due to the relative absence of stray
resistance, making their resonant frequencies very definite and precise. Because the resonant
frequency is solely dependent on the physical properties of quartz (a very stable substance, me
chanically), the resonant frequency variation over time with a quartz crystal is very, very low.
This is how quartz movement watches obtain their high accuracy: by means of an electronic
oscillator stabilized by the resonant action of a quartz crystal.
For laboratory applications, though, even greater frequency stability may be desired. To
achieve this, the crystal in question may be placed in a temperature stabilized environment
(usually an oven), thus eliminating frequency errors due to thermal expansion and contraction
of the quartz.
For the ultimate in a frequency standard though, nothing discovered thus far surpasses
the accuracy of a single resonating atom. This is the principle of the socalled atomic clock,
which uses an atom of mercury (or cesium) suspended in a vacuum, excited by outside energy
to resonate at its own unique frequency. The resulting frequency is detected as a radiowave
signal and that forms the basis for the most accurate clocks known to humanity. National
standards laboratories around the world maintain a few of these hyperaccurate clocks, and
broadcast frequency signals based on those atoms’ vibrations for scientists and technicians to
tune in and use for frequency calibration purposes.
Now we get to the practical part: once we have a source of accurate frequency, how do we
compare that against an unknown frequency to obtain a measurement? One way is to use a
CRT as a frequencycomparison device. Cathode Ray Tubes typically have means of deflecting
the electron beam in the horizontal as well as the vertical axis. If metal plates are used to
electrostatically deflect the electrons, there will be a pair of plates to the left and right of the
beam as well as a pair of plates above and below the beam as in Figure 12.16.
If we allow one AC signal to deflect the beam up and down (connect that AC voltage source
to the “vertical” deflection plates) and another AC signal to deflect the beam left and right
(using the other pair of deflection plates), patterns will be produced on the screen of the CRT
indicative of the ratio of these two AC frequencies. These patterns are called Lissajous figures
and are a common means of comparative frequency measurement in electronics.
If the two frequencies are the same, we will obtain a simple figure on the screen of the CRT,
the shape of that figure being dependent upon the phase shift between the two AC signals.
Here is a sampling of Lissajous figures for two sinewave signals of equal frequency, shown as
they would appear on the face of an oscilloscope (an AC voltagemeasuring instrument using a
CRT as its “movement”). The first picture is of the Lissajous figure formed by two AC voltages
perfectly in phase with each other: (Figure 12.17)
If the two AC voltages are not in phase with each other, a straight line will not be formed.
Rather, the Lissajous figure will take on the appearance of an oval, becoming perfectly circular
if the phase shift is exactly 90o between the two signals, and if their amplitudes are equal:
12.2. FREQUENCY AND PHASE MEASUREMENT 379
electron "gun"
electrons
electrons
light
view
screen(vacuum)
horizontal
deflection
plates
deflection
plates
vertical
Figure 12.16: Cathode ray tube (CRT) with vertical and horizontal deflection plates.
trigger
timebase
s/div
DC GND AC
X
GNDDC
V/div
vertical
OSCILLOSCOPE
Y
AC
Figure 12.17: Lissajous figure: same frequency, zero degrees phase shift.
380 CHAPTER 12. AC METERING CIRCUITS
(Figure 12.18)
trigger
timebase
s/div
DC GND AC
X
GNDDC
V/div
vertical
OSCILLOSCOPE
Y
AC
Figure 12.18: Lissajous figure: same frequency, 90 or 270 degrees phase shift.
Finally, if the two AC signals are directly opposing one another in phase (180o shift), we
will end up with a line again, only this time it will be oriented in the opposite direction: (Fig
ure 12.19)
trigger
timebase
s/div
DC GND AC
X
GNDDC
V/div
vertical
OSCILLOSCOPE
Y
AC
Figure 12.19: Lissajous figure: same frequency, 180 degrees phase shift.
When we are faced with signal frequencies that are not the same, Lissajous figures get
quite a bit more complex. Consider the following examples and their given vertical/horizontal
frequency ratios: (Figure 12.20)
The more complex the ratio between horizontal and vertical frequencies, the more com
plex the Lissajous figure. Consider the following illustration of a 3:1 frequency ratio between
horizontal and vertical: (Figure 12.21)
. . . and a 3:2 frequency ratio (horizontal = 3, vertical = 2) in Figure 12.22.
In cases where the frequencies of the two AC signals are not exactly a simple ratio of each
other (but close), the Lissajous figure will appear to “move,” slowly changing orientation as the
12.2. FREQUENCY AND PHASE MEASUREMENT 381
trigger
timebase
s/div
DC GND AC
X
GNDDC
V/div
vertical
OSCILLOSCOPE
Y
AC
Figure 12.20: Lissajous figure: Horizontal frequency is twice that of vertical.
trigger
timebase
s/div
DC GND AC
X
GNDDC
V/div
vertical
OSCILLOSCOPE
Y
AC
Figure 12.21: Lissajous figure: Horizontal frequency is three times that of vertical.
382 CHAPTER 12. AC METERING CIRCUITS
trigger
timebase
s/div DC GND AC
X
GNDDC
V/div
vertical
OSCILLOSCOPE
Y
AC
Lissajous figure: Horizontal/Vertical frequency ratio is 3:2
Figure 12.22: Lissajous figure: Horizontal/vertical frequency ratio is 3:2.
phase angle between the two waveforms rolls between 0o and 180o. If the two frequencies are
locked in an exact integer ratio between each other, the Lissajous figure will be stable on the
viewscreen of the CRT.
The physics of Lissajous figures limits their usefulness as a frequencycomparison tech
nique to cases where the frequency ratios are simple integer values (1:1, 1:2, 1:3, 2:3, 3:4,
etc.). Despite this limitation, Lissajous figures are a popular means of frequency comparison
wherever an accessible frequency standard (signal generator) exists.
• REVIEW:
• Some frequency meters work on the principle of mechanical resonance, indicating fre
quency by relative oscillation among a set of uniquely tuned “reeds” shaken at the mea
sured frequency.
• Other frequency meters use electric resonant circuits (LC tank circuits, usually) to in
dicate frequency. One or both components is made to be adjustable, with an accurately
calibrated adjustment knob, and a sensitive meter is read for maximum voltage or cur
rent at the point of resonance.
• Frequency can be measured in a comparative fashion, as is the case when using a CRT to
generate Lissajous figures. Reference frequency signals can be made with a high degree
of accuracy by oscillator circuits using quartz crystals as resonant devices. For ultra
precision, atomic clock signal standards (based on the resonant frequencies of individual
atoms) can be used.
12.3 Power measurement
Power measurement in AC circuits can be quite a bit more complex than with DC circuits
for the simple reason that phase shift complicates the matter beyond multiplying voltage by
12.3. POWER MEASUREMENT 383
current figures obtained with meters. What is needed is an instrument able to determine
the product (multiplication) of instantaneous voltage and current. Fortunately, the common
electrodynamometer movement with its stationary and moving coil does a fine job of this.
Three phase power measurement can be accomplished using two dynamometer movements
with a common shaft linking the two moving coils together so that a single pointer registers
power on a meter movement scale. This, obviously, makes for a rather expensive and complex
movement mechanism, but it is a workable solution.
An ingenious method of deriving an electronic power meter (one that generates an electric
signal representing power in the system rather than merely move a pointer) is based on the
Hall effect. The Hall effect is an unusual effect first noticed by E. H. Hall in 1879, whereby a
voltage is generated along the width of a currentcarrying conductor exposed to a perpendicular
magnetic field: (Figure 12.23)
N
S
currentcurrent
voltage
S
N
Figure 12.23: Hall effect: Voltage is proportional to current and strength of the perpendicular
magnetic field.
The voltage generated across the width of the flat, rectangular conductor is directly propor
tional to both the magnitude of the current through it and the strength of the magnetic field.
Mathematically, it is a product (multiplication) of these two variables. The amount of “Hall
Voltage” produced for any given set of conditions also depends on the type of material used
for the flat, rectangular conductor. It has been found that specially prepared “semiconductor”
materials produce a greater Hall voltage than do metals, and so modern Hall Effect devices
are made of these.
It makes sense then that if we were to build a device using a Halleffect sensor where the
current through the conductor was pushed by AC voltage from an external circuit and the
magnetic field was set up by a pair or wire coils energized by the current of the AC power
circuit, the Hall voltage would be in direct proportion to the multiple of circuit current and
384 CHAPTER 12. AC METERING CIRCUITS
voltage. Having no mass to move (unlike an electromechanical movement), this device is able
to provide instantaneous power measurement: (Figure 12.24)
voltage
LoadACpower
source
Rmultiplier
Figure 12.24: Hall effect power sensor measures instantaneous power.
Not only will the output voltage of the Hall effect device be the representation of instan
taneous power at any point in time, but it will also be a DC signal! This is because the Hall
voltage polarity is dependent upon both the polarity of the magnetic field and the direction of
current through the conductor. If both current direction and magnetic field polarity reverses –
as it would ever halfcycle of the AC power – the output voltage polarity will stay the same.
If voltage and current in the power circuit are 90o out of phase (a power factor of zero,
meaning no real power delivered to the load), the alternate peaks of Hall device current and
magnetic field will never coincide with each other: when one is at its peak, the other will be
zero. At those points in time, the Hall output voltage will likewise be zero, being the product
(multiplication) of current and magnetic field strength. Between those points in time, the
Hall output voltage will fluctuate equally between positive and negative, generating a signal
corresponding to the instantaneous absorption and release of power through the reactive load.
The net DC output voltage will be zero, indicating zero true power in the circuit.
Any phase shift between voltage and current in the power circuit less than 90o will result
in a Hall output voltage that oscillates between positive and negative, but spends more time
positive than negative. Consequently there will be a net DC output voltage. Conditioned
through a lowpass filter circuit, this net DC voltage can be separated from the AC mixed with
it, the final output signal registered on a sensitive DC meter movement.
Often it is useful to have a meter to totalize power usage over a period of time rather than
instantaneously. The output of such a meter can be set in units of Joules, or total energy
12.4. POWER QUALITY MEASUREMENT 385
consumed, since power is a measure of work being done per unit time. Or, more commonly, the
output of the meter can be set in units of WattHours.
Mechanical means for measuring WattHours are usually centered around the concept of
the motor: build an AC motor that spins at a rate of speed proportional to the instantaneous
power in a circuit, then have that motor turn an “odometer” style counting mechanism to keep
a running total of energy consumed. The “motor” used in these meters has a rotor made of
a thin aluminum disk, with the rotating magnetic field established by sets of coils energized
by line voltage and load current so that the rotational speed of the disk is dependent on both
voltage and current.
12.4 Power quality measurement
It used to be with large AC power systems that “power quality” was an unheardof concept,
aside from power factor. Almost all loads were of the “linear” variety, meaning that they did
not distort the shape of the voltage sine wave, or cause nonsinusoidal currents to flow in the
circuit. This is not true anymore. Loads controlled by “nonlinear” electronic components are
becoming more prevalent in both home and industry, meaning that the voltages and currents in
the power system(s) feeding these loads are rich in harmonics: what should be nice, clean sine
wave voltages and currents are becoming highly distorted, which is equivalent to the presence
of an infinite series of highfrequency sine waves at multiples of the fundamental power line
frequency.
Excessive harmonics in an AC power system can overheat transformers, cause exceedingly
high neutral conductor currents in threephase systems, create electromagnetic “noise” in the
form of radio emissions that can interfere with sensitive electronic equipment, reduce electric
motor horsepower output, and can be difficult to pinpoint. With problems like these plaguing
power systems, engineers and technicians require ways to precisely detect and measure these
conditions.
Power Quality is the general term given to represent an AC power system’s freedom from
harmonic content. A “power quality” meter is one that gives some form of harmonic content
indication.
A simple way for a technician to determine power quality in their system without sophis
ticated equipment is to compare voltage readings between two accurate voltmeters measuring
the same system voltage: one meter being an “averaging” type of unit (such as an electrome
chanical movement meter) and the other being a “trueRMS” type of unit (such as a high
quality digital meter). Remember that “averaging” type meters are calibrated so that their
scales indicate volts RMS, based on the assumption that the AC voltage being measured is sinu
soidal. If the voltage is anything but sinewaveshaped, the averaging meter will not register
the proper value, whereas the trueRMS meter always will, regardless of waveshape. The rule
of thumb here is this: the greater the disparity between the two meters, the worse the power
quality is, and the greater its harmonic content. A power system with good quality power
should generate equal voltage readings between the two meters, to within the rated error tol
erance of the two instruments.
Another qualitative measurement of power quality is the oscilloscope test: connect an os
cilloscope (CRT) to the AC voltage and observe the shape of the wave. Anything other than a
clean sine wave could be an indication of trouble: (Figure 12.25)
386 CHAPTER 12. AC METERING CIRCUITS
trigger
timebase
s/div
DC GND AC
X
GNDDC
V/div
vertical
OSCILLOSCOPE
Y
AC
Figure 12.25: This is a moderately ugly “sine” wave. Definite harmonic content here!
Still, if quantitative analysis (definite, numerical figures) is necessary, there is no substitute
for an instrument specifically designed for that purpose. Such an instrument is called a power
quality meter and is sometimes better known in electronic circles as a lowfrequency spectrum
analyzer. What this instrument does is provide a graphical representation on a CRT or digital
display screen of the AC voltage’s frequency “spectrum.” Just as a prism splits a beam of white
light into its constituent color components (how much red, orange, yellow, green, and blue
is in that light), the spectrum analyzer splits a mixedfrequency signal into its constituent
frequencies, and displays the result in the form of a histogram: (Figure 12.26)
1 3 5 7 9 11 13
Total distortion = 43.7 %
Power Quality Meter
Figure 12.26: Power quality meter is a low frequency spectrum analyzer.
12.5. AC BRIDGE CIRCUITS 387
Each number on the horizontal scale of this meter represents a harmonic of the fundamen
tal frequency. For American power systems, the “1” represents 60 Hz (the 1st harmonic, or
fundamental), the “3” for 180 Hz (the 3rd harmonic), the “5” for 300 Hz (the 5th harmonic),
and so on. The black rectangles represent the relative magnitudes of each of these harmonic
components in the measured AC voltage. A pure, 60 Hz sine wave would show only a tall black
bar over the “1” with no black bars showing at all over the other frequency markers on the
scale, because a pure sine wave has no harmonic content.
Power quality meters such as this might be better referred to as overtone meters, because
they are designed to display only those frequencies known to be generated by the power system.
In threephase AC power systems (predominant for large power applications), evennumbered
harmonics tend to be canceled out, and so only harmonics existing in significant measure are
the oddnumbered.
Meters like these are very useful in the hands of a skilled technician, because different
types of nonlinear loads tend to generate different spectrum “signatures” which can clue the
troubleshooter to the source of the problem. These meters work by very quickly sampling
the AC voltage at many different points along the waveform shape, digitizing those points
of information, and using a microprocessor (small computer) to perform numerical Fourier
analysis (the Fast Fourier Transform or “FFT” algorithm) on those data points to arrive at
harmonic frequency magnitudes. The process is not much unlike what the SPICE program
tells a computer to do when performing a Fourier analysis on a simulated circuit voltage or
current waveform.
12.5 AC bridge circuits
As we saw with DC measurement circuits, the circuit configuration known as a bridge can be
a very useful way to measure unknown values of resistance. This is true with AC as well, and
we can apply the very same principle to the accurate measurement of unknown impedances.
To review, the bridge circuit works as a pair of twocomponent voltage dividers connected
across the same source voltage, with a nulldetector meter movement connected between them
to indicate a condition of “balance” at zero volts: (Figure 12.27)
Any one of the four resistors in the above bridge can be the resistor of unknown value,
and its value can be determined by a ratio of the other three, which are “calibrated,” or whose
resistances are known to a precise degree. When the bridge is in a balanced condition (zero
voltage as indicated by the null detector), the ratio works out to be this:
R1
R2
=
R3
R4
In a condition of balance:
One of the advantages of using a bridge circuit to measure resistance is that the voltage of
the power source is irrelevant. Practically speaking, the higher the supply voltage, the easier
it is to detect a condition of imbalance between the four resistors with the null detector, and
thus the more sensitive it will be. A greater supply voltage leads to the possibility of increased
measurement precision. However, there will be no fundamental error introduced as a result of
a lesser or greater power supply voltage unlike other types of resistance measurement schemes.
388 CHAPTER 12. AC METERING CIRCUITS
R1
R2
R3
R4
null
Figure 12.27: A balanced bridge shows a “null”, or minimum reading, on the indicator.
Impedance bridges work the same, only the balance equation is with complex quantities, as
both magnitude and phase across the components of the two dividers must be equal in order
for the null detector to indicate “zero.” The null detector, of course, must be a device capable of
detecting very small AC voltages. An oscilloscope is often used for this, although very sensitive
electromechanical meter movements and even headphones (small speakers) may be used if the
source frequency is within audio range.
One way to maximize the effectiveness of audio headphones as a null detector is to connect
them to the signal source through an impedancematching transformer. Headphone speak
ers are typically lowimpedance units (8 Ω), requiring substantial current to drive, and so a
stepdown transformer helps “match” lowcurrent signals to the impedance of the headphone
speakers. An audio output transformer works well for this purpose: (Figure 12.28)
Test
leads
Headphones
8 Ω
made from audio headphones
Null detector for AC bridge
To test
Press button
1 kΩ
Figure 12.28: “Modern” lowOhm headphones require an impedance matching transformer for
use as a sensitive null detector.
12.5. AC BRIDGE CIRCUITS 389
Using a pair of headphones that completely surround the ears (the “closedcup” type), I’ve
been able to detect currents of less than 0.1 µA with this simple detector circuit. Roughly
equal performance was obtained using two different stepdown transformers: a small power
transformer (120/6 volt ratio), and an audio output transformer (1000:8 ohm impedance ratio).
With the pushbutton switch in place to interrupt current, this circuit is usable for detecting
signals from DC to over 2 MHz: even if the frequency is far above or below the audio range, a
“click” will be heard from the headphones each time the switch is pressed and released.
Connected to a resistive bridge, the whole circuit looks like Figure 12.29.
Headphones
R1
R2
R3
R4
Figure 12.29: Bridge with sensitive AC null detector.
Listening to the headphones as one or more of the resistor “arms” of the bridge is adjusted,
a condition of balance will be realized when the headphones fail to produce “clicks” (or tones, if
the bridge’s power source frequency is within audio range) as the switch is actuated.
When describing general AC bridges, where impedances and not just resistances must be in
proper ratio for balance, it is sometimes helpful to draw the respective bridge legs in the form
of boxshaped components, each one with a certain impedance: (Figure 12.30)
For this general form of AC bridge to balance, the impedance ratios of each branch must be
equal:
=
Z1
Z2
Z3
Z4
Again, it must be stressed that the impedance quantities in the above equation must be
complex, accounting for both magnitude and phase angle. It is insufficient that the impedance
magnitudes alone be balanced; without phase angles in balance as well, there will still be
voltage across the terminals of the null detector and the bridge will not be balanced.
Bridge circuits can be constructed to measure just about any device value desired, be it
capacitance, inductance, resistance, or even “Q.” As always in bridge measurement circuits,
the unknown quantity is always “balanced” against a known standard, obtained from a high
quality, calibrated component that can be adjusted in value until the null detector device indi
cates a condition of balance. Depending on how the bridge is set up, the unknown component’s
value may be determined directly from the setting of the calibrated standard, or derived from
390 CHAPTER 12. AC METERING CIRCUITS
Z1
Z2
Z3
Z4
null
Figure 12.30: Generalized AC impedance bridge: Z = nonspecific complex impedance.
that standard through a mathematical formula.
