2 Be able to transform a circuit with a sinusoidal source into the frequency domain using phasor concepts.. 3 Know how to use the following circuit analysis techniques to solve a circ
Trang 1A " 1L J
C H A P T E R C O N T E N T S
9.1 The Sinusoidal Source p 308
9.2 The Sinusoidal Response p 311
9.3 The Phasor p 312
9.4 The Passive Circuit Elements in the
Frequency Domain p 317
9.5 Kirchhoff's Laws in the Frequency
Domain p 321
9.6 Series, Parallel, and Delta-to-Wye
Simplifications p 322
9.7 Source Transformations and
Thevenin-Norton Equivalent Circuits p 329
9.8 The Node-Voltage Method p 332
9.9 The Mesh-Current Method p 333
9.10 The Transformer p 334
9.11 The Ideal Transformer p 338
9.12 Phasor Diagrams p 344
1 Understand phasor concepts and be able to
perform a phasor transform and an inverse
phasor transform
2 Be able to transform a circuit with a sinusoidal
source into the frequency domain using phasor
concepts
3 Know how to use the following circuit analysis
techniques to solve a circuit in the frequency
domain:
• Kirchhoffs laws;
• Series, parallel, and delta-to-wye
simplifications;
• Voltage and current division;
• Thevenin and Norton equivalents;
• Node-voltage method; and
• Mesh-current method
4 Be able to analyze circuits containing linear
transformers using phasor methods
5 Understand the ideal transformer constraints
and be able to analyze circuits containing ideal
transformers using phasor methods
306
Sinusoidal Steady-State Analysis
Thus far, we have focused on circuits with constant sources; in
this chapter we are now ready to consider circuits energized by time-varying voltage or current sources In particular, we are inter-ested in sources in which the value of the voltage or current varies sinusoidally Sinusoidal sources and their effect on circuit behavior form an important area of study for several reasons First, the gen-eration, transmission, distribution, and consumption of electric energy occur under essentially sinusoidal steady-state conditions Second, an understanding of sinusoidal behavior makes it possible
to predict the behavior of circuits with nonsinusoidal sources Third, steady-state sinusoidal behavior often simplifies the design
of electrical systems Thus a designer can spell out specifications in terms of a desired steady-state sinusoidal response and design the circuit or system to meet those characteristics If the device satis-fies the specifications, the designer knows that the circuit will respond satisfactorily to nonsinusoidal inputs
The subsequent chapters of this book are largely based on a thorough understanding of the techniques needed to analyze cir-cuits driven by sinusoidal sources Fortunately, the circuit analysis and simplification techniques first introduced in Chapters 1-4 work for circuits with sinusoidal as well as dc sources, so some of the material in this chapter will be very familiar to you The chal-lenges in first approaching sinusoidal analysis include developing the appropriate modeling equations and working in the mathe-matical realm of complex numbers
Trang 2Practical Perspective
A Household Distribution Circuit
Power systems that generate, transmit, and distribute
electri-cal power are designed to operate in the sinusoidal steady
state The standard household distribution circuit used in the
United States is the three-wire, 240/120 V circuit shown in
the accompanying figure
The transformer is used to reduce the utility distribution
voltage from 13.2 kV to 240 V The center tap on the
second-ary winding provides the 120 V service The operating
fre-quency of power systems in the United States is 60 Hz Both
50 and 60 Hz systems are found outside the United States
^ o y1 0 3
The voltage ratings alluded to above are rms values The rea-son for defining an rms value of a time-varying signal is explained in Chapter 10
307
Trang 39.1 The Sinusoidal Source
A sinusoidal voltage source (independent or dependent) produces a
volt-age that varies sinusoidally with time A sinusoidal current source
(inde-pendent or de(inde-pendent) produces a current that varies sinusoidally with
time In reviewing the sinusoidal function, we use a voltage source, but our
observations also apply to current sources
We can express a sinusoidally varying function with either the sine
function or the cosine function Although either works equally well, we
cannot use both functional forms simultaneously We will use the cosine
function throughout our discussion Hence, we write a sinusoidally varying
voltage as
v = Vm cos (art + 4>) (9.1)
