You are on page 1of 89

Exercises 135

h. Laplace transform Unidirectional, operational blocks that represent the


transfer functions of the elements of the system.
i. Linear A rule that enables the user to obtain a transfer
approximation function by tracing paths and loops within a system.
j . Linear system An electric actuator that uses an input voltage as
a control variable.
k. Mason loop rule The ratio of the Laplace transform of the output
variable to the Laplace transform of the input variable.
1. Mathematical Descriptions of the behavior of a system using
models mathematics.
A model of a system that is used to investigate the
m. Signal-flow graph behavior of a system by utilizing actual input signals.
A diagram that consists of nodes connected by several
n. Simulation directed branches and that is a graphical representation
of a set of linear relations.
o. Transfer function An approximate model that results in a linear relationship
between the output and the input of the device.

EXERCISES

Exercises are straightforward applications of the concepts


of the chapter.
Spring
E2.1 A unity, negative feedback system has a nonlinear breaks
function y = /(e) = e2, as shown in Figure E2.1. For an HIh
I fc Displacement
"' (em)
input r in the range of 0 to 4, calculate and plot the open-
loop and closed-loop output versus input and show that
the feedback system results in a more linear relationship. Springf
compresses

FIGURE E2.3 Spring behavior.

E2.4 A laser printer uses a laser beam to print copy


Close switch for closed loop rapidly for a computer. The laser is positioned by a
control input r(t), so that we have
FIGURE E2.1 Open and closed loop.
4(s + 50)
E2.2 A thermistor has a response to temperature repre- 7(5)- R(s).
s2 + 30s + 200
sented by
R = /^-01-^ The input r(t) represents the desired position of the
laser beam.
where R0 = 10,000 ft, R = resistance, and T = tem-
perature in degrees Celsius. Find the linear model for (a) If r(t) is a unit step input, find the output y(t).
the thermistor operating at T = 20C and for a small (b) What is the final value of y{t)l
range of variation of temperature. Answer: (a) y{t) = 1 + 0.6<T20' - 1.6<T10', (b) yss = 1
Answer: AR = -135AF E2.5 A noninverting amplifier uses an op-amp as shown
E2.3 The force versus displacement for a spring is shown in Figure E2.5. Assume an ideal op-amp model and
in Figure E2.3 for the spring-mass-damper system of determine v0/vm.
Figure 2.1. Graphically find the spring constant for the
equilibrium point of y = 0.5 cm and a range of opera- Answer: - 1 +
tion of 1.5 cm.
136 Chapter 2 Mathematical Models of Systems

2 +
^Wv- r~\
R(s) Gito G2W Hs)
_ . i

+ 0-
WW

TT (a)

FIGURE E2.5 A noninverting amplifier using an op-amp.


Filter

E2.6 A nonlinear device is represented by the function

y = /{X) = e*, A N

where the operating point for the input x is x0 = 1. Iris \ Opaque tube
Determine a linear approximation valid near the oper-
ating point. fb)

Answer: y = ex FIGURE E2.7 Lamp controller.


E2.7 A lamp's intensity stays constant when monitored by
an optotransistor-controlled feedback loop. When the
voltage drops, the lamp's output also drops, and opto-
transistor Q\ draws less current. As a result, a power Answer:
R(S)
transistor conducts more heavily and charges a capaci-
tor more rapidly [24]. The capacitor voltage controls KG,(*)G 2 (s)/j
the lamp voltage directly. A block diagram of the sys- 1 + G1(s)H?,(s) + <h(*yGffMiW + //*(*)] + KCh(s)GAs)/s
tem is shown in Figure E2.7. Find the closed-loop trans-
fer function, I(s)!R(s) where I{s) is the lamp intensity,
E2.9 A four-wheel antilock automobile braking system
and R(s) is the command or desired level of light.
uses electronic feedback to control automatically the
E2.8 A control engineer, N. Minorsky, designed an innov- brake force on each wheel [15]. A block diagram
ative ship steering system in the 1930s for the U.S. model of a brake control system is shown in Figure E2.9,
Navy. The system is represented by the block diagram where iy(s) and FR(s) are the braking force of the
shown in Figure E2.8, where Y(s) is the ship's course, front and rear wheels, respectively, and R{s) is the
/?(.?) is the desired course, and A(s) is the rudder angle desired automobile response on an icy road. Find
[16]. Find the transfer function Y(s)IR(s). Ff(s)/R(s).

H2{s)

A
R(s)
ky4 G?iW G2(s) -* I
s
Y(s)

ffjW -

//,(')

FIGURE E2.8 Ship steering system.


Exercises 137

H2(s)

G2(.t) * FAx) Plunger


R(s\ G,(s)

G3(s) * Fsis)

H2(s)

FIGURE E2.9 Brake control system.