A couple of simple bridge circuits are shown below, one for inductance (Figure 12.31) and
one for capacitance: (Figure 12.32)
standard
inductance
unknown
inductance
Lx Ls
R R
null
Figure 12.31: Symmetrical bridge measures unknown inductor by comparison to a standard
inductor.
Simple “symmetrical” bridges such as these are so named because they exhibit symmetry
(mirrorimage similarity) from left to right. The two bridge circuits shown above are balanced
by adjusting the calibrated reactive component (Ls or Cs). They are a bit simplified from their
reallife counterparts, as practical symmetrical bridge circuits often have a calibrated, variable
resistor in series or parallel with the reactive component to balance out stray resistance in the
unknown component. But, in the hypothetical world of perfect components, these simple bridge
12.5. AC BRIDGE CIRCUITS 391
standard
unknown
R R
capacitanceCsCx
capacitance
null
Figure 12.32: Symmetrical bridge measures unknown capacitor by comparison to a standard
capacitor.
circuits do just fine to illustrate the basic concept.
An example of a little extra complexity added to compensate for realworld effects can be
found in the socalled Wien bridge, which uses a parallel capacitorresistor standard impedance
to balance out an unknown series capacitorresistor combination. (Figure 12.33) All capacitors
have some amount of internal resistance, be it literal or equivalent (in the form of dielectric
heating losses) which tend to spoil their otherwise perfectly reactive natures. This internal
resistance may be of interest to measure, and so the Wien bridge attempts to do so by providing
a balancing impedance that isn’t “pure” either:
Being that there are two standard components to be adjusted (a resistor and a capacitor)
this bridge will take a little more time to balance than the others we’ve seen so far. The
combined effect of Rs and Cs is to alter the magnitude and phase angle until the bridge achieves
a condition of balance. Once that balance is achieved, the settings of Rs and Cs can be read from
their calibrated knobs, the parallel impedance of the two determined mathematically, and the
unknown capacitance and resistance determined mathematically from the balance equation
(Z1/Z2 = Z3/Z4).
It is assumed in the operation of the Wien bridge that the standard capacitor has negligible
internal resistance, or at least that resistance is already known so that it can be factored into
the balance equation. Wien bridges are useful for determining the values of “lossy” capacitor
designs like electrolytics, where the internal resistance is relatively high. They are also used
as frequency meters, because the balance of the bridge is frequencydependent. When used
in this fashion, the capacitors are made fixed (and usually of equal value) and the top two
resistors are made variable and are adjusted by means of the same knob.
An interesting variation on this theme is found in the next bridge circuit, used to precisely
measure inductances.
This ingenious bridge circuit is known as the MaxwellWien bridge (sometimes known
plainly as the Maxwell bridge), and is used to measure unknown inductances in terms of
calibrated resistance and capacitance. (Figure 12.34) Calibrationgrade inductors are more
392 CHAPTER 12. AC METERING CIRCUITS
R R
Cx
Rx
Cs
Rs
null
Figure 12.33: Wein Bridge measures both capacitive Cx and resistive Rx components of “real”
capacitor.
Rx
Lx
Cs
Rs
R
R
null
Figure 12.34: MaxwellWein bridge measures an inductor in terms of a capacitor standard.
12.5. AC BRIDGE CIRCUITS 393
difficult to manufacture than capacitors of similar precision, and so the use of a simple “sym
metrical” inductance bridge is not always practical. Because the phase shifts of inductors and
capacitors are exactly opposite each other, a capacitive impedance can balance out an inductive
impedance if they are located in opposite legs of a bridge, as they are here.
Another advantage of using a Maxwell bridge to measure inductance rather than a sym
metrical inductance bridge is the elimination of measurement error due to mutual inductance
between two inductors. Magnetic fields can be difficult to shield, and even a small amount of
coupling between coils in a bridge can introduce substantial errors in certain conditions. With
no second inductor to react with in the Maxwell bridge, this problem is eliminated.
For easiest operation, the standard capacitor (Cs) and the resistor in parallel with it (Rs)
are made variable, and both must be adjusted to achieve balance. However, the bridge can be
made to work if the capacitor is fixed (nonvariable) and more than one resistor made variable
(at least the resistor in parallel with the capacitor, and one of the other two). However, in
the latter configuration it takes more trialanderror adjustment to achieve balance, as the
different variable resistors interact in balancing magnitude and phase.
Unlike the plain Wien bridge, the balance of the MaxwellWien bridge is independent of
source frequency, and in some cases this bridge can be made to balance in the presence of
mixed frequencies from the AC voltage source, the limiting factor being the inductor’s stability
over a wide frequency range.
There are more variations beyond these designs, but a full discussion is not warranted here.
Generalpurpose impedance bridge circuits are manufactured which can be switched into more
than one configuration for maximum flexibility of use.
A potential problem in sensitive AC bridge circuits is that of stray capacitance between
either end of the null detector unit and ground (earth) potential. Because capacitances can
“conduct” alternating current by charging and discharging, they form stray current paths to
the AC voltage source which may affect bridge balance: (Figure 12.35)
While reedtype meters are imprecise, their operational principle is not. In lieu of mechan
ical resonance, we may substitute electrical resonance and design a frequency meter using an
inductor and capacitor in the form of a tank circuit (parallel inductor and capacitor). One or
both components are made adjustable, and a meter is placed in the circuit to indicate maxi
mum amplitude of voltage across the two components. The adjustment knob(s) are calibrated
to show resonant frequency for any given setting, and the frequency is read from them after the
device has been adjusted for maximum indication on the meter. Essentially, this is a tunable
filter circuit which is adjusted and then read in a manner similar to a bridge circuit (which
must be balanced for a “null” condition and then read). The problem is worsened if the AC
voltage source is firmly grounded at one end, the total stray impedance for leakage currents
made far less and any leakage currents through these stray capacitances made greater as a
result: (Figure 12.36)
One way of greatly reducing this effect is to keep the null detector at ground potential, so
there will be no AC voltage between it and the ground, and thus no current through stray
capacitances. However, directly connecting the null detector to ground is not an option, as it
would create a direct current path for stray currents, which would be worse than any capacitive
path. Instead, a special voltage divider circuit called a Wagner ground or Wagner earth may be
used to maintain the null detector at ground potential without the need for a direct connection
to the null detector. (Figure 12.37)
The Wagner earth circuit is nothing more than a voltage divider, designed to have the volt
394 CHAPTER 12. AC METERING CIRCUITS
R R
CsCx
null
Cstray Cstray
Cstray
Cstray
Figure 12.35: Stray capacitance to ground may introduce errors into the bridge.
R R
CsCx
null
Cstray Cstray
Figure 12.36: Stray capacitance errors are more severe if one side of the AC supply is grounded.
12.5. AC BRIDGE CIRCUITS 395
R R
CsCx
null
Cstray Cstray
Wagner
earth
Cstray
Cstray
Figure 12.37: Wagner ground for AC supply minimizes the effects of stray capacitance to
ground on the bridge.
age ratio and phase shift as each side of the bridge. Because the midpoint of the Wagner divider
is directly grounded, any other divider circuit (including either side of the bridge) having the
same voltage proportions and phases as the Wagner divider, and powered by the same AC
voltage source, will be at ground potential as well. Thus, the Wagner earth divider forces the
null detector to be at ground potential, without a direct connection between the detector and
ground.
There is often a provision made in the null detector connection to confirm proper setting
of the Wagner earth divider circuit: a twoposition switch, (Figure 12.38) so that one end of
the null detector may be connected to either the bridge or the Wagner earth. When the null
detector registers zero signal in both switch positions, the bridge is not only guaranteed to
be balanced, but the null detector is also guaranteed to be at zero potential with respect to
ground, thus eliminating any errors due to leakage currents through stray detectortoground
capacitances:
• REVIEW:
• AC bridge circuits work on the same basic principle as DC bridge circuits: that a bal
anced ratio of impedances (rather than resistances) will result in a “balanced” condition
as indicated by the nulldetector device.
• Null detectors for AC bridges may be sensitive electromechanical meter movements, os
cilloscopes (CRT’s), headphones (amplified or unamplified), or any other device capable of
registering very small AC voltage levels. Like DC null detectors, its only required point
of calibration accuracy is at zero.
396 CHAPTER 12. AC METERING CIRCUITS
R R
CsCx
null
Cstray Cstray
Cstray
Cstray
Figure 12.38: Switchup position allows adjustment of the Wagner ground.
• AC bridge circuits can be of the “symmetrical” type where an unknown impedance is
balanced by a standard impedance of similar type on the same side (top or bottom) of the
bridge. Or, they can be “nonsymmetrical,” using parallel impedances to balance series
impedances, or even capacitances balancing out inductances.
• AC bridge circuits often have more than one adjustment, since both impedance magni
tude and phase angle must be properly matched to balance.
• Some impedance bridge circuits are frequencysensitive while others are not. The frequency
sensitive types may be used as frequency measurement devices if all component values
are accurately known.
• A Wagner earth or Wagner ground is a voltage divider circuit added to AC bridges to help
reduce errors due to stray capacitance coupling the null detector to ground.
12.6 AC instrumentation transducers
Just as devices have been made to measure certain physical quantities and repeat that infor
mation in the form of DC electrical signals (thermocouples, strain gauges, pH probes, etc.),
special devices have been made that do the same with AC.
It is often necessary to be able to detect and transmit the physical position of mechanical
parts via electrical signals. This is especially true in the fields of automated machine tool
control and robotics. A simple and easy way to do this is with a potentiometer: (Figure 12.39)
12.6. AC INSTRUMENTATION TRANSDUCERS 397
+
V

potentiometer shaft moved
by physical motion of an object
voltmeter indicates
position of that object
Figure 12.39: Potentiometer tap voltage indicates position of an object slaved to the shaft.
However, potentiometers have their own unique problems. For one, they rely on physi
cal contact between the “wiper” and the resistance strip, which means they suffer the effects
of physical wear over time. As potentiometers wear, their proportional output versus shaft
position becomes less and less certain. You might have already experienced this effect when
adjusting the volume control on an old radio: when twisting the knob, you might hear “scratch
ing” sounds coming out of the speakers. Those noises are the result of poor wiper contact in
the volume control potentiometer.
Also, this physical contact between wiper and strip creates the possibility of arcing (spark
ing) between the two as the wiper is moved. With most potentiometer circuits, the current is so
low that wiper arcing is negligible, but it is a possibility to be considered. If the potentiometer
is to be operated in an environment where combustible vapor or dust is present, this potential
for arcing translates into a potential for an explosion!
Using AC instead of DC, we are able to completely avoid sliding contact between parts if
we use a variable transformer instead of a potentiometer. Devices made for this purpose are
called LVDT’s, which stands for Linear Variable Differential Transformers. The design of an
LVDT looks like this: (Figure 12.40)
Obviously, this device is a transformer: it has a single primary winding powered by an
external source of AC voltage, and two secondary windings connected in seriesbucking fashion.
It is variable because the core is free to move between the windings. It is differential because
of the way the two secondary windings are connected. Being arranged to oppose each other
(180o out of phase) means that the output of this device will be the difference between the
voltage output of the two secondary windings. When the core is centered and both windings
are outputting the same voltage, the net result at the output terminals will be zero volts. It is
called linear because the core’s freedom of motion is straightline.
The AC voltage output by an LVDT indicates the position of the movable core. Zero volts
means that the core is centered. The further away the core is from center position, the greater
percentage of input (“excitation”) voltage will be seen at the output. The phase of the output
voltage relative to the excitation voltage indicates which direction from center the core is offset.
The primary advantage of an LVDT over a potentiometer for position sensing is the absence
of physical contact between the moving and stationary parts. The core does not contact the wire
windings, but slides in and out within a nonconducting tube. Thus, the LVDT does not “wear”
like a potentiometer, nor is there the possibility of creating an arc.
Excitation of the LVDT is typically 10 volts RMS or less, at frequencies ranging from power
398 CHAPTER 12. AC METERING CIRCUITS
movable core
AC "excitation"
voltage
AC output
voltage
Figure 12.40: AC output of linear variable differential transformer (LVDT) indicates core posi
tion.
line to the high audio (20 kHz) range. One potential disadvantage of the LVDT is its response
time, which is mostly dependent on the frequency of the AC voltage source. If very quick
response times are desired, the frequency must be higher to allow whatever voltagesensing
circuits enough cycles of AC to determine voltage level as the core is moved. To illustrate
the potential problem here, imagine this exaggerated scenario: an LVDT powered by a 60
Hz voltage source, with the core being moved in and out hundreds of times per second. The
output of this LVDT wouldn’t even look like a sine wave because the core would be moved
throughout its range of motion before the AC source voltage could complete a single cycle! It
would be almost impossible to determine instantaneous core position if it moves faster than
the instantaneous source voltage does.
A variation on the LVDT is the RVDT, or Rotary Variable Differential Transformer. This
device works on almost the same principle, except that the core revolves on a shaft instead of
moving in a straight line. RVDT’s can be constructed for limited motion of 360o (fullcircle)
motion.
Continuing with this principle, we have what is known as a Synchro or Selsyn, which is a
device constructed a lot like a woundrotor polyphase AC motor or generator. The rotor is free
to revolve a full 360o, just like a motor. On the rotor is a single winding connected to a source
of AC voltage, much like the primary winding of an LVDT. The stator windings are usually in
the form of a threephase Y, although synchros with more than three phases have been built.
(Figure 12.41) A device with a twophase stator is known as a resolver. A resolver produces
sine and cosine outputs which indicate shaft position.
Voltages induced in the stator windings from the rotor’s AC excitation are not phaseshifted
by 120o as in a real threephase generator. If the rotor were energized with DC current rather
than AC and the shaft spun continuously, then the voltages would be true threephase. But this
is not how a synchro is designed to be operated. Rather, this is a positionsensing device much
like an RVDT, except that its output signal is much more definite. With the rotor energized
by AC, the stator winding voltages will be proportional in magnitude to the angular position
12.6. AC INSTRUMENTATION TRANSDUCERS 399
rotor
winding
threephase
stator winding
AC voltage
source
modern schematic symbol
stator
connections
rotor
connections
Synchro (a.k.a "Selsyn")
rotor
winding
twophase
stator winding
Resolver
stator
connections
rotor
connections
Figure 12.41: A synchro is wound with a threephase stator winding, and a rotating field. A
resolver has a twophase stator.
of the rotor, phase either 0o or 180o shifted, like a regular LVDT or RVDT. You could think of
it as a transformer with one primary winding and three secondary windings, each secondary
winding oriented at a unique angle. As the rotor is slowly turned, each winding in turn will
line up directly with the rotor, producing full voltage, while the other windings will produce
something less than full voltage.
Synchros are often used in pairs. With their rotors connected in parallel and energized
by the same AC voltage source, their shafts will match position to a high degree of accuracy:
(Figure 12.42)
Synchro "transmitter" Synchro "receiver"
The receiver rotor will turn to match position with the
transmitter rotor so long as the two rotors remain energized.
Figure 12.42: Synchro shafts are slaved to each other. Rotating one moves the other.
Such “transmitter/receiver” pairs have been used on ships to relay rudder position, or to
400 CHAPTER 12. AC METERING CIRCUITS
relay navigational gyro position over fairly long distances. The only difference between the
“transmitter” and the “receiver” is which one gets turned by an outside force. The “receiver”
can just as easily be used as the “transmitter” by forcing its shaft to turn and letting the
synchro on the left match position.
If the receiver’s rotor is left unpowered, it will act as a positionerror detector, generating
an AC voltage at the rotor if the shaft is anything other than 90o or 270o shifted from the
shaft position of the transmitter. The receiver rotor will no longer generate any torque and
consequently will no longer automatically match position with the transmitter’s: (Figure 12.43)
Synchro "transmitter" Synchro "receiver"
AC voltmeter
Figure 12.43: AC voltmeter registers voltage if the receiver rotor is not rotated exactly 90 or
270 degrees from the transmitter rotor.
This can be thought of almost as a sort of bridge circuit that achieves balance only if the
receiver shaft is brought to one of two (matching) positions with the transmitter shaft.
One rather ingenious application of the synchro is in the creation of a phaseshifting device,
provided that the stator is energized by threephase AC: (Figure 12.44)
threephase AC voltage
source (can be Y or Delta)
Synchro
voltage signal
output
Figure 12.44: Full rotation of the rotor will smoothly shift the phase from 0o all the way to
360o (back to 0o).
As the synchro’s rotor is turned, the rotor coil will progressively align with each stator coil,
their respective magnetic fields being 120o phaseshifted from one another. In between those
positions, these phaseshifted fields will mix to produce a rotor voltage somewhere between 0o,
12.6. AC INSTRUMENTATION TRANSDUCERS 401
120o, or 240o shift. The practical result is a device capable of providing an infinitely variable
phase AC voltage with the twist of a knob (attached to the rotor shaft).
A synchro or a resolver may measure linear motion if geared with a rack and pinion mech
anism. A linear movement of a few inches (or cm) resulting in multiple revolutions of the
synchro (resolver) generates a train of sinewaves. An Inductosyn R© is a linear version of the
resolver. It outputs signals like a resolver; though, it bears slight resemblance.
The Inductosyn consists of two parts: a fixed serpentine winding having a 0.1 in or 2 mm
pitch, and a movable winding known as a slider. (Figure 12.45) The slider has a pair of wind
ings having the same pitch as the fixed winding. The slider windings are offset by a quarter
pitch so both sine and cosine waves are produced by movement. One slider winding is adequate
for counting pulses, but provides no direction information. The 2phase windings provide direc
tion information in the phasing of the sine and cosine waves. Movement by one pitch produces
a cycle of sine and cosine waves; multiple pitches produce a train of waves.
sin(θ)
θ
cos(θ)
fixed
slider
P
fixed
slider
(a) (b)
Figure 12.45: Inductosyn: (a) Fixed serpentine winding, (b) movable slider 2phase windings.
Adapted from Figure 6.16 [1]
When we say sine and cosine waves are produces as a function of linear movement, we
really mean a high frequency carrier is amplitude modulated as the slider moves. The two
slider AC signals must be measured to determine position within a pitch, the fine position.
How many pitches has the slider moved? The sine and cosine signals’ relationship does not
reveal that. However, the number of pitches (number of waves) may be counted from a known
starting point yielding coarse position. This is an incremental encoder. If absolute position
must be known regardless of the starting point, an auxiliary resolver geared for one revolution
per length gives a coarse position. This constitutes an absolute encoder.
A linear Inductosyn has a transformer ratio of 100:1. Compare this to the 1:1 ratio for a
resolver. A few volts AC excitation into an Inductosyn yields a few millivolts out. This low
signal level is converted to to a 12bit digital format by a resolver to digital converter (RDC).
Resolution of 25 microinches is achievable.
402 CHAPTER 12. AC METERING CIRCUITS
There is also a rotary version of the Inductosyn having 360 pattern pitches per revolution.
When used with a 12bit resolver to digital converter, better that 1 arc second resolution is
achievable. This is an incremental encoder. Counting of pitches from a known starting point
is necessary to determine absolute position. Alternatively, a resolver may determine coarse
absolute position. [1]
So far the transducers discussed have all been of the inductive variety. However, it is
possible to make transducers which operate on variable capacitance as well, AC being used to
sense the change in capacitance and generate a variable output voltage.