To aid discussion of the parameters in Eq 9.1, we show the voltage
versus time plot in Fig 9.1
Note that the sinusoidal function repeats at regular intervals Such a
function is called periodic One parameter of interest is the length of time
required for the sinusoidal function to pass through all its possible values
This time is referred to as the period of the function and is denoted T It is
measured in seconds The reciprocal of T gives the number of cycles per
second, or the frequency, of the sine function and is denoted /, or
f = f- (9.2)
A cycle per second is referred to as a hertz, abbreviated Hz (The term
cycles per second rarely is used in contemporary technical literature.) The
coefficient of t in Eq 9.1 contains the numerical value of Torf Omega (co)
represents the angular frequency of the sinusoidal function, or
Equation 9.3 is based on the fact that the cosine (or sine) function passes
through a complete set of values each time its argument, cot, passes
through 2-7T rad (360°) From Eq 9.3, note that, whenever t is an integral
multiple of T, the argument cot increases by an integral multiple of 2TT rad
The coefficient V m gives the maximum amplitude of the sinusoidal
voltage Because ±1 bounds the cosine function, ±V m bounds the
ampli-tude Figure 9.1 shows these characteristics
The angle cp in Eq 9.1 is known as the phase angle of the sinusoidal
voltage It determines the value of the sinusoidal function at t = 0;
there-fore, it fixes the point on the periodic wave at which we start measuring
time Changing the phase angle cp shifts the sinusoidal function along the
time axis but has no effect on either the amplitude (V m ) or the angular
fre-quency (co) Note, for example, that reducing cp to zero shifts the sinusoidal
function shown in Fig 9.1 cp/co time units to the right, as shown in Fig 9.2
Note also that if cp is positive, the sinusoidal function shifts to the left,
whereas if cp is negative, the function shifts to the right (See Problem 9.5.)
A comment with regard to the phase angle is in order: cot and cp must
carry the same units, because they are added together in the argument of
the sinusoidal function With cot expressed in radians, you would expect cp
to be also However, cp normally is given in degrees, and cot is converted
from radians to degrees before the two quantities are added We continue
Trang 49.1 The Sinusoidal Source 3 0 9
this bias t o w a r d degrees by expressing t h e phase angle in degrees Recall
from your studies of trigonometry that t h e conversion from radians to
degrees is given by
( n u m b e r of degrees) 18CT
TT ( n u m b e r of radians) (9.4)
Another important characteristic of the sinusoidal voltage (or
cur-rent) is its rms value The rms value of a periodic function is defined as the
square root of the mean value of the squared function Hence, if
v = V m cos (cot + 4>), the rms value of v is
t»+ T
V 2m COS 2 (cot + ¢) dt (9.5)
Note from Eq 9.5 that we obtain the mean value of the squared voltage by
integrating v 2 over one period (that is, from t 0 to t Q + T) and then dividing
by the range of integration, T Note further that the starting point for the
integration t (] is arbitrary
The quantity under the radical sign in Eq 9.5 reduces to V 2 „/2 (See
Problem 9.6.) Hence the rms value of v is
V
-y rms
Vm
vr
(9.6) M rms value of a sinusoidal voltage source
T h e r m s value of t h e sinusoidal voltage d e p e n d s only o n t h e m a x i m u m
amplitude of v, namely, V m T h e r m s value is n o t a function of either t h e
frequency or t h e p h a s e angle We stress t h e i m p o r t a n c e of t h e rms value as
it relates t o p o w e r calculations in C h a p t e r 10 (see Section 10.3)
Thus, we can completely describe a specific sinusoidal signal if we know
its frequency, phase angle, a n d amplitude (either t h e m a x i m u m o r the r m s
value) Examples 9.1, 9.2, and 9.3 illustrate these basic properties of the
sinusoidal function In Example 9.4, we calculate t h e rms value of a periodic
function, and in so doing we clarify the meaning of root mean square
Example 9 1 Finding the Characteristics of a Sinusoidal Current
A sinusoidal current has a maximum amplitude of
20 A The current passes through o n e complete cycle
in 1 ms T h e m a g n i t u d e of the current at zero time
is 10 A
a) What is the frequency of the current in hertz?
b) W h a t is t h e frequency in radians p e r second?
c) Write t h e expression for i(t) using t h e cosine
function Express <£ in degrees
d) What is the rms value of the current?