Remember that the capacitance between two conductive surfaces varies with three major
factors: the overlapping area of those two surfaces, the distance between them, and the di
electric constant of the material in between the surfaces. If two out of three of these variables
can be fixed (stabilized) and the third allowed to vary, then any measurement of capacitance
between the surfaces will be solely indicative of changes in that third variable.
Medical researchers have longmade use of capacitive sensing to detect physiological changes
in living bodies. As early as 1907, a German researcher named H. Cremer placed two metal
plates on either side of a beating frog heart and measured the capacitance changes resulting
from the heart alternately filling and emptying itself of blood. Similar measurements have
been performed on human beings with metal plates placed on the chest and back, recording
respiratory and cardiac action by means of capacitance changes. For more precise capacitive
measurements of organ activity, metal probes have been inserted into organs (especially the
heart) on the tips of catheter tubes, capacitance being measured between the metal probe and
the body of the subject. With a sufficiently high AC excitation frequency and sensitive enough
voltage detector, not just the pumping action but also the sounds of the active heart may be
readily interpreted.
Like inductive transducers, capacitive transducers can also be made to be selfcontained
units, unlike the direct physiological examples described above. Some transducers work by
making one of the capacitor plates movable, either in such a way as to vary the overlapping
area or the distance between the plates. Other transducers work by moving a dielectric mate
rial in and out between two fixed plates: (Figure 12.46)
(a) (b) (c)
Figure 12.46: Variable capacitive transducer varies; (a) area of overlap, (b) distance between
plates, (c) amount of dielectric between plates.
Transducers with greater sensitivity and immunity to changes in other variables can be
obtained by way of differential design, much like the concept behind the LVDT (Linear Vari
able Differential Transformer). Here are a few examples of differential capacitive transducers:
(Figure 12.47)
12.6. AC INSTRUMENTATION TRANSDUCERS 403
(a) (b) (c)
Figure 12.47: Differential capacitive transducer varies capacitance ratio by changing: (a) area
of overlap, (b) distance between plates, (c) dielectric between plates.
As you can see, all of the differential devices shown in the above illustration have three wire
connections rather than two: one wire for each of the “end” plates and one for the “common”
plate. As the capacitance between one of the “end” plates and the “common” plate changes,
the capacitance between the other “end” plate and the “common” plate is such to change in the
opposite direction. This kind of transducer lends itself very well to implementation in a bridge
circuit: (Figure 12.48)
V
Pictoral diagram
Schematic diagram
sensor
capacitive
Figure 12.48: Differential capacitive transducer bridge measurement circuit.
Capacitive transducers provide relatively small capacitances for a measurement circuit to
operate with, typically in the picofarad range. Because of this, high power supply frequencies
(in the megahertz range!) are usually required to reduce these capacitive reactances to rea
sonable levels. Given the small capacitances provided by typical capacitive transducers, stray
capacitances have the potential of being major sources of measurement error. Good conductor
shielding is essential for reliable and accurate capacitive transducer circuitry!
The bridge circuit is not the only way to effectively interpret the differential capacitance
output of such a transducer, but it is one of the simplest to implement and understand. As
with the LVDT, the voltage output of the bridge is proportional to the displacement of the
transducer action from its center position, and the direction of offset will be indicated by phase
404 CHAPTER 12. AC METERING CIRCUITS
shift. This kind of bridge circuit is similar in function to the kind used with strain gauges: it
is not intended to be in a “balanced” condition all the time, but rather the degree of imbalance
represents the magnitude of the quantity being measured.
An interesting alternative to the bridge circuit for interpreting differential capacitance is
the twinT. It requires the use of diodes, those “oneway valves” for electric current mentioned
earlier in the chapter: (Figure 12.49)
R
R
Rload EoutC1 C2
+

+

Figure 12.49: Differential capacitive transducer “TwinT” measurement circuit.
This circuit might be better understood if redrawn to resemble more of a bridge configura
tion: (Figure 12.50)
RR
Rload
C1 C2
+

+

Figure 12.50: Differential capacitor transducer “TwinT” measurement circuit redrawn as a
bridge.Output is across Rload.
Capacitor C1 is charged by the AC voltage source during every positive halfcycle (positive
as measured in reference to the ground point), while C2 is charged during every negative half
cycle. While one capacitor is being charged, the other capacitor discharges (at a slower rate
than it was charged) through the threeresistor network. As a consequence, C1 maintains
a positive DC voltage with respect to ground, and C2 a negative DC voltage with respect to
12.6. AC INSTRUMENTATION TRANSDUCERS 405
ground.
If the capacitive transducer is displaced from center position, one capacitor will increase
in capacitance while the other will decrease. This has little effect on the peak voltage charge
of each capacitor, as there is negligible resistance in the charging current path from source to
capacitor, resulting in a very short time constant (τ ). However, when it comes time to discharge
through the resistors, the capacitor with the greater capacitance value will hold its charge
longer, resulting in a greater average DC voltage over time than the lesservalue capacitor.
The load resistor (Rload), connected at one end to the point between the two equalvalue
resistors (R) and at the other end to ground, will drop no DC voltage if the two capacitors’
DC voltage charges are equal in magnitude. If, on the other hand, one capacitor maintains a
greater DC voltage charge than the other due to a difference in capacitance, the load resistor
will drop a voltage proportional to the difference between these voltages. Thus, differential
capacitance is translated into a DC voltage across the load resistor.
Across the load resistor, there is both AC and DC voltage present, with only the DC voltage
being significant to the difference in capacitance. If desired, a lowpass filter may be added
to the output of this circuit to block the AC, leaving only a DC signal to be interpreted by
measurement circuitry: (Figure 12.51)
R
R
EoutC1 C2
+
 +

Lowpass
filter
Rfilter
CfilterRload
Figure 12.51: Addition of lowpass filter to “twinT” feeds pure DC to measurement indicator.
As a measurement circuit for differential capacitive sensors, the twinT configuration en
joys many advantages over the standard bridge configuration. First and foremost, transducer
displacement is indicated by a simple DC voltage, not an AC voltage whose magnitude and
phase must be interpreted to tell which capacitance is greater. Furthermore, given the proper
component values and power supply output, this DC output signal may be strong enough to
directly drive an electromechanical meter movement, eliminating the need for an amplifier
circuit. Another important advantage is that all important circuit elements have one terminal
directly connected to ground: the source, the load resistor, and both capacitors are all ground
referenced. This helps minimize the ill effects of stray capacitance commonly plaguing bridge
measurement circuits, likewise eliminating the need for compensatory measures such as the
Wagner earth.
This circuit is also easy to specify parts for. Normally, a measurement circuit incorporating
complementary diodes requires the selection of “matched” diodes for good accuracy. Not so with
406 CHAPTER 12. AC METERING CIRCUITS
this circuit! So long as the power supply voltage is significantly greater than the deviation in
voltage drop between the two diodes, the effects of mismatch are minimal and contribute little
to measurement error. Furthermore, supply frequency variations have a relatively low impact
on gain (how much output voltage is developed for a given amount of transducer displacement),
and squarewave supply voltage works as well as sinewave, assuming a 50% duty cycle (equal
positive and negative halfcycles), of course.
Personal experience with using this circuit has confirmed its impressive performance. Not
only is it easy to prototype and test, but its relative insensitivity to stray capacitance and its
high output voltage as compared to traditional bridge circuits makes it a very robust alterna
tive.
12.7 Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
Jason Starck (June 2000): HTML document formatting, which led to a much better
looking second edition.
Bibliography
[1] Waltt Kestler, “Position and Motion Sensors”, Analog De
vices. http://www.analog.com/UploadedFiles/Associated Docs/
324695618448506532114843952501435805318549066180119988Fsect6.PDF
Chapter 13
AC MOTORS
Contents
13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 408
13.1.1 Hysteresis and Eddy Current . . . . . . . . . . . . . . . . . . . . . . . . 409
13.2 Synchronous Motors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412
13.3 Synchronous condenser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 420
13.4 Reluctance motor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
13.4.1 Synchronous reluctance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
13.4.2 Switched reluctance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422
13.4.3 Electronic driven variable reluctance motor . . . . . . . . . . . . . . . . . 424
13.5 Stepper motors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 426
13.5.1 Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 426
13.5.2 Variable reluctance stepper . . . . . . . . . . . . . . . . . . . . . . . . . . 428
13.5.3 Permanent magnet stepper . . . . . . . . . . . . . . . . . . . . . . . . . . 431
13.5.4 Hybrid stepper motor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
13.6 Brushless DC motor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438
13.7 Tesla polyphase induction motors . . . . . . . . . . . . . . . . . . . . . . . . 442
13.7.1 Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
13.7.2 Theory of operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445
13.7.3 Nola power factor corrector . . . . . . . . . . . . . . . . . . . . . . . . . . 453
13.7.4 Induction motor alternator . . . . . . . . . . . . . . . . . . . . . . . . . . . 454
13.7.5 Motor starting and speed control . . . . . . . . . . . . . . . . . . . . . . . 455
13.7.6 Linear induction motor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460
13.8 Wound rotor induction motors . . . . . . . . . . . . . . . . . . . . . . . . . . 460
13.8.1 Speed control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 462
13.8.2 Doublyfed induction generator . . . . . . . . . . . . . . . . . . . . . . . . 462
13.9 Singlephase induction motors . . . . . . . . . . . . . . . . . . . . . . . . . . 464
13.9.1 Permanentsplit capacitor motor . . . . . . . . . . . . . . . . . . . . . . . 465
13.9.2 Capacitorstart induction motor . . . . . . . . . . . . . . . . . . . . . . . 466
407
408 CHAPTER 13. AC MOTORS
13.9.3 Capacitorrun motor induction motor . . . . . . . . . . . . . . . . . . . . 467
13.9.4 Resistance splitphase motor induction motor . . . . . . . . . . . . . . . . 467
13.9.5 Nola power factor corrrector . . . . . . . . . . . . . . . . . . . . . . . . . . 467
13.10 Other specialized motors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469
13.10.1 Shaded pole induction motor . . . . . . . . . . . . . . . . . . . . . . . . . 469
13.10.2 2phase servo motor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469
13.10.3 Hysteresis motor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 470
13.10.4 Eddy current clutch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 470
13.11 Selsyn (synchro) motors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471
13.11.1Transmitter  receiver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 472
13.11.2 Differential transmitter  receiver . . . . . . . . . . . . . . . . . . . . . . 473
13.11.3 Control transformer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 476
13.11.4 Resolver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 478
13.12 AC commutator motors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479
13.12.1 Single phase series motor . . . . . . . . . . . . . . . . . . . . . . . . . . . 480
13.12.2 Compensated series motor . . . . . . . . . . . . . . . . . . . . . . . . . . 480
13.12.3 Universal motor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 480
13.12.4 Repulsion motor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 481
13.12.5 Repulsion start induction motor . . . . . . . . . . . . . . . . . . . . . . . 481
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 482
Original author: Dennis Crunkilton
Figure 13.1: Conductors of squirrel cage induction motor removed from rotor.
13.1 Introduction
After the introduction of the DC electrical distribution system by Edison in the United States,
a gradual transition to the more economical AC system commenced. Lighting worked as well
on AC as on DC. Transmission of electrical energy covered longer distances at lower loss with
13.1. INTRODUCTION 409
alternating current. However, motors were a problem with alternating current. Initially, AC
motors were constructed like DC motors. Numerous problems were encountered due to chang
ing magnetic fields, as compared to the static fields in DC motor motor field coils.
Charles P. Steinmetz contributed to solving these problems with his investigation of hys
teresis losses in iron armatures. Nikola Tesla envisioned an entirely new type of motor when
he visualized a spinning turbine, not spun by water or steam, but by a rotating magnetic field.
His new type of motor, the AC induction motor, is the workhorse of industry to this day. Its
ruggedness and simplicity (Figure 13.1) make for long life, high reliability, and low mainte
nance. Yet small brushed AC motors, similar to the DC variety, persist in small appliances
along with small Tesla induction motors. Above one horsepower (750 W), the Tesla motor
reigns supreme.
Modern solid state electronic circuits drive brushless DC motors with AC waveforms gen
erated from a DC source. The brushless DC motor, actually an AC motor, is replacing the
conventional brushed DC motor in many applications. And, the stepper motor, a digital ver
sion of motor, is driven by alternating current square waves, again, generated by solid state
circuitry Figure 13.2 shows the family tree of the AC motors described in this chapter.
Cruise ships and other large vessels replace reduction geared drive shafts with large multi
megawatt generators and motors. Such has been the case with dieselelectric locomotives on a
smaller scale for many years.
At the system level, (Figure 13.3) a motor takes in electrical energy in terms of a potential
difference and a current flow, converting it to mechanical work. Alas, electric motors are not
100% efficient. Some of the electric energy is lost to heat, another form of energy, due to I2R
losses in the motor windings. The heat is an undesired byproduct of the conversion. It must
be removed from the motor and may adversely affect longevity. Thus, one goal is to maximize
motor efficiency, reducing the heat loss. AC motors also have some losses not encountered by
DC motors: hysteresis and eddy currents.
13.1.1 Hysteresis and Eddy Current
Early designers of AC motors encountered problems traced to losses unique to alternating cur
rent magnetics. These problems were encountered when adapting DC motors to AC operation.
Though few AC motors today bear any resemblance to DC motors, these problems had to be
solved before AC motors of any type could be properly designed before they were built.
Both rotor and stator cores of AC motors are composed of a stack of insulated laminations.
The laminations are coated with insulating varnish before stacking and bolting into the final
form. Eddy currents are minimized by breaking the potential conductive loop into smaller less
lossy segments. (Figure 13.4) The current loops look like shorted transformer secondary turns.
The thin isolated laminations break these loops. Also, the silicon (a semiconductor) added to
the alloy used in the laminations increases electrical resistance which decreases the magnitude
of eddy currents.
If the laminations are made of silicon alloy grain oriented steel, hysteresis losses are min
imized. Magnetic hysteresis is a lagging behind of magnetic field strength as compared to
magnetizing force. If a soft iron nail is temporarily magnetized by a solenoid, one would expect
the nail to lose the magnetic field once the solenoid is deenergized. However, a small amount
of residual magnetization, Br due to hysteresis remains. (Figure 13.5) An alternating current
has to expend energy, Hc the coercive force, in overcoming this residual magnetization before
410 CHAPTER 13. AC MOTORS
El
ec
tri
c
m
o
to
r
AC
D
C
As
yn
ch
ro
no
us
Po
lyp
ha
se
Si
ng
le
ph
as
e
Sq
ui
rre
l
ca
ge
W
ou
nd
ro
to
r
Sy
nc
hr
os
&
re
so
lv
er
s
Ca
pa
cit
or
st
ar
t
Pe
rm
an
en
t
sp
lit
ca
pa
cit
or
Sp
lit
ph
as
e
Ca
pa
cit
or
ru
n
Sh
ad
ed
po
le
Va
ria
bl
e
re
lu
ct
an
ce
El
ec
tri
c
m
ot
or
fa
m
ily
tr
ee
Sy
nc
hr
on
ou
s
co
n
de
ns
er
Si
ne
w
a
ve
Br
us
hl
es
s
St
ep
pe
r
R
el
uc
ta
nc
e
Va
ria
bl
e
re
lu
ct
an
ce
PM
H
yb
rid
Sw
itc
he
d
re
lu
ct
an
ce
Sy
nc
hr
on
ou
s
re
lu
ct
an
ce
Sy
nc
hr
on
ou
s
O
th
er
D
C
m
o
to
rs
W
ou
nd
ro
to
r
PM
ro
to
r
AC br
us
he
d
Un
ive
rs
al
Figure 13.2: AC electric motor family diagram.
13.1. INTRODUCTION 411
Heat
Electrical energy
Mechanical enegy
Figure 13.3: Motor system level diagram.
solid core laminated core
Figure 13.4: Eddy currents in iron cores.
it can magnetize the core back to zero, let alone in the opposite direction. Hysteresis loss is
encountered each time the polarity of the AC reverses. The loss is proportional to the area en
closed by the hysteresis loop on the BH curve. “Soft” iron alloys have lower losses than “hard”
high carbon steel alloys. Silicon grain oriented steel, 4% silicon, rolled to preferentially orient
the grain or crystalline structure, has still lower losses.
BR
BR
B Teslas
H
Aturns/m
HC
HC
B
H
low hysteresis loss high loss
Figure 13.5: Hysteresis curves for low and high loss alloys.
Once Steinmetz’s Laws of hysteresis could predict iron core losses, it was possible to design
412 CHAPTER 13. AC MOTORS
AC motors which performed as designed. This was akin to being able to design a bridge ahead
of time that would not collapse once it was actually built. This knowledge of eddy current
and hysteresis was first applied to building AC commutator motors similar to their DC coun
terparts. Today this is but a minor category of AC motors. Others invented new types of AC
motors bearing little resemblance to their DC kin.
13.2 Synchronous Motors
Single phase synchronous motors are available in small sizes for applications requiring precise
timing such as time keeping, (clocks) and tape players. Though battery powered quartz regu
lated clocks are widely available, the AC line operated variety has better long term accuracy−−
over a period of months. This is due to power plant operators purposely maintaining the long
term accuracy of the frequency of the AC distribution system. If it falls behind by a few cycles,
they will make up the lost cycles of AC so that clocks lose no time.
Above 10 Horsepower (10 kW) the higher efficiency and leading powerfactor make large
synchronous motors useful in industry. Large synchronous motors are a few percent more
efficient than the more common induction motors. Though, the synchronous motor is more
complex.
Since motors and generators are similar in construction, it should be possible to use a gen
erator as a motor, conversely, use a motor as a generator. A synchronous motor is similar to an
alternator with a rotating field. The figure below shows small alternators with a permanent
magnet rotating field. This figure 13.6 could either be two paralleled and synchronized alter
nators driven by a mechanical energy sources, or an alternator driving a synchronous motor.
Or, it could be two motors, if an external power source were connected. The point is that in
either case the rotors must run at the same nominal frequency, and be in phase with each
other. That is, they must be synchronized. The procedure for synchronizing two alternators
is to (1) open the switch, (2) drive both alternators at the same rotational rate, (3) advance or
retard the phase of one unit until both AC outputs are in phase, (4) close the switch before they
drift out of phase. Once synchronized, the alternators will be locked to each other, requiring
considerable torque to break one unit loose (out of synchronization) from the other.
N
S
N
S
N
S
1 2
N
S
N
S
3
N
S
torque angle
Figure 13.6: Synchronous motor running in step with alternator.
If more torque in the direction of rotation is applied to the rotor of one of the above rotating
13.2. SYNCHRONOUS MOTORS 413
alternators, the angle of the rotor will advance (opposite of (3)) with respect to the magnetic
field in the stator coils while still synchronized and the rotor will deliver energy to the AC
line like an alternator. The rotor will also be advanced with respect to the rotor in the other
alternator. If a load such as a brake is applied to one of the above units, the angle of the
rotor will lag the stator field as at (3), extracting energy from the AC line, like a motor. If
excessive torque or drag is applied, the rotor will exceed the maximum torque angle advancing
or lagging so much that synchronization is lost. Torque is developed only when synchronization
of the motor is maintained.
In the case of a small synchronous motor in place of the alternator Figure 13.6 right, it is
not necessary to go through the elaborate synchronization procedure for alternators. However,
the synchronous motor is not self starting and must still be brought up to the approximate
alternator electrical speed before it will lock (synchronize) to the generator rotational rate.
Once up to speed, the synchronous motor will maintain synchronism with the AC power source
and develop torque.
N
S
N
S
NS
N
S
N S
N
S
1 2 3
1
2
3
N
S
N
S
4
4
α
Figure 13.7: Sinewave drives synchronous motor.
Assuming that the motor is up to synchronous speed, as the sine wave changes to positive
in Figure 13.7 (1), the lower north coil pushes the north rotor pole, while the upper south coil
attracts that rotor north pole. In a similar manner the rotor south pole is repelled by the upper
south coil and attracted to the lower north coil. By the time that the sine wave reaches a peak
at (2), the torque holding the north pole of the rotor up is at a maximum. This torque decreases
as the sine wave decreases to 0 VDC at (3) with the torque at a minimum.
As the sine wave changes to negative between (3&4), the lower south coil pushes the south
rotor pole, while attracting rotor north rotor pole. In a similar manner the rotor north pole
is repelled by the upper north coil and attracted to the lower south coil. At (4) the sinewave
reaches a negative peak with holding torque again at a maximum. As the sine wave changes
from negative to 0 VDC to positive, The process repeats for a new cycle of sine wave.
Note, the above figure illustrates the rotor position for a noload condition (α=0o). In actual
practice, loading the rotor will cause the rotor to lag the positions shown by angle α. This
angle increases with loading until the maximum motor torque is reached at α=90o electrical.
Synchronization and torque are lost beyond this angle.
The current in the coils of a single phase synchronous motor pulsates while alternating
414 CHAPTER 13. AC MOTORS
polarity. If the permanent magnet rotor speed is close to the frequency of this alternation, it
synchronizes to this alternation. Since the coil field pulsates and does not rotate, it is necessary
to bring the permanent magnet rotor up to speed with an auxiliary motor. This is a small
induction motor similar to those in the next section.
N
S
Figure 13.8: Addition of field poles decreases speed.
A 2pole (pair of NS poles) alternator will generate a 60 Hz sine wave when rotated at
3600 rpm (revolutions per minute). The 3600 rpm corresponds to 60 revolutions per second.
A similar 2pole permanent magnet synchronous motor will also rotate at 3600 rpm. A lower
speed motor may be constructed by adding more pole pairs. A 4pole motor would rotate at
1800 rpm, a 12pole motor at 600 rpm. The style of construction shown (Figure 13.8) is for
illustration. Higher efficiency higher torque multipole stator synchronous motors actually
have multiple poles in the rotor.
Figure 13.9: Onewinding 12pole synchronous motor.
Rather than wind 12coils for a 12pole motor, wind a single coil with twelve interdigitated
steel poles pieces as shown in Figure 13.9. Though the polarity of the coil alternates due to the
appplied AC, assume that the top is temporarily north, the bottom south. Pole pieces route the
south flux from the bottom and outside of the coil to the top. These 6souths are interleaved
with 6north tabs bent up from the top of the steel pole piece of the coil. Thus, a permanent
magnet rotor bar will encounter 6pole pairs corresponding to 6cycles of AC in one physical
rotation of the bar magnet. The rotation speed will be 1/6 of the electrical speed of the AC.
Rotor speed will be 1/6 of that experienced with a 2pole synchronous motor. Example: 60 Hz
would rotate a 2pole motor at 3600 rpm, or 600 rpm for a 12pole motor.
The stator (Figure 13.10) shows a 12pole Westclox synchronous clock motor. Construc
tion is similar to the previous figure with a single coil. The one coil style of construction is
13.2. SYNCHRONOUS MOTORS 415
Figure 13.10: Reprinted by permission of Westclox History at www.clockHistory.com
economical for low torque motors. This 600 rpm motor drives reduction gears moving clock
hands.
If the Westclox motor were to run at 600 rpm from a 50 Hz power source, how many poles
would be required? A 10pole motor would have 5pairs of NS poles. It would rotate at 50/5 =
10 rotations per second or 600 rpm (10 s−1 x 60 s/minute.)
Figure 13.11: Reprinted by permission of Westclox History at www.clockHistory.com
The rotor (Figure 13.11) consists of a permanent magnet bar and a steel induction motor
cup. The synchronous motor bar rotating within the pole tabs keeps accurate time. The induc
tion motor cup outside of the bar magnet fits outside and over the tabs for self starting. At one
time nonselfstarting motors without the induction motor cup were manufactured.
A 3phase synchronous motor as shown in Figure 13.12 generates an electrically rotating
field in the stator. Such motors are not self starting if started from a fixed frequency power
416 CHAPTER 13. AC MOTORS
source such as 50 or 60 Hz as found in an industrial setting. Furthermore, the rotor is not a
permanent magnet as shown below for the multihorsepower (multikilowatt) motors used in
industry, but an electromagnet. Large industrial synchronous motors are more efficient than
induction motors. They are used when constant speed is required. Having a leading power
factor, they can correct the AC line for a lagging power factor.
The three phases of stator excitation add vectorially to produce a single resultant magnetic
field which rotates f/2n times per second, where f is the power line frequency, 50 or 60 Hz
for industrial power line operated motors. The number of poles is n. For rotor speed in rpm,
multiply by 60.
S = f120/n
where: S = rotor speed in rpm
f = AC line frequency
n = number of poles per phase
The 3phase 4pole (per phase) synchronous motor (Figure 13.12) will rotate at 1800 rpm
with 60 Hz power or 1500 rpm with 50 Hz power. If the coils are energized one at a time in
the sequence φ1, φ2, φ3, the rotor should point to the corresponding poles in turn. Since
the sine waves actually overlap, the resultant field will rotate, not in steps, but smoothly. For
example, when the φ1 and φ2 sinewaves coincide, the field will be at a peak pointing between
these poles. The bar magnet rotor shown is only appropriate for small motors. The rotor
with multiple magnet poles (below right) is used in any efficient motor driving a substantial
load. These will be slip ring fed electromagnets in large industrial motors. Large industrial
synchronous motors are self started by embedded squirrel cage conductors in the armature,
acting like an induction motor. The electromagnetic armature is only energized after the rotor
is brought up to near synchronous speed.
N
S
φ1
φ2
φ3
φ3
φ1
φ2
N
S S
N
Figure 13.12: Three phase, 4pole synchronous motor
Small multiphase synchronous motors (Figure 13.12) may be started by ramping the drive
frequency from zero to the final running frequency. The multiphase drive signals are gener
ated by electronic circuits, and will be square waves in all but the most demanding applications.
Such motors are known as brushless DC motors. True synchronous motors are driven by sine
waveforms. Two or three phase drive may be used by supplying the appropriate number of
windings in the stator. Only 3phase is shown above.
13.2. SYNCHRONOUS MOTORS 417
φ1
φ2
φ3
waveform
gen &
power
drive
torque
output
motor
hall effect
Figure 13.13: Electronic synchronous motor
The block diagram (Figure 13.13) shows the drive electronics associated with a low voltage
(12 VDC) synchronous motor. These motors have a position sensor integrated within the motor,
which provides a low level signal with a frequency proportional to the speed of rotation of the
motor. The position sensor could be as simple as as solid state magnetic field sensors such
as Hall effect devices providing commutation (armature current direction) timing to the drive
electronics The position sensor could be a high resolution angular sensor such as a resolver,
inductosyn (magnetic encoder), or an optical encoder.
If constant and accurate speed of rotation is required, (as for a disk drive) a tachometer
and phase locked loop may be included. (Figure 13.14) This tachometer signal, a pulse train
proportional to motor speed, is fed back to a phase locked loop, which compares the tachometer
frequency and phase to a stable reference frequency source such as a crystal oscillator.
phase
locked
loop
φ1
φ2
φ3
waveform
gen &
power
drive
tachometer
torque
output
motor
reference
frequency
position sensor
Figure 13.14: Phase locked loop controls synchronous motor speed.
A motor driven by square waves of current, as provided by simple hall effect sensors, is
known as a brushless DC motor. This type of motor has higher ripple torque torque variation
through a shaft revolution than a sine wave driven motor. This is not a problem for many
applications. Though, we are primarily interested in synchronous motors in this section.
Ripple torque, or cogging is caused by magnetic attraction of the rotor poles to the stator
pole pieces. (Figure 13.15) Note that there are no stator coils, not even a motor. The PM rotor
may be rotated by hand but will encounter attraction to the pole pieces when near them. This
is analogous to the mechanical situation. Would ripple torque be a problem for a motor used in
a tape player? Yes, we do not want the motor to alternately speed and slow as it moves audio
tape past a tape playback head. Would ripple torque be a problem for a fan motor? No.
If a motor is driven by sinewaves of current synchronous with the motor back emf, it is
classified as a synchronous AC motor, regardless of whether the drive waveforms are generated
by electronic means. A synchronous motor will generate a sinusoidal back emf if the stator
418 CHAPTER 13. AC MOTORS
N
S
Ripple torque mechanical analog
Figure 13.15: Motor ripple torque and mechanical analog.
3φ distributed winding Single phase belt
c’
b’
a’
a
b
c
d
a
b
c
d
c’
b’
a’
ph1 ph1’
ph1’ ph1ph2ph2’ ph3ph3’
Figure 13.16: Windings distributed in a belt produce a more sinusoidal field.
13.2. SYNCHRONOUS MOTORS 419
magnetic field has a sinusoidal distribution. It will be more sinusoidal if pole windings are
distributed in a belt (Figure 13.16) across many slots instead of concentrated on one large pole
(as drawn in most of our simplified illustrations). This arrangement cancels many of the stator
field odd harmonics. Slots having fewer windings at the edge of the phase winding may share
the space with other phases. Winding belts may take on an alternate concentric form as shown
in Figure 13.78.
For a 2phase motor, driven by a sinewave, the torque is constant throughout a revolution
by the trigonometric identity:
sin2θ + cos2θ = 1
The generation and synchronization of the drive waveform requires a more precise rotor
position indication than provided by the hall effect sensors used in brushless DC motors. A
resolver, or optical or magnetic encoder provides resolution of hundreds to thousands of parts
(pulses) per revolution. A resolver provides analog angular position signals in the form of
signals proportional to the sine and cosine of shaft angle. Encoders provide a digital angular
position indication in either serial or parallel format. The sine wave drive may actually be
from a PWM, Pulse Width Modulator, a high efficiency method of approximating a sinewave
with a digital waveform. (Figure 13.17) Each phase requires drive electronics for this wave
form phaseshifted by the appropriate amount per phase.
PWM
Figure 13.17: PWM approximates a sinewave.
Synchronous motor efficiency is higher than that of induction motors. The synchronous
motor can also be smaller, especially if high energy permanent magnets are used in the rotor.
The advent of modern solid state electronics makes it possible to drive these motors at variable
speed. Induction motors are mostly used in railway traction. However, a small synchronous
motor, which mounts inside a drive wheel, makes it attractive for such applications. The high
temperature superconducting version of this motor is one fifth to one third the weight of a cop
per wound motor.[1] The largest experimental superconducting synchronous motor is capable
of driving a naval destroyer class ship. In all these applications the electronic variable speed
drive is essential.
420 CHAPTER 13. AC MOTORS
The variable speed drive must also reduce the drive voltage at low speed due to decreased
inductive reactance at lower frequency. To develop maximum torque, the rotor needs to lag the
stator field direction by 90o. Any more, it loses synchronization. Much less results in reduced
torque. Thus, the position of the rotor needs to be known accurately. And the position of the
rotor with respect to the stator field needs to be calculated, and controlled. This type of control
is known as vector phase control. It is implemented with a fast microprocessor driving a pulse
width modulator for the stator phases.
The stator of a synchronous motor is the same as that of the more popular induction motor.
As a result the industrial grade electronic speed control used with induction motors is also
applicable to large industrial synchronous motors.
If the rotor and stator of a conventional rotary synchronous motor are unrolled, a syn
chronous linear motor results. This type of motor is applied to precise high speed linear
positioning.[2]
A larger version of the linear synchronous motor with a movable carriage containing high
energy NdBFe permanent magnets is being developed to launch aircraft from naval aricraft
carriers.[3]
13.3 Synchronous condenser
Synchronous motors load the power line with a leading power factor. This is often usefull
in cancelling out the more commonly encountered lagging power factor caused by induction
motors and other inductive loads. Originally, large industrial synchronous motors came into
wide use because of this ability to correct the lagging power factor of induction motors.
This leading power factor can be exaggerated by removing the mechanical load and over
exciting the field of the synchronous motor. Such a device is known as a synchronous condenser.
Furthermore, the leading power factor can be adjusted by varying the field excitation. This
makes it possible to nearly cancel an arbitrary lagging power factor to unity by paralleling the
lagging load with a synchronous motor. A synchronous condenser is operated in a borderline
condition between a motor and a generator with no mechanical load to fulfill this function. It
can compensate either a leading or lagging power factor, by absorbing or supplying reactive
power to the line. This enhances power line voltage regulation.
Since a synchronous condenser does not supply a torque, the output shaft may be dispensed
with and the unit easily enclosed in a gas tight shell. The synchronous condenser may then be
filled with hydrogen to aid cooling and reduce windage losses. Since the density of hydrogen is
7% of that of air, the windage loss for a hydrogen filled unit is 7% of that encountered in air.
Furthermore, the thermal conductivity of hydrogen is ten times that of air. Thus, heat removal
is ten times more efficient. As a result, a hydrogen filled synchronous condenser can be driven
harder than an air cooled unit, or it may be physically smaller for a given capacity. There is
no explosion hazard as long as the hydrogen concentration is maintained above 70%, typically
above 91%.
The efficiency of long power transmission lines may be increased by placing synchronous
condensers along the line to compensate lagging currents caused by line inductance. More real
power may be transmitted through a fixed size line if the power factor is brought closer to unity
by synchronous condensers absorbing reactive power.
13.4. RELUCTANCE MOTOR 421
The ability of synchronous condensers to absorb or produce reactive power on a transient
basis stabilizes the power grid against short circuits and other transient fault conditions. Tran
sient sags and dips of milliseconds duration are stabilized. This supplements longer response
times of quick acting voltage regulation and excitation of generating equipment. The syn
chronous condenser aids voltage regulation by drawing leading current when the line voltage
sags, which increases generator excitation thereby restoring line voltage. (Figure 13.18) A
capacitor bank does not have this ability.
Line current
G
en
er
at
or
V
ol
ta
ge
100%
80%
60%
40%
20%
20% 40% 60% 80% 100%
synchronous condenser
w/o synchronous condenser
Figure 13.18: Synchronous condenser improves power line voltage regulation.
The capacity of a synchronous condenser can be increased by replacing the copper wound
iron field rotor with an ironless rotor of high temperature superconducting wire, which must be
cooled to the liquid nitrogen boiling point of 77oK (196oC). The superconducting wire carries
160 times the current of comparable copper wire, while producing a flux density of 3 Teslas or
higher. An iron core would saturate at 2 Teslas in the rotor air gap. Thus, an iron core, approx
imate µr=1000, is of no more use than air, or any other material with a relative permeability
µr=1, in the rotor. Such a machine is said to have considerable additional transient ability to
supply reactive power to troublesome loads like metal melting arc furnaces. The manufacturer
describes it as being a “reactive power shock absorber”. Such a synchronous condenser has a
higher power density (smaller physically) than a switched capacitor bank. The ability to ab
sorb or produce reactive power on a transient basis stabilizes the overall power grid against
fault conditions.
13.4 Reluctance motor
The variable reluctance motor is based on the principle that an unrestrained piece of iron
will move to complete a magnetic flux path with minimum reluctance, the magnetic analog of
electrical resistance. (Figure 13.19)
13.4.1 Synchronous reluctance
If the rotating field of a large synchronous motor with salient poles is deenergized, it will
still develop 10 or 15% of synchronous torque. This is due to variable reluctance throughout
422 CHAPTER 13. AC MOTORS
a rotor revolution. There is no practical application for a large synchronous reluctance motor.
However, it is practical in small sizes.
If slots are cut into the conductorless rotor of an induction motor, corresponding to the
stator slots, a synchronous reluctance motor results. It starts like an induction motor but
runs with a small amount of synchronous torque. The synchronous torque is due to changes in
reluctance of the magnetic path from the stator through the rotor as the slots align. This motor
is an inexpensive means of developing a moderate synchronous torque. Low power factor, low
pullout torque, and low efficiency are characteristics of the direct power line driven variable
reluctance motor. Such was the status of the variable reluctance motor for a century before the
development of semiconductor power control.
13.4.2 Switched reluctance
If an iron rotor with poles, but without any conductors, is fitted to a multiphase stator, a
switched reluctance motor, capable of synchronizing with the stator field results. When a sta
tor coil pole pair is energized, the rotor will move to the lowest magnetic reluctance path.
(Figure 13.19) A switched reluctance motor is also known as a variable reluctance motor. The
reluctance of the rotor to stator flux path varies with the position of the rotor.
φ3
high reluctance low reluctance
S S
N N
φ3
Figure 13.19: Reluctance is a function of rotor position in a variable reluctance motor.
Sequential switching (Figure 13.20) of the stator phases moves the rotor from one position
to the next. The mangetic flux seeks the path of least reluctance, the magnetic analog of electric
resistance. This is an over simplified rotor and waveforms to illustrate operation.
If one end of each 3phase winding of the switched reluctance motor is brought out via a
common lead wire, we can explain operation as if it were a stepper motor. (Figure 13.20) The
other coil connections are successively pulled to ground, one at a time, in a wave drive pattern.
This attracts the rotor to the clockwise rotating magnetic field in 60o increments.
Various waveforms may drive variable reluctance motors. (Figure 13.21) Wave drive (a)
is simple, requiring only a single ended unipolar switch. That is, one which only switches in
one direction. More torque is provided by the bipolar drive (b), but requires a bipolar switch.
The power driver must pull alternately high and low. Waveforms (a & b) are applicable to the
stepper motor version of the variable reluctance motor. For smooth vibration free operation
the 6step approximation of a sine wave (c) is desirable and easy to generate. Sine wave drive
(d) may be generated by a pulse width modulator (PWM), or drawn from the power line.
13.4. RELUCTANCE MOTOR 423
φ1
φ2
φ3
φ1
φ2φ3
V+
φ3
φ1
φ2
φ1
φ2
φ3
Figure 13.20: Variable reluctance motor, oversimplified operation.
φ1
φ2
φ3 φ3
φ1
φ2
φ3
φ1
φ2
φ1
φ2
φ3 (a)
(b) (d)
(c)
Figure 13.21: Variable reluctance motor drive waveforms: (a) unipolar wave drive, (b) bipolar
full step (c) sinewave (d) bipolar 6step.
424 CHAPTER 13. AC MOTORS
Doubling the number of stator poles decreases the rotating speed and increases torque.
This might eliminate a gear reduction drive. A variable reluctance motor intended to move in
discrete steps, stop, and start is a variable reluctance stepper motor, covered in another section.
If smooth rotation is the goal, there is an electronic driven version of the switched reluctance
motor. Variable reluctance motors or steppers actually use rotors like those in Figure 13.22.
13.4.3 Electronic driven variable reluctance motor
Variable reluctance motors are poor performers when direct power line driven. However, mi
croprocessors and solid state power drive makes this motor an economical high performance
solution in some high volume applications.
Though difficult to control, this motor is easy to spin. Sequential switching of the field coils
creates a rotating magnetic field which drags the irregularly shaped rotor around with it as
it seeks out the lowest magnetic reluctance path. The relationship between torque and stator
current is highly nonlinear– difficult to control.