Solution
a) From the statement of the problem, T = 1 ms;
h e n c e / = 1/T = 1000 Hz
b) to « 277-/ = 2000TT r a d / s c) We h a v e i(t) = I m cos (<ot + ¢) = 20 COS(2000TT/
+ <f>), b u t /(0) = 10 A Therefore 10 = 20 cos 4>
a n d cl> = 60° Thus t h e expression for i(t) becomes
/(f) = 20cos(20007rf + 60°)
d) From the derivation of Eq 9.6, the rms value of a sinusoidal current is / „ , / V 2 Therefore the rms value is 2 0 / V 2 , or 14.14 A
Trang 5310 Sinusoidal Steady-State Analysis
Finding the Characteristics of a Sinusoidal Voltage
A sinusoidal voltage is given by the expression
v = 300 cos (12()77/ + 30°)
a) What is the period of the voltage in milliseconds?
b) What is the frequency in hertz?
c) What is the magnitude of v at t = 2.778 ms?
d) What is the rms value of V?
Solution
a) From the expression for v, to = 12077 rad/s
Because (0 = 2TT/7\ T = 2TT/<O = ^ s,
or 16.667 ms
b) The frequency is 1/71, or 60 Hz
c) From (a), co = 2 77-/ 16.667; thus, at t = 2.778 ms,
at is nearly 1.047 rad, or 60° Therefore,
y(2.778ms) = 300 cos (60° + 30°) = 0 V
d)V ms = 300/ V2 = 212.13 V
Example 9.3 Translating a Sine Expression to a Cosine Expression
We can translate the sine function to the cosine
function by subtracting 90° (TT/2 rad) from the
argu-ment of the sine function
a) Verify this translation by showing that
sin (tot + 0) = cos (tot + 8 - 90°)
b) Use the result in (a) to express sin (cot + 30°) as
a cosine function
Solution
a) Verification involves direct application of the trigonometric identity
cos(a — /3) = cos a cos /3 + sin a sin /3
We let a = ait + 0 and /3 = 90° As cos 90° = 0 and
sin 90° = 1, we have
cos(a -/3)= sin a = sin(atf + 0) = cos(a>/ + 0 - 90°)
b) From (a) we have sin(wr + 30°) = cos(o>/ + 30° - 90°) = cos(atf - 60°)
Example 9.4 Calculating the rms Value of a Triangular Waveform
Calculate the rms value of the periodic triangular
current shown in Fig 9.3 Express your answer in
terms of the peak current I p
7 7 2 \
-Figure 9.3 A Periodic triangular current
Solution
From Eq 9.5, the rms value of i is
^rms \l
-T-Interpreting the integral under the radical sign as the area under the squared function for an interval
of one period is helpful in finding the rms value The squared function with the area between 0 and
T shaded is shown in Fig 9.4, which also indicates
that for this particular function, the area under the
Trang 69.2 The Sinusoidal Response 311
squared current for an interval of one period is
equal to four times the area under the squared
cur-rent for the interval 0 to TfA seconds; that is,
t«+T /.7/4
i 2 dt = 4 / i 2 dt
etc
- 7 / 2 - 7 / 4 0
Figure 9.4 • r versus t
7/4 7/2 37/4 7
The analytical expression for /' in the interval 0 to
774is
4 7
i = -jrt, 0 < / < 774
The area under the squared function for one period is
/ i 2 dt = 4 /
Tlj
r2 3
The mean, or average, value of the function is simply the area for one period divided by the period Thus
1 ^ - i,2
— in'
7 3 3 p '
The rms value of the current is the square root of this mean value Hence
rms V 3 '
NOTE: Assess your understanding of this material by trying Chapter Problems 9.1, 9.4, 9.8
9.2 The Sinusoidal Response
Before focusing on the steady-state response to sinusoidal sources, let's
consider the problem in broader terms, that is, in terms of the total
response Such an overview will help you keep the steady-state solution in
perspective The circuit shown in Fig 9.5 describes the general nature of
the problem There, v s is a sinusoidal voltage, or
v s = V m c o s i^t + </>)• (9.7)
For convenience, we assume the initial current in the circuit to be zero and
measure time from the moment the switch is closed The task is to derive
the expression for /(0 when t > 0 It is similar to finding the step response
of an RL circuit, as in Chapter 7 The only difference is that the voltage
source is now a time-varying sinusoidal voltage rather than a constant, or
dc, voltage Direct application of Kirchhoffs voltage law to the circuit
shown in Fig 9.5 leads to the ordinary differential equation
Figure 9.5 • An RL circuit excited by a sinusoidal
voltage source
L— + Ri = V m cos {a)t + 4>\ (9.8)
the formal solution of which is discussed in an introductory course in
dif-ferential equations We ask those of you who have not yet studied
differ-ential equations to accept that the solution for / is
-v.,
VR 2 + <o 2 L 2 cos (0 -6)e~W L)t + V„
VR 2 + <o2L2
cos (cot + 4> - 0),
(9.9)
Trang 7where 0 is defined as the angle whose tangent is coL/R Thus we can easily
determine 0 for a circuit driven by a sinusoidal source of known frequency
We can check the validity of Eq 9.9 by determining that it satisfies
Eq 9.8 for all values of t > 0; this exercise is left for your exploration in
Problem 9.10
The first term on the right-hand side of Eq 9.9 is referred to as the
transient component of the current because it becomes infinitesimal as
time elapses The second term on the right-hand side is known as the
steady-state component of the solution It exists as long as the switch
remains closed and the source continues to supply the sinusoidal voltage
In this chapter, we develop a technique for calculating the steady-state
response directly, thus avoiding the problem of solving the differential
equation However, in using this technique we forfeit obtaining either the
transient component or the total response, which is the sum of the