φ1
φ2
φ3
φ1
φ2
φ3
φ1
φ2
φ3
φ3
φ1
φ2
φ1
φ2
φ3
φ3
φ1
φ2
Figure 13.22: Electronic driven variable reluctance motor.
An electronic driven variable reluctance motor (Figure 13.23) resembles a brushless DC
motor without a permanent magnet rotor. This makes the motor simple and inexpensive.
However, this is offset by the cost of the electronic control, which is not nearly as simple as
that for a brushless DC motor.
While the variable reluctance motor is simple, even more so than an induction motor, it is
difficult to control. Electronic control solves this problem and makes it practical to drive the
motor well above and below the power line frequency. A variable reluctance motor driven by
a servo, an electronic feedback system, controls torque and speed, minimizing ripple torque.
Figure 13.23
This is the opposite of the high ripple torque desired in stepper motors. Rather than a
stepper, a variable reluctance motor is optimized for continuous high speed rotation with min
imum ripple torque. It is necessary to measure the rotor position with a rotary position sensor
like an optical or magnetic encoder, or derive this from monitoring the stator back EMF. A
microprocessor performs complex calculations for switching the windings at the proper time
with solid state devices. This must be done precisely to minimize audible noise and ripple
13.4. RELUCTANCE MOTOR 425
rotor position
stator current
µprocessor
control
stator
drive
variable
reluctance
motor
Figure 13.23: Electronic driven variable reluctance motor.
torque. For lowest ripple torque, winding current must be monitored and controlled. The strict
drive requirements make this motor only practical for high volume applications like energy
efficient vacuum cleaner motors, fan motors, or pump motors. One such vacuum cleaner uses
a compact high efficiency electronic driven 100,000 rpm fan motor. The simplicity of the motor
compensates for the drive electronics cost. No brushes, no commutator, no rotor windings, no
permanent magnets, simplifies motor manufacture. The efficiency of this electronic driven mo
tor can be high. But, it requires considerable optimization, using specialized design techniques,
which is only justified for large manufacturing volumes.
Advantages
• Simple construction no brushes, commutator, or permanent magnets, no Cu or Al in the
rotor.
• High efficiency and reliability compared to conventional AC or DC motors.
• High starting torque.
• Cost effective compared to bushless DC motor in high volumes.
• Adaptable to very high ambient temperature.
• Low cost accurate speed control possible if volume is high enough.
Disadvantages
• Current versus torque is highly nonlinear
• Phase switching must be precise to minimize ripple torque
• Phase current must be controlled to minimize ripple torque
• Acoustic and electrical noise
• Not applicable to low volumes due to complex control issues
426 CHAPTER 13. AC MOTORS
13.5 Stepper motors
A stepper motor is a “digital” version of the electric motor. The rotor moves in discrete steps
as commanded, rather than rotating continuously like a conventional motor. When stopped
but energized, a stepper (short for stepper motor) holds its load steady with a holding torque.
Wide spread acceptance of the stepper motor within the last two decades was driven by the
ascendancy of digital electronics. Modern solid state driver electronics was a key to its success.
And, microprocessors readily interface to stepper motor driver circuits.
Application wise, the predecessor of the stepper motor was the servo motor. Today this
is a higher cost solution to high performance motion control applications. The expense and
complexity of a servomotor is due to the additional system components: position sensor and
error amplifier. (Figure 13.24) It is still the way to position heavy loads beyond the grasp
of lower power steppers. High acceleration or unusually high accuracy still requires a servo
motor. Otherwise, the default is the stepper due to low cost, simple drive electronics, good
accuracy, good torque, moderate speed, and low cost.
servo motorstepper motor load position
sensor
error load
command
command
Figure 13.24: Stepper motor vs servo motor.
A stepper motor positions the readwrite heads in a floppy drive. They were once used for
the same purpose in harddrives. However, the high speed and accuracy required of modern
harddrive head positioning dictates the use of a linear servomotor (voice coil).
The servo amplifier is a linear amplifier with some difficult to integrate discrete compo
nents. A considerable design effort is required to optimize the servo amplifier gain vs phase
response to the mechanical components. The stepper motor drivers are less complex solid state
switches, being either “on” or “off”. Thus, a stepper motor controller is less complex and costly
than a servo motor controller.
Slosyn synchronous motors can run from AC line voltage like a singlephase permanent
capacitor induction motor. The capacitor generates a 90o second phase. With the direct line
voltage, we have a 2phase drive. Drive waveforms of bipolar (±) square waves of 224V are
more common these days. The bipolar magnetic fields may also be generated from unipolar
(one polarity) voltages applied to alternate ends of a center tapped winding. (Figure 13.25) In
other words, DC can be switched to the motor so that it sees AC. As the windings are energized
in sequence, the rotor synchronizes with the consequent stator magnetic field. Thus, we treat
stepper motors as a class of AC synchronous motor.
13.5.1 Characteristics
Stepper motors are rugged and inexpensive because the rotor contains no winding slip rings,
or commutator. The rotor is a cylindrical solid, which may also have either salient poles or
13.5. STEPPER MOTORS 427
V
V+ V
V+
V+
unipolarbipolar
Figure 13.25: Unipolar drive of center tapped coil at (b), emulates AC current in single coil at
(a).
fine teeth. More often than not the rotor is a permanent magnet. Determine that the rotor
is a permanent magnet by unpowered hand rotation showing detent torque, torque pulsations.
Stepper motor coils are wound within a laminated stator, except for can stack construction.
There may be as few as two winding phases or as many as five. These phases are frequently
split into pairs. Thus, a 4pole stepper motor may have two phases composed of inline pairs of
poles spaced 90o apart. There may also be multiple pole pairs per phase. For example a 12pole
stepper has 6pairs of poles, three pairs per phase.
Since stepper motors do not necessarily rotate continuously, there is no horsepower rating.
If they do rotate continuously, they do not even approach a subfractional hp rated capability.
They are truly small low power devices compared to other motors. They have torque ratings to
a thousand inoz (inchounces) or ten nm (newtonmeters) for a 4 kg size unit. A small “dime”
size stepper has a torque of a hundredth of a newtonmeter or a few inchounces. Most steppers
are a few inches in diameter with a fraction of a nm or a few inoz torque. The torque available
is a function of motor speed, load inertia, load torque, and drive electronics as illustrated on
the speed vs torque curve. (Figure 13.26) An energized, holding stepper has a relatively high
holding torque rating. There is less torque available for a running motor, decreasing to zero at
some high speed. This speed is frequently not attainable due to mechanical resonance of the
motor load combination.
Speed
To
rq
ue
cutoff speed
maximum speed
holding torque
Figure 13.26: Stepper speed characteristics.
Stepper motors move one step at a time, the step angle, when the drive waveforms are
changed. The step angle is related to motor construction details: number of coils, number
of poles, number of teeth. It can be from 90o to 0.75o, corresponding to 4 to 500 steps per
revolution. Drive electronics may halve the step angle by moving the rotor in halfsteps.
Steppers cannot achieve the speeds on the speed torque curve instantaneously. The max
imum start frequency is the highest rate at which a stopped and unloaded stepper can be
428 CHAPTER 13. AC MOTORS
started. Any load will make this parameter unattainable. In practice, the step rate is ramped
up during starting from well below the maximum start frequency. When stopping a stepper
motor, the step rate may be decreased before stopping.
The maximum torque at which a stepper can start and stop is the pullin torque. This torque
load on the stepper is due to frictional (brake) and inertial (flywheel) loads on the motor shaft.
Once the motor is up to speed, pullout torque is the maximum sustainable torque without
losing steps.
There are three types of stepper motors in order of increasing complexity: variable reluc
tance, permanent magnet, and hybrid. The variable reluctance stepper has s solid soft steel
rotor with salient poles. The permanent magnet stepper has a cylindrical permanent mag
net rotor. The hybrid stepper has soft steel teeth added to the permanent magnet rotor for a
smaller step angle.
13.5.2 Variable reluctance stepper
A variable reluctance stepper motor relies upon magnetic flux seeking the lowest reluctance
path through a magnetic circuit. This means that an irregularly shaped soft magnetic rotor
will move to complete a magnetic circuit, minimizing the length of any high reluctance air gap.
The stator typically has three windings distributed between pole pairs , the rotor four salient
poles, yielding a 30o step angle.(Figure 13.27) A deenergized stepper with no detent torque
when hand rotated is identifiable as a variable reluctance type stepper.
φ1
φ2
φ3
φ2
φ1
φ3
φ1
φ2φ3
15o step30o step
V+
φ4
φ1
φ2
φ3
φ4
V+
Figure 13.27: Three phase and four phase variable reluctance stepper motors.
The drive waveforms for the 3φ stepper can be seen in the “Reluctance motor” section. The
drive for a 4φ stepper is shown in Figure 13.28. Sequentially switching the stator phases
produces a rotating magnetic field which the rotor follows. However, due to the lesser number
of rotor poles, the rotor moves less than the stator angle for each step. For a variable reluctance
stepper motor, the step angle is given by:
ΘS = 360o/NS
ΘR = 360o/NR
ΘST = ΘR  ΘS
where: ΘS = stator angle, ΘR = Rotor angle, ΘST = step
angle
NS = number stator poles, NP = number rotor poles
13.5. STEPPER MOTORS 429
φ1
φ1
φ2
φ3
φ4
V+
φ3φ2
φ4
Ν
S
Ν
Ν
Ν
S
S
S
φ1
φ2
φ3
φ4
φ1
φ2
φ3
φ4
counterclockwise 15o step reverse step, clockwise
Figure 13.28: Stepping sequence for variable reluctance stepper.
In Figure 13.28, moving from φ1 to φ2, etc., the stator magnetic field rotates clockwise. The
rotor moves counterclockwise (CCW). Note what does not happen! The dotted rotor tooth does
not move to the next stator tooth. Instead, the φ2 stator field attracts a different tooth in
moving the rotor CCW, which is a smaller angle (15o) than the stator angle of 30o. The rotor
tooth angle of 45o enters into the calculation by the above equation. The rotor moved CCW to
the next rotor tooth at 45o, but it aligns with a CW by 30o stator tooth. Thus, the actual step
angle is the difference between a stator angle of 45o and a rotor angle of 30o . How far would
the stepper rotate if the rotor and stator had the same number of teeth? Zero– no notation.
Starting at rest with phase φ1 energized, three pulses are required (φ2, φ3, φ4) to align the
“dotted” rotor tooth to the next CCW stator Tooth, which is 45o. With 3pulses per stator tooth,
and 8stator teeth, 24pulses or steps move the rotor through 360o.
By reversing the sequence of pulses, the direction of rotation is reversed above right. The
direction, step rate, and number of steps are controlled by a stepper motor controller feeding
a driver or amplifier. This could be combined into a single circuit board. The controller could
be a microprocessor or a specialized integrated circuit. The driver is not a linear amplifier, but
a simple onoff switch capable of high enough current to energize the stepper. In principle,
the driver could be a relay or even a toggle switch for each phase. In practice, the driver is
either discrete transistor switches or an integrated circuit. Both driver and controller may be
combined into a single integrated circuit accepting a direction command and step pulse. It
outputs current to the proper phases in sequence.
Disassemble a reluctance stepper to view the internal components. Otherwise, we show
the internal construction of a variable reluctance stepper motor in Figure 13.29. The rotor has
protruding poles so that they may be attracted to the rotating stator field as it is switched. An
actual motor, is much longer than our simplified illustration.
The shaft is frequently fitted with a drive screw. (Figure 13.30) This may move the heads
of a floppy drive upon command by the floppy drive controller.
Variable reluctance stepper motors are applied when only a moderate level of torque is
required and a coarse step angle is adequate. A screw drive, as used in a floppy disk drive
is such an application. When the controller powersup, it does not know the position of the
carriage. However, it can drive the carriage toward the optical interrupter, calibrating the
430 CHAPTER 13. AC MOTORS
Figure 13.29: Variable reluctance stepper motor.
stepper
motor
optical
interrupter
knife edge
guide rails carriage
Figure 13.30: Variable reluctance stepper drives lead screw.
13.5. STEPPER MOTORS 431
position at which the knife edge cuts the interrupter as “home”. The controller counts step
pulses from this position. As long as the load torque does not exceed the motor torque, the
controller will know the carriage position.
Summary: variable reluctance stepper motor
• The rotor is a soft iron cylinder with salient (protruding) poles.
• This is the least complex, most inexpensive stepper motor.
• The only type stepper with no detent torque in hand rotation of a deenergized motor
shaft.
• Large step angle
• A lead screw is often mounted to the shaft for linear stepping motion.
13.5.3 Permanent magnet stepper
A permanent magnet stepper motor has a cylindrical permanent magnet rotor. The stator usu
ally has two windings. The windings could be center tapped to allow for a unipolar driver
circuit where the polarity of the magnetic field is changed by switching a voltage from one
end to the other of the winding. A bipolar drive of alternating polarity is required to power
windings without the center tap. A pure permanent magnet stepper usually has a large step
angle. Rotation of the shaft of a deenergized motor exhibits detent torque. If the detent angle
is large, say 7.5o to 90o, it is likely a permanent magnet stepper rather than a hybrid stepper
(next subsection).
Permanent magnet stepper motors require phased alternating currents applied to the two
(or more) windings. In practice, this is almost always square waves generated from DC by
solid state electronics. Bipolar drive is square waves alternating between (+) and () polarities,
say, +2.5 V to 2.5 V. Unipolar drive supplies a (+) and () alternating magnetic flux to the coils
developed from a pair of positive square waves applied to opposite ends of a center tapped coil.
The timing of the bipolar or unipolar wave is wave drive, full step, or half step.
Wave drive
NS
a b c
N
S
d
Sφ−1
φ−2
φ−1
φ−2
N N S
N
S
φ−1
φ−2
φ−1
φ−2
S N
− +
+
−
+
−
+ −
SN
N
S
Wave drive
Figure 13.31: PM wave drive sequence (a) φ1+ , (b) φ2+ , (c) φ1 , (d) φ2.
432 CHAPTER 13. AC MOTORS
Conceptually, the simplest drive is wave drive. (Figure 13.31) The rotation sequence left to
right is positive φ1 points rotor north pole up, (+) φ2 points rotor north right, negative φ1
attracts rotor north down, () φ2 points rotor left. The wave drive waveforms below show that
only one coil is energized at a time. While simple, this does not produce as much torque as
other drive techniques.
+

+

a b c d
φ2
φ1
φ1’
φ1
φ2’φ2
Figure 13.32: Waveforms: bipolar wave drive.
The waveforms (Figure 13.32) are bipolar because both polarities , (+) and () drive the
stepper. The coil magnetic field reverses because the polarity of the drive current reverses.
φ1φ2
V+
φ1
φ2
φ1’
φ2’
φ1’
φ2’V+
6wire
Figure 13.33: Waveforms: unipolar wave drive.
The (Figure 13.33) waveforms are unipolar because only one polarity is required. This
simplifies the drive electronics, but requires twice as many drivers. There are twice as many
waveforms because a pair of (+) waves is required to produce an alternating magnetic field by
application to opposite ends of a center tapped coil. The motor requires alternating magnetic
fields. These may be produced by either unipolar or bipolar waves. However, motor coils must
have center taps for unipolar drive.
Permanent magnet stepper motors are manufactured with various leadwire configurations.
(Figure 13.34)
φ1φ2
V+
φ1’
φ2’V+ φ1φ2
φ1’
φ2’V+
6wire
unipolar
5wire
unipolar
φ1a
φ2a
φ1b’
φ2b’
8wire bipolar
or unipolar
φ1a’
φ1b
φ2a’ φ2b’
φ1’
φ1
φ2’φ2
2wire
bipolar
Figure 13.34: Stepper motor wiring diagrams.
The 4wire motor can only be driven by bipolar waveforms. The 6wire motor, the most
13.5. STEPPER MOTORS 433
common arrangement, is intended for unipolar drive because of the center taps. Though, it
may be driven by bipolar waves if the center taps are ignored. The 5wire motor can only be
driven by unipolar waves, as the common center tap interferes if both windings are energized
simultaneously. The 8wire configuration is rare, but provides maximum flexibility. It may be
wired for unipolar drive as for the 6wire or 5wire motor. A pair of coils may be connected
in series for high voltage bipolar low current drive, or in parallel for low voltage high current
drive.
A bifilar winding is produced by winding the coils with two wires in parallel, often a red and
green enamelled wire. This method produces exact 1:1 turns ratios for center tapped windings.
This winding method is applicable to all but the 4wire arrangement above.
Full step drive
Full step drive provides more torque than wave drive because both coils are energized at the
same time. This attracts the rotor poles midway between the two field poles. (Figure 13.35)
a b c
N
S
d
Sφ−1
φ−2
φ−1
φ−2
N
φ−1
φ−2
φ−1
φ−2
S N
− +
+
−
+
−
+ −
SN
N
S
+

+

0
0
a b c d
φ−1
φ−2
N
SN
NS
S
S N
S
N
+ −
+
−
+
−
+−
N S
N
S
Figure 13.35: Full step, bipolar drive.
Full step bipolar drive as shown in Figure 13.35 has the same step angle as wave drive.
Unipolar drive (not shown) would require a pair of unipolar waveforms for each of the above
bipolar waveforms applied to the ends of a center tapped winding. Unipolar drive uses a less
complex, less expensive driver circuit. The additional cost of bipolar drive is justified when
more torque is required.
Half step drive
The step angle for a given stepper motor geometry is cut in half with half step drive. This
corresponds to twice as many step pulses per revolution. (Figure 13.36) Half stepping provides
greater resolution in positioning of the motor shaft. For example, half stepping the motor
moving the print head across the paper of an inkjet printer would double the dot density.
Half step drive is a combination of wave drive and full step drive with one winding ener
gized, followed by both windings energized, yielding twice as many steps. The unipolar wave
forms for half step drive are shown above. The rotor aligns with the field poles as for wave
drive and between the poles as for full step drive.
434 CHAPTER 13. AC MOTORS
a
N
S
Sφ−1
φ−2
+
−
N
++

+

0
0
a b c d
b
φ−1
φ−2
N
− +
S
N
S
S
N
+
−
dc
φ−1
φ−2
+
−
S
N
N
S
+−
N SNS
φ−1
φ−2
N
− +
S
Half step
Figure 13.36: Half step, bipolar drive.
Microstepping is possible with specialized controllers. By varying the currents to the wind
ings sinusoidally many microsteps can be interpolated between the normal positions.
Construction
The contruction of a permanent magnet stepper motor is considerably different from the draw
ings above. It is desirable to increase the number of poles beyond that illustrated to produce
a smaller step angle. It is also desirable to reduce the number of windings, or at least not
increase the number of windings for ease of manufacture.
north
south
ceramic permanent magnet
rotor
φ1 coil φ2 coil
SN
Figure 13.37: Permanent magnet stepper motor, 24pole canstack construction.
The permanent magnet stepper (Figure 13.37) only has two windings, yet has 24poles in
each of two phases. This style of construction is known as can stack. A phase winding is
wrapped with a mild steel shell, with fingers brought to the center. One phase, on a tran
sient basis, will have a north side and a south side. Each side wraps around to the center
13.5. STEPPER MOTORS 435
of the doughnut with twelve interdigitated fingers for a total of 24 poles. These alternating
northsouth fingers will attract the permanent magnet rotor. If the polarity of the phase were
reversed, the rotor would jump 360o/24 = 15o. We do not know which direction, which is not
usefull. However, if we energize φ1 followed by φ2, the rotor will move 7.5o because the φ2
is offset (rotated) by 7.5o from φ1. See below for offset. And, it will rotate in a reproducible
direction if the phases are alternated. Application of any of the above waveforms will rotate
the permanent magnet rotor.
Note that the rotor is a gray ferrite ceramic cylinder magnetized in the 24pole pattern
shown. This can be viewed with magnet viewer film or iron filings applied to a paper wrapping.
Though, the colors will be green for both north and south poles with the film.
Can stack permanent magnet stepper
dust coverφ−1 coil
φ−2 coil
PM rotor
90° offset
φ−1 coil
φ−2 coil
xx’
x’
x
Figure 13.38: (a) External view of can stack, (b) field offset detail.
Canstack style construction of a PM stepper is distinctive and easy to identify by the
stacked “cans”. (Figure 13.38) Note the rotational offset between the two phase sections. This
is key to making the rotor follow the switching of the fields between the two phases.
Summary: permanent magnet stepper motor
• The rotor is a permanent magnet, often a ferrite sleeve magnetized with numerous poles.
• Canstack construction provides numerous poles from a single coil with interleaved fin
gers of soft iron.
• Large to moderate step angle.
• Often used in computer printers to advance paper.
13.5.4 Hybrid stepper motor
The hybrid stepper motor combines features of both the variable reluctance stepper and the
permanent magnet stepper to produce a smaller step angle. The rotor is a cylindrical perma
436 CHAPTER 13. AC MOTORS
nent magnet, magnetized along the axis with radial soft iron teeth (Figure 13.39). The stator
coils are wound on alternating poles with corresponding teeth. There are typically two winding
phases distributed between pole pairs. This winding may be center tapped for unipolar drive.
The center tap is achieved by a bifilar winding, a pair of wires wound physically in parallel, but
wired in series. The northsouth poles of a phase swap polarity when the phase drive current
is reversed. Bipolar drive is required for untapped windings.
N S
N S
rotor pole detail
permanent magnet
rotor, 96pole
8pole stator
Figure 13.39: Hybrid stepper motor.
Note that the 48teeth on one rotor section are offset by half a pitch from the other. See
rotor pole detail above. This rotor tooth offset is also shown below. Due to this offset, the rotor
effectively has 96 interleaved poles of opposite polarity. This offset allows for rotation in 1/96
th of a revolution steps by reversing the field polarity of one phase. Two phase windings are
common as shown above and below. Though, there could be as many as five phases.
The stator teeth on the 8poles correspond to the 48rotor teeth, except for missing teeth in
the space between the poles. Thus, one pole of the rotor, say the south pole, may align with the
stator in 48 distinct positions. However, the teeth of the south pole are offset from the north
teeth by half a tooth. Therefore, the rotor may align with the stator in 96 distinct positions.
This half tooth offset shows in the rotor pole detail above, or Figure 13.30.
As if this were not complicated enough, the stator main poles are divided into two phases
(φ1, φ2). These stator phases are offset from one another by onequarter of a tooth. This
detail is only discernable on the schematic diagrams below. The result is that the rotor moves
in steps of a quarter of a tooth when the phases are alternately energized. In other words, the
rotor moves in 2×96=192 steps per revolution for the above stepper.
The above drawing is representative of an actual hybrid stepper motor. However, we pro
vide a simplified pictorial and schematic representation (Figure 13.40) to illustrate details not
obvious above. Note the reduced number of coils and teeth in rotor and stator for simplicity. In
the next two figures, we attempt to illustrate the quarter tooth rotation produced by the two
stator phases offset by a quarter tooth, and the rotor half tooth offset. The quarter tooth stator
offset in conjunction with drive current timing also defines direction of rotation.
Features of hybrid stepper schematic (Figure 13.40)
13.5. STEPPER MOTORS 437
PM South
PM
North
Stator South
stator North
alignment
alignment N φ1
S φ1′
φ2′ φ2
1/2 tooth
offset
1/4 tooth offset
Figure 13.40: Hybrid stepper motor schematic diagram.
• The top of the permanent magnet rotor is the south pole, the bottom north.
• The rotor northsouth teeth are offset by half a tooth.
• If the φ1 stator is temporarily energized north top, south bottom.
• The top φ1 stator teeth align north to rotor top south teeth.
• The bottom φ1’ stator teeth align south to rotor bottom north teeth.
• Enough torque applied to the shaft to overcome the holdin torque would move the rotor
by one tooth.
• If the polarity of φ1 were reversed, the rotor would move by onehalf tooth, direction
unknown. The alignment would be south stator top to north rotor bottom, north stator
bottom to south rotor.
• The φ2 stator teeth are not aligned with the rotor teeth when φ1 is energized. In fact,
the φ2 stator teeth are offset by onequarter tooth. This will allow for rotation by that
amount if φ1 is deenergized and φ2 energized. Polarity of φ1 and ¡phi2¿ drive deter
mines direction of rotation.
Hybrid stepper motor rotation (Figure 13.41)
• Rotor top is permanent magnet south, bottom north. Fields φ1, φ2 are switchable: on,
off, reverse.
• (a) φ1=on=northtop, φ2=off. Align (top to bottom): φ1 statorN:rotortopS, φ1’
statorS: rotorbottomN. Start position, rotation=0.
• (b) φ1=off, φ2=on. Align (right to left): φ2 statorNright:rotortopS, φ2’ statorS:
rotorbottomN. Rotate 1/4 tooth, total rotation=1/4 tooth.
438 CHAPTER 13. AC MOTORS
(a) (b) (c)
N
N
NS
S
S
align top align right align bottom
φ2
φ1
φ2′
φ1′
φ1
φ1′
φ1
φ1′
φ2′ φ2φ2′ φ2
off
off
off
off off
off
Figure 13.41: Hybrid stepper motor rotation sequence.
• (c) φ1=reverse(on), φ2=off. Align (bottom to top): φ1 statorS:rotorbottomN, φ1’
statorN:rotortopS. Rotate 1/4 tooth from last position. Total rotation from start: 1/2
tooth.
• Not shown: φ1=off, φ2=reverse(on). Align (left to right): Total rotation: 3/4 tooth.
• Not shown: φ1=on, φ2=off (same as (a)). Align (top to bottom): Total rotation 1tooth.
An unpowered stepper motor with detent torque is either a permanent magnet stepper or
a hybrid stepper. The hybrid stepper will have a small step angle, much less than the 7.5o of
permanent magnet steppers. The step angle could be a fraction of a degree, corresponding to a
few hundred steps per revolution.
Summary: hybrid stepper motor
• The step angle is smaller than variable reluctance or permanent magnet steppers.
• The rotor is a permanent magnet with fine teeth. North and south teeth are offset by half
a tooth for a smaller step angle.
• The stator poles have matching fine teeth of the same pitch as the rotor.
• The stator windings are divided into no less than two phases.
• The poles of one stator windings are offset by a quarter tooth for an even smaller step
angle.
13.6 Brushless DC motor
Brushless DC motors were developed from conventional brushed DC motors with the avail
ability of solid state power semiconductors. So, why do we discuss brushless DC motors in a
chapter on AC motors? Brushless DC motors are similar to AC synchronous motors. The major
13.6. BRUSHLESS DC MOTOR 439
difference is that synchronous motors develop a sinusoidal back EMF, as compared to a rect
angular, or trapezoidal, back EMF for brushless DC motors. Both have stator created rotating
magnetic fields producing torque in a magnetic rotor.
Synchronous motors are usually large multikilowatt size, often with electromagnet rotors.
True synchronous motors are considered to be single speed, a submultiple of the powerline
frequency. Brushless DC motors tend to be small– a few watts to tens of watts, with permanent
magnet rotors. The speed of a brushless DC motor is not fixed unless driven by a phased locked
loop slaved to a reference frequency. The style of construction is either cylindrical or pancake.
(Figures 13.42 and 13.43)
(a)
Stator
Stator
Rotor
Rotor
(b)
Figure 13.42: Cylindrical construction: (a) outside rotor, (b) inside rotor.
The most usual construction, cylindrical, can take on two forms (Figure 13.42). The most
common cylindrical style is with the rotor on the inside, above right. This style motor is used
in hard disk drives. It is also possible to put the rotor on the outside surrounding the stator.
Such is the case with brushless DC fan motors, sans the shaft. This style of construction may
be short and fat. However, the direction of the magnetic flux is radial with respect to the
rotational axis.
Rotor
RotorStator
Stator
Stator
(a) (b)
Figure 13.43: Pancake motor construction: (a) single stator, (b) double stator.
High torque pancake motors may have stator coils on both sides of the rotor (Figure 13.43
b). Lower torque applications like floppy disk drive motors suffice with a stator coil on one side
440 CHAPTER 13. AC MOTORS
of the rotor, (Figure 13.43a). The direction of the magnetic flux is axial, that is, parallel to the
axis of rotation.
The commutation function may be performed by various shaft position sensors: optical
encoder, magnetic encoder (resolver, synchro, etc), or Hall effect magnetic sensors. Small inex
pensive motors use Hall effect sensors. (Figure 13.44) A Hall effect sensor is a semiconductor
device where the electron flow is affected by a magnetic field perpendicular to the direction of
current flow.. It looks like a four terminal variable resistor network. The voltages at the two
outputs are complementary. Application of a magnetic field to the sensor causes a small voltage
change at the output. The Hall output may drive a comparator to provide for more stable drive
to the power device. Or, it may drive a compound transistor stage if properly biased. More
modern Hall effect sensors may contain an integrated amplifier, and digital circuitry. This 3
lead device may directly drive the power transistor feeding a phase winding. The sensor must
be mounted close to the permanent magnet rotor to sense its position.
N
S
H2H1
H3
V+
V+
V+
V+
V+
NC
H3 H2
V+
H1
V+
A3
A2
A1
Figure 13.44: Hall effect sensors commutate 3φ brushless DC motor.
The simple cylindrical 3φmotor Figure 13.44 is commutated by a Hall effect device for each
of the three stator phases. The changing position of the permanent magnet rotor is sensed by
the Hall device as the polarity of the passing rotor pole changes. This Hall signal is amplified
so that the stator coils are driven with the proper current. Not shown here, the Hall signals
may be processed by combinatorial logic for more efficient drive waveforms.
The above cylindrical motor could drive a harddrive if it were equipped with a phased locked
loop (PLL) to maintain constant speed. Similar circuitry could drive the pancake floppy disk
drive motor (Figure 13.45). Again, it would need a PLL to maintain constant speed.
The 3φ pancake motor (Figure 13.45) has 6stator poles and 8rotor poles. The rotor is a flat
ferrite ring magnetized with eight axially magnetized alternating poles. We do not show that
the rotor is capped by a mild steel plate for mounting to the bearing in the middle of the stator.
The steel plate also helps complete the magnetic circuit. The stator poles are also mounted
atop a steel plate, helping to close the magnetic circuit. The flat stator coils are trapezoidal to
more closely fit the coils, and approximate the rotor poles. The 6stator coils comprise three
winding phases.
If the three stator phases were successively energized, a rotating magnetic field would be
generated. The permanent magnet rotor would follow as in the case of a synchronous motor. A
13.6. BRUSHLESS DC MOTOR 441
NS
N
N
N S
S
S
RotorStator
Hall effect sensor
φ−1 φ−2φ−3
φ−1′
Figure 13.45: Brushless pancake motor
two pole rotor would follow this field at the same rotation rate as the rotating field. However,
our 8pole rotor will rotate at a submultiple of this rate due the the extra poles in the rotor.
The brushless DC fan motor (Figure 13.46) has these feature:
N
N
S
S
N
N
N
N
S
S
S
S
H1
H2
2φ brushless fan motor
Figure 13.46: Brushless fan motor, 2φ.
• The stator has 2phases distributed between 4poles
• There are 4salient poles with no windings to eliminate zero torque points.
• The rotor has four main drive poles.
• The rotor has 8poles superimposed to help eliminate zero torque points.
• The Hall effect sensors are spaced at 45o physical.
• The fan housing is placed atop the rotor, which is placed over the stator.
The goal of a brushless fan motor is to minimize the cost of manufacture. This is an incen
tive to move lower performance products from a 3φ to a 2φ configuration. Depending on how
it is driven, it may be called a 4φ motor.
442 CHAPTER 13. AC MOTORS
You may recall that conventional DC motors cannot have an even number of armature
poles (2,4, etc) if they are to be selfstarting, 3,5,7 being common. Thus, it is possible for a
hypothetical 4pole motor to come to rest at a torque minima, where it cannot be started from
rest. The addition of the four small salient poles with no windings superimposes a ripple torque
upon the torque vs position curve. When this ripple torque is added to normal energizedtorque
curve, the result is that torque minima are partially removed. This makes it possible to start
the motor for all possible stopping positions. The addition of eight permanant magnet poles
to the normal 4pole permanent magnet rotor superimposes a small second harmonic ripple
torque upon the normal 4pole ripple torque. This further removes the torque minima. As long
as the torque minima does not drop to zero, we should be able to start the motor. The more
successful we are in removing the torque minima, the easier the motor starting.
The 2φ stator requires that the Hall sensors be spaced apart by 90o electrical. If the rotor
was a 2pole rotor, the Hall sensors would be placed 90o physical. Since we have a 4pole
permanent magnet rotor, the sensors must be placed 45o physical to achieve the 90o electrical
spacing. Note Hall spacing above. The majority of the torque is due to the interaction of the
inside stator 2φ coils with the 4pole section of the rotor. Moreover, the 4pole section of the
rotor must be on the bottom so that the Hall sensors will sense the proper commutation signals.
The 8poles rotor section is only for improving motor starting.
N
S H1
H2
V+
H2
V+
H1
V+
A3 A2
A1
A2
V+
V+
V+
Figure 13.47: Brushless DC motor 2φ pushpull drive.
In Figure 13.47, the 2φ pushpull drive (also known as 4φ drive) uses two Hall effect sen
sors to drive four windings. The sensors are spaced 90o electrical apart, which is 90o physical
for a single pole rotor. Since the Hall sensor has two complementary outputs, one sensor pro
vides commutation for two opposing windings.
13.7 Tesla polyphase induction motors
Most AC motors are induction motors. Induction motors are favored due to their ruggedness
and simplicity. In fact, 90% of industrial motors are induction motors.
Nikola Tesla conceived the basic principals of the polyphase induction motor in 1883, and
had a half horsepower (400 watt) model by 1888. Tesla sold the manufacturing rights to George
13.7. TESLA POLYPHASE INDUCTION MOTORS 443
Westinghouse for $65,000.
Most large ( > 1 hp or 1 kW) industrial motors are polyphase induction motors. By poly
phase, we mean that the stator contains multiple distinct windings per motor pole, driven
by corresponding time shifted sine waves. In practice, this is two or three phases. Large in
dustrial motors are 3phase. While we include numerous illustrations of twophase motors for
simplicity, we must emphasize that nearly all polyphase motors are threephase. By induction
motor, we mean that the stator windings induce a current flow in the rotor conductors, like a
transformer, unlike a brushed DC commutator motor.
13.7.1 Construction
An induction motor is composed of a rotor, known as an armature, and a stator containing
windings connected to a polyphase energy source as shown in Figure 13.48. The simple 2
phase induction motor below is similar to the 1/2 horsepower motor which Nikola Tesla intro
duced in 1888.
Stator
Rotor
Figure 13.48: Tesla polyphase induction motor.
The stator in Figure 13.48 is wound with pairs of coils corresponding to the phases of elec
trical energy available. The 2phase induction motor stator above has 2pairs of coils, one pair
for each of the two phases of AC. The individual coils of a pair are connected in series and
correspond to the opposite poles of an electromagnet. That is, one coil corresponds to a Npole,
the other to a Spole until the phase of AC changes polarity. The other pair of coils is oriented
90o in space to the first pair. This pair of coils is connected to AC shifted in time by 90o in
the case of a 2phase motor. In Tesla’s time, the source of the two phases of AC was a 2phase
alternator.
The stator in Figure 13.48 has salient, obvious protruding poles, as used on Tesla’s early
induction motor. This design is used to this day for subfractional horsepower motors (<50
watts). However, for larger motors less torque pulsation and higher efficiency results if the
444 CHAPTER 13. AC MOTORS
coils are embedded into slots cut into the stator laminations. (Figure 13.49)
Figure 13.49: Stator frame showing slots for windings.
The stator laminations are thin insulated rings with slots punched from sheets of electrical
grade steel. A stack of these is secured by end screws, which may also hold the end housings.
(a)
φ−3
φ−1
φ−1φ−2 φ−2
(b)
Figure 13.50: Stator with (a) 2φ and (b) 3φ windings.
In Figure 13.50, the windings for both a twophase motor and a threephase motor have
been installed in the stator slots. The coils are wound on an external fixture, then worked into
the slots. Insulation wedged between the coil periphery and the slot protects against abrasion.
Actual stator windings are more complex than the single windings per pole in Figure 13.50.
Comparing the 2φ motor to Tesla’s 2φ motor with salient poles, the number of coils is the
13.7. TESLA POLYPHASE INDUCTION MOTORS 445
same. In actual large motors, a pole winding, is divided into identical coils inserted into many
smaller slots than above. This group is called a phase belt. See Figure 13.16. The distributed
coils of the phase belt cancel some of the odd harmonics, producing a more sinusoidal magnetic
field distribution across the pole. This is shown in the synchronous motor section. The slots
at the edge of the pole may have fewer turns than the other slots. Edge slots may contain
windings from two phases. That is, the phase belts overlap.
The key to the popularity of the AC induction motor is simplicity as evidenced by the simple
rotor (Figure 13.51). The rotor consists of a shaft, a steel laminated rotor, and an embedded
copper or aluminum squirrel cage, shown at (b) removed from the rotor. As compared to a
DC motor armature, there is no commutator. This eliminates the brushes, arcing, sparking,
graphite dust, brush adjustment and replacement, and remachining of the commutator.
(a) (b)
Figure 13.51: Laminated rotor with (a) embedded squirrel cage, (b) conductive cage removed
from rotor.
The squirrel cage conductors may be skewed, twisted, with respsect to the shaft. The mis
alignment with the stator slots reduces torque pulsations.
Both rotor and stator cores are composed of a stack of insulated laminations. The lamina
tions are coated with insulating oxide or varnish to minimize eddy current losses. The alloy
used in the laminations is selected for low hysteresis losses.
13.7.2 Theory of operation
A short explanation of operation is that the stator creates a rotating magnetic field which drags
the rotor around.
The theory of operation of induction motors is based on a rotating magnetic field. One
means of creating a rotating magnetic field is to rotate a permanent magnet as shown in Fig
ure 13.52. If the moving magnetic lines of flux cut a conductive disk, it will follow the motion
of the magnet. The lines of flux cutting the conductor will induce a voltage, and consequent
current flow, in the conductive disk. This current flow creates an electromagnet whose polarity
opposes the motion of the permanent magnet– Lenz’s Law. The polarity of the electromagnet
is such that it pulls against the permanent magnet. The disk follows with a little less speed
than the permanent magnet.
446 CHAPTER 13. AC MOTORS
Figure 13.52: Rotating magnetic field produces torque in conductive disk.