tran-sient and steady-state components
We now focus on the steady-state portion of Eq 9.9 It is important to
remember the following characteristics of the steady-state solution:
1 The steady-state solution is a sinusoidal function
2 The frequency of the response signal is identical to the frequency of
the source signal This condition is always true in a linear circuit
when the circuit parameters, R, L, and C, are constant (If
frequen-cies in the response signals are not present in the source signals,
there is a nonlinear element in the circuit.)
3 The maximum amplitude of the steady-state response, in general,
differs from the maximum amplitude of the source For the circuit
being discussed, the maximum amplitude of the response signal is
VJ\/R 2 + arL 2 , and the maximum amplitude of the signal source
is V m
4 The phase angle of the response signal, in general, differs from the
phase angle of the source For the circuit being discussed, the phase
angle of the current is 4> - 0 and that of the voltage source is <f>
These characteristics are worth remembering because they help you
understand the motivation for the phasor method, which we introduce in
Section 9.3 In particular, note that once the decision has been made to
find only the steady-state response, the task is reduced to finding the
max-imum amplitude and phase angle of the response signal The waveform
and frequency of the response are already known
NOTE: Assess your understanding of this material by trying Chapter
Problem 9.9
9.3 The Phasor
The phasor is a complex number that carries the amplitude and phase
angle information of a sinusoidal function.1 The phasor concept is rooted
in Euler's identity, which relates the exponential function to the
trigono-metric function:
e ±jd = cos6» ± / s i n 0 (9.10)
Equation 9.10 is important here because it gives us another way of
express-ing the cosine and sine functions We can think of the cosine function as the
If you feel a bit uneasy about complex numbers, peruse Appendix B
Trang 89.3 The Phasor 313
real part of the exponential function and the sine function as the imaginary
part of the exponential function; that is,
cosfl = Sfc{tf**}, (9.11) and
sin0 = 3{<?'0}, (9.12) where 5ft means "the real part of1 and S means "the imaginary part of."
Because we have already chosen to use the cosine function in
analyz-ing the sinusoidal steady state (see Section 9.1), we can apply Eq 9.11
directly In particular, we write the sinusoidal voltage function given by
Eq 9.1 in the form suggested by Eq 9.11:
v = V m cos (cot 4- (f>)
We can move the coefficient V m inside the argument of the real part of the
function without altering the result We can also reverse the order of the
two exponential functions inside the argument and write Eq 9.13 as
In Eq 9.14, note that the quantity V m e® is a complex number that carries
the amplitude and phase angle of the given sinusoidal function This
complex number is by definition the phasor representation, or phasor
transform, of the given sinusoidal function Thus
V = V m ef* = V{V m cos(cot + ¢)), (9,15) < Phasor transform
where the notation V{V m cos (cot + <f>)} is read "the phasor transform of
Vmcos (cot 4- <£)."Thus the phasor transform transfers the sinusoidal
func-tion from the time domain to the complex-number domain, which is also
called the frequency domain, since the response depends, in general, on to
As in Eq 9.15, throughout this book we represent a phasor quantity by
using a boldface letter
Equation 9.15 is the polar form of a phasor, but we also can express a
phasor in rectangular form Thus we rewrite Eq 9.15 as
V = V m cos $ + jV m sin </> (9.16)
Both polar and rectangular forms are useful in circuit applications of the
phasor concept
One additional comment regarding Eq 9.15 is in order The frequent
occurrence of the exponential function e^ has led to an abbreviation that
lends itself to text material This abbreviation is the angle notation
We use this notation extensively in the material that follows
Trang 9Inverse Phasor Transform
So far we have emphasized moving from the sinusoidal function to its
pha-sor transform However, we may also reverse the process That is, for a
phasor we may write the expression for the sinusoidal function Thus for
V = l ( ) 0 / - 2 6 ° , the expression for v is 100cos (a)t - 26°) because we
have decided to use the cosine function for all sinusoids Observe that we
cannot deduce the value of co from the phasor The phasor carries only
amplitude and phase information The step of going from the phasor
transform to the time-domain expression is referred to as finding the
inverse phasor transform and is formalized by the equation
V~ x {V m e^} = $t{V m e i4 >e> m }, (9.17)
where the notation V~ l {V m e^} is read as "the inverse phasor transform of
V m e'^." Equation 9.17 indicates that to find the inverse phasor transform, we
multiply the phasor by <?/W and then extract the real part of the product
The phasor transform is useful in circuit analysis because it reduces
the task of finding the maximum amplitude and phase angle of the
steady-state sinusoidal response to the algebra of complex numbers The
follow-ing observations verify this conclusion:
1 The transient component vanishes as time elapses, so the
steady-state component of the solution must also satisfy the differential
equation (See Problem 9.10[b].)