The torque developed by the disk is proportional to the number of flux lines cutting the disk
and the rate at which it cuts the disk. If the disk were to spin at the same rate as the permanent
magnet, there would be no flux cutting the disk, no induced current flow, no electromagnet field,
no torque. Thus, the disk speed will always fall behind that of the rotating permanent magnet,
so that lines of flux cut the disk induce a current, create an electromagnetic field in the disk,
which follows the permanent magnet. If a load is applied to the disk, slowing it, more torque
will be developed as more lines of flux cut the disk. Torque is proportional to slip, the degree
to which the disk falls behind the rotating magnet. More slip corresponds to more flux cutting
the conductive disk, developing more torque.
An analog automotive eddy current speedometer is based on the principle illustrated above.
With the disk restrained by a spring, disk and needle deflection is proportional to magnet
rotation rate.
A rotating magnetic field is created by two coils placed at right angles to each other, driven
by currents which are 90o out of phase. This should not be surprising if you are familiar with
oscilloscope Lissajous patterns.
In Figure 13.53, a circular Lissajous is produced by driving the horizontal and vertical
oscilloscope inputs with 90o out of phase sine waves. Starting at (a) with maximum “X” and
minimum “Y” deflection, the trace moves up and left toward (b). Between (a) and (b) the
two waveforms are equal to 0.707 Vpk at 45o. This point (0.707, 0.707) falls on the radius
of the circle between (a) and (b) The trace moves to (b) with minimum “X” and maximum
“Y” deflection. With maximum negative “X” and minimum “Y” deflection, the trace moves to
(c). Then with minimum “X” and maximum negative “Y”, it moves to (d), and on back to (a),
completing one cycle.
Figure 13.54 shows the two 90o phase shifted sine waves applied to oscilloscope deflection
plates which are at right angles in space. If this were not the case, a one dimensional line
would display. The combination of 90o phased sine waves and right angle deflection, results
in a two dimensional pattern– a circle. This circle is traced out by a counterclockwise rotating
13.7. TESLA POLYPHASE INDUCTION MOTORS 447
trigger
timebase
s/div
DC GND AC
X
GNDDC
V/div
vertical
OSCILLOSCOPE
Y
AC
horiz’tl
a a’ b c d a
a
b
c
d
Figure 13.53: Out of phase (90o) sine waves produce circular Lissajous pattern.
a
b
c
d
a’ a’
a’
a
b
c
d
b
c
d
a
a
a
horizontal
deflection
verticall
deflection
Figure 13.54: Xaxis sine and Yaxis cosine trace circle.
448 CHAPTER 13. AC MOTORS
electron beam.
For reference, Figure 13.55shows why inphase sine waves will not produce a circular pat
tern. Equal “X” and “Y” deflection moves the illuminated spot from the origin at (a) up to right
(1,1) at (b), back down left to origin at (c),down left to (1.1) at (d), and back up right to origin.
The line is produced by equal deflections along both axes; y=x is a straight line.
trigger
timebase
s/div
DC GND AC
X
GNDDC
V/div
vertical
OSCILLOSCOPE
Y
AC
horiz’tl
a
b
c
d
a a’ b c d a
Figure 13.55: No circular motion from inphase waveforms.
If a pair of 90o out of phase sine waves produces a circular Lissajous, a similar pair of
currents should be able to produce a circular rotating magnetic field. Such is the case for a 2
phase motor. By analogy three windings placed 120o apart in space, and fed with corresponding
120o phased currents will also produce a rotating magnetic field.
φ−2
a a’ b c d
φ−1
φ−2
a
c
a’ b
d
φ−1 φ−1
φ−2
φ−1
φ−2
φ−1
φ−2
φ−1
φ−2
Figure 13.56: Rotating magnetic field from 90o phased sinewaves.
13.7. TESLA POLYPHASE INDUCTION MOTORS 449
As the 90o phased sinewaves, Figure 13.56, progress from points (a) through (d), the mag
netic field rotates counterclockwise (figures ad) as follows:
• (a) φ1 maximum, φ2 zero
• (a’) φ1 70%, φ2 70%
• (b) φ1 zero, φ2 maximum
• (c) φ1 maximum negative, φ2 zero
• (d) φ1 zero, φ2 maximum negative
Motor speed
The rotation rate of a stator rotating magnetic field is related to the number of pole pairs per
stator phase. The “full speed” Figure 13.57 has a total of six poles or three polepairs and three
phases. However,there is but one pole pair per phase– the number we need. The magnetic field
will rotate once per sine wave cycle. In the case of 60 Hz power, the field rotates at 60 times per
second or 3600 revolutions per minute (rpm). For 50 Hz power, it rotates at 50 rotations per
second, or 3000 rpm. The 3600 and 3000 rpm, are the synchronous speed of the motor. Though
the rotor of an induction motor never achieves this speed, it certainly is an upper limit. If we
double the number of motor poles, the synchronous speed is cut in half because the magnetic
field rotates 180o in space for 360o of electrical sine wave.
φ1
φ2
φ3
φ3
φ1
φ2
full speed half speed
Figure 13.57: Doubling the stator poles halves the synchronous speed.
The synchronous speed is given by:
Ns = 120·f/P
Ns = synchronous speed in rpm
f = frequency of applied power, Hz
P = total number of poles per phase, a multiple of 2
450 CHAPTER 13. AC MOTORS
Example:
The “half speed” Figure 13.57 has four poles per phase (3phase). The synchronous speed
for 50 Hz power is:
S = 120·50/4 = 1500 rpm
The short explanation of the induction motor is that the rotating magnetic field produced
by the stator drags the rotor around with it.
The longer more correct explanation is that the stator’s magnetic field induces an alternat
ing current into the rotor squirrel cage conductors which constitutes a transformer secondary.
This induced rotor current in turn creates a magnetic field. The rotating stator magnetic field
interacts with this rotor field. The rotor field attempts to align with the rotating stator field.
The result is rotation of the squirrel cage rotor. If there were no mechanical motor torque load,
no bearing, windage, or other losses, the rotor would rotate at the synchronous speed. How
ever, the slip between the rotor and the synchronous speed stator field develops torque. It is
the magnetic flux cutting the rotor conductors as it slips which develops torque. Thus, a loaded
motor will slip in proportion to the mechanical load. If the rotor were to run at synchronous
speed, there would be no stator flux cutting the rotor, no current induced in the rotor, no torque.
Torque
When power is first applied to the motor, the rotor is at rest, while the stator magnetic field
rotates at the synchronous speed Ns. The stator field is cutting the rotor at the synchronous
speed Ns. The current induced in the rotor shorted turns is maximum, as is the frequency of
the current, the line frequency. As the rotor speeds up, the rate at which stator flux cuts the
rotor is the difference between synchronous speed Ns and actual rotor speed N, or (Ns  N). The
ratio of actual flux cutting the rotor to synchronous speed is defined as slip:
s = (Ns  N)/Ns
where: Ns = synchronous speed, N = rotor speed
The frequency of the current induced into the rotor conductors is only as high as the line fre
quency at motor start, decreasing as the rotor approaches synchronous speed. Rotor frequency
is given by:
fr = s·f
where: s = slip, f = stator power line frequency
Slip at 100% torque is typically 5% or less in induction motors. Thus for f = 50 Hz line fre
quency, the frequency of the induced current in the rotor fr = 0.05·50 = 2.5 Hz. Why is it so low?
The stator magnetic field rotates at 50 Hz. The rotor speed is 5% less. The rotating magnetic
field is only cutting the rotor at 2.5 Hz. The 2.5 Hz is the difference between the synchronous
speed and the actual rotor speed. If the rotor spins a little faster, at the synchronous speed, no
flux will cut the rotor at all, fr = 0.
The Figure 13.58 graph shows that starting torque known as locked rotor torque (LRT) is
higher than 100% of the full load torque (FLT), the safe continuous torque rating. The locked
13.7. TESLA POLYPHASE INDUCTION MOTORS 451
100 80 60 40 20 0 % slip
0 20 40 60 80 100 % Ns
100%
200%
300%
400%
%
fu
ll l
oa
d
to
rq
ue
&
c
ur
re
nt
current
torq
ue
locked rotor current
locked rotor torque
breakdown torque
full load torque/ current
500%
pullup torque
Figure 13.58: Torque and speed vs %Slip. %Ns=%Synchronous Speed.
rotor torque is about 175% of FLT for the example motor graphed above. Starting current
known as locked rotor current (LRC) is 500% of full load current (FLC), the safe running cur
rent. The current is high because this is analogous to a shorted secondary on a transformer.
As the rotor starts to rotate the torque may decrease a bit for certain classes of motors to a
value known as the pull up torque. This is the lowest value of torque ever encountered by the
starting motor. As the rotor gains 80% of synchronous speed, torque increases from 175% up to
300% of the full load torque. This breakdown torque is due to the larger than normal 20% slip.
The current has decreased only slightly at this point, but will decrease rapidly beyond this
point. As the rotor accelerates to within a few percent of synchronous speed, both torque and
current will decrease substantially. Slip will be only a few percent during normal operation.
For a running motor, any portion of the torque curve below 100% rated torque is normal. The
motor load determines the operating point on the torque curve. While the motor torque and
current may exceed 100% for a few seconds during starting, continuous operation above 100%
can damage the motor. Any motor torque load above the breakdown torque will stall the motor.
The torque, slip, and current will approach zero for a “no mechanical torque” load condition.
This condition is analogous to an open secondary transformer.
There are several basic induction motor designs (Figure 13.59) showing consideable varia
tion from the torque curve above. The different designs are optimized for starting and running
different types of loads. The locked rotor torque (LRT) for various motor designs and sizes
ranges from 60% to 350% of full load torque (FLT). Starting current or locked rotor current
(LRC) can range from 500% to 1400% of full load current (FLC). This current draw can present
a starting problem for large induction motors.
NEMA design classes
Various standard classes (or designs) for motors, corresponding to the torque curves (Fig
ure 13.59) have been developed to better drive various type loads. The National Electrical
Manufacturers Association (NEMA) has specified motor classes A, B, C, and D to meet these
452 CHAPTER 13. AC MOTORS
drive requirements. Similar International Electrotechnical Commission (IEC) classes N and H
correspond to NEMA B and C designs respectively.
100 80 60 40 20 0 % slip
0 20 40 60 80 100 % Ns
100%
200%
300%
400%
%
fu
ll l
oa
d
to
rq
ue
A
D
C
B
Figure 13.59: Characteristics for NEMA designs.
All motors, except class D, operate at %5 slip or less at full load.
• Class B (IEC Class N) motors are the default motor to use in most applications. With a
starting torque of LRT = 150% to 170% of FLT, it can start most loads, without excessive
starting current (LRT). Efficiency and power factor are high. It typically drives pumps,
fans, and machine tools.
• Class A starting torque is the same as class B. Drop out torque and starting current
(LRT)are higher. This motor handles transient overloads as encountered in injection
molding machines.
• Class C (IEC Class H) has higher starting torque than class A and B at LRT = 200%
of FLT. This motor is applied to hardstarting loads which need to be driven at constant
speed like conveyors, crushers, and reciprocating pumps and compressors.
• Class D motors have the highest starting torque (LRT) coupled with low starting cur
rent due to high slip ( 5% to 13% at FLT). The high slip results in lower speed. Speed
regulation is poor. However, the motor excels at driving highly variable speed loads like
those requiring an energy storage flywheel. Applications include punch presses, shears,
and elevators.
• Class E motors are a higher efficiency version of class B.
• Class F motors have much lower LRC, LRT, and break down torque than class B. They
drive constant easily started loads.
Power factor
Induction motors present a lagging (inductive) power factor to the power line.The power factor
in large fully loaded high speed motors can be as favorable as 90% for large high speed motors.
13.7. TESLA POLYPHASE INDUCTION MOTORS 453
At 3/4 full load the largest high speed motor power factor can be 92%. The power factor for
small low speed motors can be as low as 50%. At starting, the power factor can be in the range
of 10% to 25%, rising as the rotor achieves speed.
Power factor (PF) varies considerably with the motor mechanical load (Figure 13.60). An
unloaded motor is analogous to a transformer with no resistive load on the secondary. Little
resistance is reflected from the secondary (rotor) to the primary (stator). Thus the power line
sees a reactive load, as low as 10% PF. As the rotor is loaded an increasing resistive component
is reflected from rotor to stator, increasing the power factor.
0 20 40 60 80 100 % load
20%
40%
e
ffi
cie
nc
e,
po
we
r f
ac
to
r
60%
80%
100%
power factor
efficiency
Figure 13.60: Induction motor power factor and efficiency.
Efficiency
Large three phase motors are more efficient than smaller 3phase motors, and most all single
phase motors. Large induction motor efficiency can be as high as 95% at full load, though 90%
is more common. Efficiency for a lightly load or noloaded induction motor is poor because most
of the current is involved with maintaining magnetizing flux. As the torque load is increased,
more current is consumed in generating torque, while current associated with magnetizing re
mains fixed. Efficiency at 75% FLT can be slightly higher than that at 100% FLT. Efficiency
is decreased a few percent at 50% FLT, and decreased a few more percent at 25% FLT. Effi
ciency only becomes poor below 25% FLT. The variation of efficiency with loading is shown in
Figure 13.60
Induction motors are typically oversized to guarantee that their mechanical load can be
started and driven under all operating conditions. If a polyphase motor is loaded at less than
75% of rated torque where efficiency peaks, efficiency suffers only slightly down to 25% FLT.
13.7.3 Nola power factor corrector
Frank Nola of NASA proposed a power factor corrector (PFC) as an energy saving device for
single phase induction motors in the late 1970’s. It is based on the premise that a less than
fully loaded induction motor is less efficient and has a lower power factor than a fully loaded
motor. Thus, there is energy to be saved in partially loaded motors, 1φ motors in particular.
The energy consumed in maintaining the stator magnetic field is relatively fixed with respect
to load changes. While there is nothing to be saved in a fully loaded motor, the voltage to a par
tially loaded motor may be reduced to decrease the energy required to maintain the magnetic
field. This will increase power factor and efficiency. This was a good concept for the notoriously
inefficient single phase motors for which it was intended.
454 CHAPTER 13. AC MOTORS
This concept is not very applicable to large 3phase motors. Because of their high efficiency
(90%+), there is not much energy to be saved. Moreover, a 95% efficient motor is still 94%
efficient at 50% full load torque (FLT) and 90% efficient at 25% FLT. The potential energy
savings in going from 100% FLT to 25% FLT is the difference in efficiency 95%  90% = 5%. This
is not 5% of the full load wattage but 5% of the wattage at the reduced load. The Nola power
factor corrector might be applicable to a 3phase motor which idles most of the time (below 25%
FLT), like a punch press. The payback period for the expensive electronic controller has been
estimated to be unattractive for most applications. Though, it might be economical as part of
an electronic motor starter or speed Control. [7]
13.7.4 Induction motor alternator
An induction motor may function as an alternator if it is driven by a torque at greater than
100% of the synchronous speed. (Figure 13.61) This corresponds to a few % of “negative” slip,
say 1% slip. This means that as we are rotating the motor faster than the synchronous speed,
the rotor is advancing 1% faster than the stator rotating magnetic field. It normally lags by
1% in a motor. Since the rotor is cutting the stator magnetic field in the opposite direction
(leading), the rotor induces a voltage into the stator feeding electrical energy back into the
power line.
100 80 60 40 20 0 % slip
0 20 40 60 80 100 % Ns
100%
200%
300%
400%
%
fu
ll l
oa
d
to
rq
ue
&
c
ur
re
nt
current
torq
ue
500%
200%
400%
current
torq
ue
500%
300%
100%
20 40 60 80 100 % slip
120 140 160 180 200 % Ns
%
fu
ll l
oa
d
to
rq
ue
&
c
ur
re
nt
Motor mode
Generator mode
Figure 13.61: Negative torque makes induction motor into generator.
Such an induction generator must be excited by a “live” source of 50 or 60 Hz power. No
power can be generated in the event of a power company power failure. This type of alternator
appears to be unsuited as a standby power source. As an auxiliary power wind turbine gen
erator, it has the advantage of not requiring an automatic power failure disconnect switch to
protect repair crews. It is failsafe.
13.7. TESLA POLYPHASE INDUCTION MOTORS 455
Small remote (from the power grid) installations may be make selfexciting by placing ca
pacitors in parallel with the stator phases. If the load is removed residual magnetism may
generate a small amount of current flow. This current is allowed to flow by the capacitors with
out dissipating power. As the generator is brought up to full speed, the current flow increases
to supply a magnetizing current to the stator. The load may be applied at this point. Voltage
regulation is poor. An induction motor may be converted to a selfexcited generator by the
addition of capacitors.[6]
Start up procedure is to bring the wind turbine up to speed in motor mode by application
of normal power line voltage to the stator. Any wind induced turbine speed in excess of syn
chronous speed will develop negative torque, feeding power back into the power line, reversing
the normal direction of the electric kilowatthour meter. Whereas an induction motor presents
a lagging power factor to the power line, an induction alternator presents a leading power
factor. Induction generators are not widely used in conventional power plants. The speed of
the steam turbine drive is steady and controllable as required by synchronous alternators.
Synchronous alternators are also more efficient.
The speed of a wind turbine is difficult to control, and subject to wind speed variation by
gusts. An induction alternator is better able to cope with these variations due to the inher
ent slip. This stresses the gear train and mechanical components less than a synchronous
genertor. However, this allowable speed variation only amounts to about 1%. Thus, a di
rect line connected induction generator is considered to be fixedspeed in a wind turbine. See
Doublyfed induction generator for a true variable speed alternator. Multiple generators or
multiple windings on a common shaft may be switched to provide a high and low speed to
accomodate variable wind conditions.
13.7.5 Motor starting and speed control
Some induction motors can draw over 1000% of full load current during starting; though, a few
hundred percent is more common. Small motors of a few kilowatts or smaller can be started by
direct connection to the power line. Starting larger motors can cause line voltage sag, affecting
other loads. Motorstart rated circuit breakers (analogous to slow blow fuses) should replace
standard circuit breakers for starting motors of a few kilowatts. This breaker accepts high
overcurrent for the duration of starting.
Motors over 50 kW use motor starters to reduce line current from several hundred to a few
hundred percent of full load current. An intermittent duty autotarnsformer may reduce the
stator voltage for a fraction of a minute during the start interval, followed by application of
full line voltage as in Figure 13.62. Closure of the S contacts applies reduced voltage during
the start interval. The S contacts open and the R contacts close after starting. This reduces
starting current to, say, 200% of full load current. Since the autotransformer is only used for
the short start interval, it may be sized considerably smaller than a continuous duty unit.
Running 3phase motors on 1phase
Threephase motors will run on single phase as readily as single phase motors. The only
problem for either motor is starting. Sometimes 3phase motors are purchased for use on
single phase if threephase provisioning is anticipated. The power rating needs to be 50%
larger than for a comparable single phase motor to make up for one unused winding. Single
456 CHAPTER 13. AC MOTORS
φ1
φ2
φ3
S
R
S
S
R
S
S = start, R= run
M1
T1
T2
Figure 13.62: Autotransformer induction motor starter.
phase is applied to a pair of windings simultanous with a start capacitor in series with the
third winding. The start switch is opened in Figure 13.63 upon motor start. Sometimes a
smaller capacitor than the start capacitor is retained while running.
φ1 R
S = start, R= run
M1
Rφ1’
S
optional run capacitor
start capacitor
return
φsynthetic
φsynthetic
φ1
φ1’
φ1
φ2
φ3
synthetic 3φ standard 3φ
Figure 13.63: Starting a threephase motor on single phase.