2 In a linear circuit driven by sinusoidal sources, the steady-state
response also is sinusoidal, and the frequency of the sinusoidal
response is the same as the frequency of the sinusoidal source
3 Using the notation introduced in Eq 9.11, we can postulate that the
steady-state solution is of the form lR{Ae' li e JM '}, where A is the
maximum amplitude of the response and /3 is the phase angle of the
response
4 When we substitute the postulated steady-state solution into the
differential equation, the exponential term e ,u>l cancels out, leaving
the solution for A and ft in the domain of
complex numbers
We illustrate these observations with the circuit shown in Fig 9.5 (see
p 311) We know that the steady-state solution for the current i is of the form
W O = K { / / f l , (9.18)
where the subscript uss" emphasizes that we are dealing with the
steady-state solution When we substitute Eq 9.18 into Eq 9.8, we generate the
expression
®.{}<oLI m e®ei«*} + n{RI m e iP e J(M } = ^{V m e^ wt \ (9.19)
In deriving Eq 9.19 we recognized that both differentiation and
multiplica-tion by a constant can be taken inside the real part of an operamultiplica-tion We also
rewrote the right-hand side of Eq 9.8, using the notation of Eq 9.11 From
Trang 10the algebra of complex numbers, we know that the sum of the real parts is the
same as the real part of the sum Therefore we may reduce the left-hand side
of Eq 9.19 to a single term:
9R{(/wL + K ^ ' V * } = ftlV^V*"} (9.20)
Recall that our decision to use the cosine function in analyzing the
response of a circuit in the sinusoidal steady state results in the use of
the 5ft operator in deriving Eq 9.20 If instead we had chosen to use the
sine function in our sinusoidal steady-state analysis, we would have
applied Eq 9.12 directly, in place of Eq 9.11, and the result would be
Eq.9.21:
Q{(j(ol + R)I m e ifi e^} = ^{V m e j ^ wt } (9.21)
Note that the complex quantities on either side of Eq 9.21 are identical to
those on either side of Eq 9.20 When both the real and imaginary parts of
two complex quantities are equal, then the complex quantities are
them-selves equal Therefore, from Eqs 9.20 and 9.21,
(J(oL + R)I m e® = V m e i<!>
or
Note that e Ja>t has been eliminated from the determination of the
ampli-tude (/,„) and phase angle (/3) of the response Thus, for this circuit, the
task of finding /,„ and /3 involves the algebraic manipulation of the
com-plex quantities V m e® and R + jwL Note that we encountered both polar
and rectangular forms
An important warning is in order: The phasor transform, along with
the inverse phasor transform, allows you to go back and forth between
the time domain and the frequency domain Therefore, when you obtain
a solution, you are either in the time domain or the frequency domain
You cannot be in both domains simultaneously Any solution that
con-tains a mixture of time domain and phasor domain nomenclature is
nonsensical
The phasor transform is also useful in circuit analysis because it applies
directly to the sum of sinusoidal functions Circuit analysis involves
sum-ming currents and voltages, so the importance of this observation is
obvi-ous We can formalize this property as follows: If
where all the voltages on the right-hand side are sinusoidal voltages of the
same frequency, then