The circuit in Figure 13.63 for running a threephase motor on single phase is known as a
static phase converter if the motor shaft is not loaded. Moreover, the motor acts as a 3phase
generator. Three phase power may be tapped off from the three stator windings for powering
other 3phase equipment. The capacitor supplies a synthetic phase approximately midway
6 90o between the 6 180o single phase power source terminals for starting. While running, the
motor generates approximately standard 3φ, as shown in Figure 13.63. Matt Isserstedt shows
a complete design for powering a home machine shop. [8]
Since a static phase converter has no torque load, it may be started with a capacitor con
siderably smaller than a normal start capacitor. If it is small enough, it may be left in circuit
as a runcapacitor. See Figure 13.64. However, smaller runcapacitors result in better 3phase
power output as in Figure 13.65. Moreover, adjustment of these capacitors to equalize the
currents as measured in the three phases results in the most efficient machine.[9] However, a
large start capacitor is required for about a second to quickly start the converter. Hanrahan
provides construction details.[9]
13.7. TESLA POLYPHASE INDUCTION MOTORS 457
1 HP
motor
L1
L2
L3
L1
L2
220V
single phase in
3phase
out
Run capacitor = 2530 µF per HP
Figure 13.64: Selfstarting static phase converter. Run capacitor = 2530µF per HP. Adapted
from Figure 7, Hanrahan [9]
1 HP
motor
L1
L2
L3
L1
L2
220V
single phase
in
3phase
out
Start capacitor = 50100 µF/HP. Run capacitors = 1216 µF/HP.
startrun
run
Figure 13.65: More efficient static phase converter. Start capacitor = 50100µF/HP. Run capac
itors = 1216µF/HP. Adapted from Figure 1, Hanrahan [9]
458 CHAPTER 13. AC MOTORS
Multiple fields
Induction motors may contain multiple field windings, for example a 4pole and an 8pole
winding corresponding to 1800 and 900 rpm synchronous speeds. Energizing one field or the
other is less complex than rewiring the stator coils in Figure 13.66.
4pole 2pole
φ1 φ1
Figure 13.66: Multiple fields allow speed change.
If the field is segmented with leads brought out, it may be rewired (or switched) from 4
pole to 2pole as shown above for a 2phase motor. The 22.5o segments are switchable to 45o
segments. Only the wiring for one phase is shown above for clarity. Thus, our induction motor
may run at multiple speeds. When switching the above 60 Hz motor from 4 poles to 2 poles
the synchronous speed increases from 1800 rpm to 3600 rpm. If the motor is driven by 50 Hz,
what would be the corresponding 4pole and 2pole synchronous speeds?
Ns = 120f/P = 120*50/4 = 1500 rpm (4pole)
Ns = 3000 rpm (2pole)
Variable voltage
The speed of small squirrel cage induction motors for applications such as driving fans, may
be changed by reducing the line voltage. This reduces the torque available to the load which
reduces the speed. (Figure 13.67)
Electronic speed control
Modern solid state electronics increase the options for speed control. By changing the 50 or
60 Hz line frequency to higher or lower values, the synchronous speed of the motor may be
changed. However, decreasing the frequency of the current fed to the motor also decreases
reactance XL which increases the stator current. This may cause the stator magnetic circuit
to saturate with disastrous results. In practice, the voltage to the motor needs to be decreased
when frequency is decreased.
Conversely, the drive frequency may be increased to increase the synchronous speed of the
motor. However, the voltage needs to be increased to overcome increasing reactance to keep
13.7. TESLA POLYPHASE INDUCTION MOTORS 459
100 80 60 40 20 0 % slip
0 20 40 60 80 100 % Ns
100%
75%
50%
%
to
rq
ue
100% V
25%
50% V
load torqu
e
reduced
speedmotor torque
Figure 13.67: Variable voltage controls induction motor speed.
M1
AC line Inverter,variable
frequency
& voltagespeed
Figure 13.68: Electronic variable speed drive.
current up to a normal value and maintain torque. The inverter (Figure¡ref¿02480.eps¡x1¿)
approximates sinewaves to the motor with pulse width modulation outputs. This is a chopped
waveform which is either on or off, high or low, the percentage of “on” time corresponds to the
instantaneous sine wave voltage.
Once electronics is applied to induction motor control, many control methods are available,
varying from the simple to complex:
Summary: Speed control
• Scaler Control Low cost method described above to control only voltage and frequency,
without feedback.
• Vector Control Also known as vector phase control. The flux and torque producing com
ponents of stator current are measured or estimated on a realtime basis to enhance the
motor torquespeed curve. This is computation intensive.
• Direct Torque Control An elaborate adaptive motor model allows more direct control of
flux and torque without feedback. This method quickly responds to load changes.
Summary: Tesla polyphase induction motors
• A polyphase induction motor consists of a polyphase winding embedded in a laminated
stator and a conductive squirrel cage embedded in a laminated rotor.
460 CHAPTER 13. AC MOTORS
• Three phase currents flowing within the stator create a rotating magnetic field which
induces a current, and consequent magnetic field in the rotor. Rotor torque is developed
as the rotor slips a little behind the rotating stator field.
• Unlike single phase motors, polyphase induction motors are selfstarting.
• Motor starters minimize loading of the power line while providing a larger starting torque
than required during running. Line current reducing starters are only required for large
motors.
• Three phase motors will run on single phase, if started.
• A static phase converter is three phase motor running on single phase having no shaft
load, generating a 3phase output.
• Multiple field windings can be rewired for multiple discrete motor speeds by changing
the number of poles.
13.7.6 Linear induction motor
The wound stator and the squirrel cage rotor of an induction motor may be cut at the circum
ference and unrolled into a linear induction motor. The direction of linear travel is controlled
by the sequence of the drive to the stator phases.
The linear induction motor has been proposed as a drive for high speed passenger trains. Up
to this time, the linear induction motor with the accompanying magnetic repulsion levitation
system required for a smooth ride has been too costly for all but experimental installations.
However, the linear induction motor is scheduled to replace steam driven catapult aircraft
launch systems on the next generation of naval aircraft carrier, CVNX1, in 2013. This will
increase efficiency and reduce maintenance.[4] [5]
13.8 Wound rotor induction motors
A wound rotor induction motor has a stator like the squirrel cage induction motor, but a rotor
with insulated windings brought out via slip rings and brushes. However, no power is applied
to the slip rings. Their sole purpose is to allow resistance to be placed in series with the rotor
windings while starting. (Figure 13.69) This resistance is shorted out once the motor is started
to make the rotor look electrically like the squirrel cage counterpart.
Why put resistance in series with the rotor? Squirrel cage induction motors draw 500% to
over 1000% of full load current (FLC) during starting. While this is not a severe problem for
small motors, it is for large (10’s of kW) motors. Placing resistance in series with the rotor
windings not only decreases start current, locked rotor current (LRC), but also increases the
starting torque, locked rotor torque (LRT). Figure 13.70 shows that by increasing the rotor
resistance from R0 to R1 to R2, the breakdown torque peak is shifted left to zero speed.Note
that this torque peak is much higher than the starting torque available with no rotor resistance
(R0) Slip is proportional to rotor resistance, and pullout torque is proportional to slip. Thus,
high torque is produced while starting.
13.8. WOUND ROTOR INDUCTION MOTORS 461
M1
Stator Rotor Start resistance
φ2
φ1
φ3
Figure 13.69: Wound rotor induction motor.
100 80 60 40 20 0 % slip
0 20 40 60 80 100 % Ns
100%
200%
300%
400%
%
fu
ll l
oa
d
to
rq
ue
breakdown torque
R0
R1R2
Figure 13.70: Breakdown torque peak is shifted to zero speed by increasing rotor resistance.
462 CHAPTER 13. AC MOTORS
The resistance decreases the torque available at full running speed. But that resistance
is shorted out by the time the rotor is started. A shorted rotor operates like a squirrel cage
rotor. Heat generated during starting is mostly dissipated external to the motor in the starting
resistance. The complication and maintenance associated with brushes and slip rings is a
disadvantage of the wound rotor as compared to the simple squirrel cage rotor.
This motor is suited for starting high inertial loads. A high starting resistance makes the
high pull out torque available at zero speed. For comparison, a squirrel cage rotor only exhibits
pull out (peak) torque at 80% of its’ synchronous speed.
13.8.1 Speed control
Motor speed may be varied by putting variable resistance back into the rotor circuit. This
reduces rotor current and speed. The high starting torque available at zero speed, the down
shifted break down torque, is not available at high speed. See R2 curve at 90%Ns, Figure 13.71.
Resistors R0R1R2R3 increase in value from zero. A higher resistance at R3 reduces the speed
further. Speed regulation is poor with respect to changing torque loads. This speed control
technique is only usefull over a range of 50% to 100% of full speed. Speed control works well
with variable speed loads like elevators and printing presses.
100 80 60 40 20 0 % slip
0 20 40 60 80 100 % Ns
100%
200%
300%
400%
%
fu
ll l
oa
d
to
rq
ue
R0
R1R2
load torqu
e
R3
speed
reduction
R0, 1, 2, 3 = motor torque
Figure 13.71: Rotor resistance controls speed of wound rotor induction motor.
13.8.2 Doublyfed induction generator
We previously described a squirrel cage induction motor acting like a generator if driven faster
than the synchronous speed. (See Induction motor alternator) This is a singlyfed induction
generator, having electrical connections only to the stator windings. A wound rotor induction
motor may also act as a generator when driven above the synchronous speed. Since there are
connections to both the stator and rotor, such a machine is known as a doublyfed induction
generator (DFIG).
The singlyfed induction generator only had a usable slip range of 1% when driven by trou
blesome wind torque. Since the speed of a wound rotor induction motor may be controlled over
13.8. WOUND ROTOR INDUCTION MOTORS 463
StatorRotor φ2
φ1
φ3
Overspeed
torqure
30% Waste heat
70%
Electric energy
100%
Torque energy
Figure 13.72: Rotor resistance allows overspeed of doublyfed induction generator.
a range of 50100% by inserting resistance in the rotor, we may expect the same of the doubly
fed induction generator. Not only can we slow the rotor by 50%, we can also overspeed it by
50%. That is, we can vary the speed of a doubly fed induction generator by ±50% from the
synchronous speed. In actual practice, ±30% is more practical.
If the generator overspeeds, resistance placed in the rotor circuit will absorb excess energy
while the stator feeds constant 60 Hz to the power line. (Figure 13.72) In the case of under
speed, negative resistance inserted into the rotor circuit can make up the energy deficit, still
allowing the stator to feed the power line with 60 Hz power.
StatorRotor φ2
φ1
φ3
Overspeed
torqure
70%
Electric energyConverter φ2
φ1
φ3
30% Electric energy
100%
Torque energy
Figure 13.73: Converter recovers energy from rotor of doublyfed induction generator.
In actual practice, the rotor resistance may be replaced by a converter (Figure 13.73) ab
sorbing power from the rotor, and feeding power into the power line instead of dissipating it.
This improves the efficiency of the generator.
The converter may “borrow” power from the line for the underspeed rotor, which passes it
on to the stator. (Figure 13.74) The borrowed power, along with the larger shaft energy, passes
to the stator which is connected to the power line. The stator appears to be supplying 130% of
power to the line. Keep in mind that the rotor “borrows” 30%, leaving, leaving the line with
100% for the theoretical lossless DFIG.
Wound rotor induction motor qualities.
• Excellent starting torque for high inertia loads.
464 CHAPTER 13. AC MOTORS
StatorRotor φ2
φ1
φ3
Underspeed
torqure
130%
Electric energyConverter φ2
φ1
φ3
30% Electric energy
100%
Torque energy
Figure 13.74: Converter borrows energy from power line for rotor of doubly fed induction gen
erator, allowing it to function well under synchronous speed.
• Low starting current compared to squirrel cage induction motor.
• Speed is resistance variable over 50% to 100% full speed.
• Higher maintenance of brushes and slip rings compared to squirrel cage motor.
• The generator version of the wound rotor machine is known as a doublyfed induction
generator, a variable speed machine.
13.9 Singlephase induction motors
A three phase motor may be run from a single phase power source. (Figure 13.75) However,
it will not selfstart. It may be hand started in either direction, coming up to speed in a few
seconds. It will only develop 2/3 of the 3φ power rating because one winding is not used.
open
3φ motor, 1φ powered
nostart 1φ motor
Figure 13.75: 3φmotor runs from 1φ power, but does not start.
The single coil of a single phase induction motor does not produce a rotating magnetic field,
but a pulsating field reaching maximum intensity at 0o and 180o electrical. (Figure 13.76)
13.9. SINGLEPHASE INDUCTION MOTORS 465
∠90o
∠180o ∠0o
∠90o
a b c d e
Figure 13.76: Single phase stator produces a nonrotating, pulsating magnetic field.
Another view is that the single coil excited by a single phase current produces two counter
rotating magnetic field phasors, coinciding twice per revolution at 0o (Figure 13.76a) and 180o
(figure e). When the phasors rotate to 90o and 90o they cancel in figure b. At 45o and 45o
(figure c) they are partially additive along the +x axis and cancel along the y axis. An analogous
situation exists in figure d. The sum of these two phasors is a phasor stationary in space, but
alternating polarity in time. Thus, no starting torque is developed.
However, if the rotor is rotated forward at a bit less than the synchronous speed, It will
develop maximum torque at 10% slip with respect to the forward rotating phasor. Less torque
will be developed above or below 10% slip. The rotor will see 200%  10% slip with respect to
the counter rotating magnetic field phasor. Little torque (see torque vs slip curve) other than
a double freqency ripple is developed from the counter rotating phasor. Thus, the single phase
coil will develop torque, once the rotor is started. If the rotor is started in the reverse direction,
it will develop a similar large torque as it nears the speed of the backward rotating phasor.
Single phase induction motors have a copper or aluminum squirrel cage embedded in a
cylinder of steel laminations, typical of polyphase induction motors.
13.9.1 Permanentsplit capacitor motor
One way to solve the single phase problem is to build a 2phase motor, deriving 2phase power
from single phase. This requires a motor with two windings spaced apart 90o electrical, fed
with two phases of current displaced 90o in time. This is called a permanentsplit capacitor
motor in Figure 13.77.
Figure 13.77: Permanentsplit capacitor induction motor.
This type of motor suffers increased current magnitude and backward time shift as the
motor comes up to speed, with torque pulsations at full speed. The solution is to keep the ca
pacitor (impedance) small to minimize losses. The losses are less than for a shaded pole motor.
466 CHAPTER 13. AC MOTORS
This motor configuration works well up to 1/4 horsepower (200watt), though, usually applied
to smaller motors. The direction of the motor is easily reversed by switching the capacitor in
series with the other winding. This type of motor can be adapted for use as a servo motor,
described elsewhere is this chapter.
Figure 13.78: Single phase induction motor with embedded stator coils.
Single phase induction motors may have coils embedded into the stator as shown in Fig
ure 13.78 for larger size motors. Though, the smaller sizes use less complex to build concen
trated windings with salient poles.
13.9.2 Capacitorstart induction motor
In Figure 13.79 a larger capacitor may be used to start a single phase induction motor via
the auxiliary winding if it is switched out by a centrifugal switch once the motor is up to
speed. Moreover, the auxiliary winding may be many more turns of heavier wire than used
in a resistance splitphase motor to mitigate excessive temperature rise. The result is that
more starting torque is available for heavy loads like air conditioning compressors. This motor
configuration works so well that it is available in multihorsepower (multikilowatt) sizes.
Figure 13.79: Capacitorstart induction motor.
13.9. SINGLEPHASE INDUCTION MOTORS 467
13.9.3 Capacitorrun motor induction motor
A variation of the capacitorstart motor (Figure 13.80) is to start the motor with a relatively
large capacitor for high starting torque, but leave a smaller value capacitor in place after start
ing to improve running characteristics while not drawing excessive current. The additional
complexity of the capacitorrun motor is justified for larger size motors.
Figure 13.80: Capacitorrun motor induction motor.
A motor starting capacitor may be a doubleanode nonpolar electrolytic capacitor which
could be two + to + (or  to ) series connected polarized electrolytic capacitors. Such AC rated
electrolytic capacitors have such high losses that they can only be used for intermittent duty
(1 second on, 60 seconds off) like motor starting. A capacitor for motor running must not be of
electrolytic construction, but a lower loss polymer type.
13.9.4 Resistance splitphase motor induction motor
If an auxiliary winding of much fewer turns of smaller wire is placed at 90o electrical to the
main winding, it can start a single phase induction motor. (Figure 13.81) With lower induc
tance and higher resistance, the current will experience less phase shift than the main wind
ing. About 30o of phase difference may be obtained. This coil produces a moderate starting
torque, which is disconnected by a centrifugal switch at 3/4 of synchronous speed. This simple
(no capacitor) arrangement serves well for motors up to 1/3 horsepower (250 watts) driving
easily started loads.
This motor has more starting torque than a shaded pole motor (next section), but not as
much as a two phase motor built from the same parts. The current density in the auxiliary
winding is so high during starting that the consequent rapid temperature rise precludes fre
quent restarting or slow starting loads.
13.9.5 Nola power factor corrrector
Frank Nola of NASA proposed a power factor corrector for improving the efficiency of AC induc
tion motors in the mid 1970’s. It is based on the premise that induction motors are inefficient
at less than full load. This inefficiency correlates with a low power factor. The less than unity
power factor is due to magnetizing current required by the stator. This fixed current is a larger
proportion of total motor current as motor load is decreased. At light load, the full magnetiz
ing current is not required. It could be reduced by decreasing the applied voltage, improving
468 CHAPTER 13. AC MOTORS
VT
IT IStart
IRun
IStart
IStart
IRun IT
IR
IX
φ
VT
φ ≅ 30ο
Figure 13.81: Resistance splitphase motor induction motor.
the power factor and efficiency. The power factor corrector senses power factor, and decreases
motor voltage, thus restoring a higher power factor and decreasing losses.
Since singlephase motors are about 2 to 4 times as inefficient as threephase motors, there
is potential energy savings for 1φ motors. There is no savings for a fully loaded motor since
all the stator magnetizing current is required. The voltage cannot be reduced. But there is
potential savings from a less than fully loaded motor. A nominal 117 VAC motor is designed to
work at as high as 127 VAC, as low as 104 VAC. That means that it is not fully loaded when
operated at greater than 104 VAC, for example, a 117 VAC refrigerator. It is safe for the power
factor controller to lower the line voltage to 104110 VAC. The higher the initial line voltage,
the greater the potential savings. Of course, if the power company delivers closer to 110 VAC,
the motor will operate more efficiently without any addon device.
Any substantially idle, 25% FLC or less, single phase induction motor is a candidate for
a PFC. Though, it needs to operate a large number of hours per year. And the more time it
idles, as in a lumber saw, punch press, or conveyor, the greater the possibility of paying for
the controller in a few years operation. It should be easier to pay for it by a factor of three as
compared to the more efficient 3φmotor. The cost of a PFC cannot be recovered for a motor
operating only a few hours per day. [7]
Summary: Singlephase induction motors
• Singlephase induction motors are not selfstarting without an auxiliary stator winding
driven by an out of phase current of near 90o. Once started the auxiliary winding is
optional.
• The auxiliary winding of a permanentsplit capacitor motor has a capacitor in series with
it during starting and running.
• A capacitorstart induction motoronly has a capacitor in series with the auxiliary winding
during starting.