Damping
orifice
E2.10 One of the most potentially beneficial applications Piston rod Piston travel
of an automotive control system is the active control of
the suspension system. One feedback control system FIGURE E2.10 Shock absorber.
uses a shock absorber consisting of a cylinder filled
with a compressible fluid that provides both spring and
damping forces [17].The cylinder has a plunger activat-
ed by a gear motor, a displacement-measuring sensor,
and a piston. Spring force is generated by piston dis-
placement, which compresses the fluid. During piston
displacement, the pressure unbalance across the piston
is used to control damping. The plunger varies the in-
ternal volume of the cylinder. This feedback system is
shown in Figure E2.10. Develop a linear model for this
device using a block diagram model.
E2.ll A spring exhibits a force-versus-displacement
characteristic as shown in Figure E2.ll. For small de-
viations from the operating point x0, find the spring
constant when x0 is (a) -1.4; (b) 0; (c) 3.5.
FIGURE E2.11 Spring characteristic.
E2.12 Off-road vehicles experience many disturbance
inputs as they traverse over rough roads. An active
suspension system can be controlled by a sensor that gain Kx so that the vehicle does not bounce when the
looks "ahead" at the road conditions. An example of a desired deflection is R{s) = 0 and the disturbance is
simple suspension system that can accommodate the Us).
bumps is shown in Figure E2.12. Find the appropriate Answer: K^K^ = 1

Bump disturbance

Preview of disturbance
1
'j( r;

<

*i

dynamics Bounce of
Desired >
deflection
J
+ S J
s K2
+ ..

T<~>
G(.v)
m auto or

._ horizontal

FIGURE E2.12 Active suspension system.


138 Chapter 2 Mathematical Models of Systems

E2.13 Consider the feedback system in Figure E2.13. Com-


- = 06*
pute the transfer functions Y(s)/Td(s) and Y(s)/N(s).
E2.14 Find the transfer function V2 = lV\.
Yj(s) The variables involved are as follows:
R2(s)
r(t) = desired platform position
for the multivariate system in Figure E2.14.
p{t) = actual platform position
E2.15 Obtain the differential equations for the circuit in
Figure E2.15 in terms of ^ and i2. V\{l) = amplifier input voltage
E2.16 The position control system for a spacecraft plat- v2(t) = amplifier output voltage
form is governed by the following equations: 0(/) = motor shaft position
d2p dp
f + 2 - p + 4p = B Sketch a signal-flow diagram or a block diagram of
2
dt dt the system, identifying the component parts and de-
Vi=r- p termine the system transfer function P(s)/R(s).

Ms)

FIGURE E2.13 Feedback system with measurement noise, A/(s), and plant
disturbances, Td(s).

Hl(5)

Gfr)
+
r~\ +x~
+v
.(*) G2(s) Gjlis) " W
+
1 k

G7(s) G%(s) G9{s)

i k

R2(s) G4{s) -iF k


G5{s) Gb{s) Ki(.V)

H2(s)

FIGURE E2.14 Multivariate system.


Exercises 139
E2.21 A high-precision positioning slide is shown in Figure
E2.21. Determine the transfer function Xp(s)/Xm(s)
when the drive shaft friction is bd = 0.7, the drive shaft
spring constant is kd = 2, mc = 1, and the sliding
friction is bs = 0.8.
(f)

FIGURE E2.15 Electric circuit.

E2.17 A spring develops a force /represented by the rela- Sliding


tion friction, b.

/ = kx2,
where x is the displacement of the spring. Determine
a linear model for the spring when x0 = j - FIGURE E2.21 Precision slide.
E2.18 The output y and input x of a device are related by
y = x + 1.4x3.
E2.22 The rotational velocity &> of the satellite shown in
(a) Find the values of the output for steady-state op- Figure E2.22 is adjusted by changing the length of the
eration at the two operating points x0 = 1 and x0 = 2. beam L. The transfer function between <x)(s) and the
(b) Obtain a linearized model for both operating incremental change in beam length AL(s) is
points and compare them.
w(s) 2{s + 4)
E2.19 The transfer function of a system is
AZ-(.v) (s + 5)(s + 1)2
Y(s) _ 15(.f + 1)
R(s) ~ s2 + 9s + 14' The beam length change is AL(i) = 1/s. Determine
the response of the rotation co(t).
Determine y{t) when r(t) is a unit step input.
Answer: (r) = 1.6 + 0.025e~5' - 1.625-' - 1.5te-'
Answer: y(t) = 1.07 + l i e - * - 2.57e-7', t s 0
E2.20 Determine the transfer function VQ(s)/V{s) of the op-
erational amplifier circuit shown in Figure E2.20. Assume
an ideal operational amplifier. Determine the transfer
function when /?, = R2 = 100 kfl, Cx = 10 jttF, and
C2 = 5 fiF.

C,

-1(-
i^l'
Rotation

t
* o+
FIGURE E2.22 Satellite with adjustable rotational velocity.
-o

E2.23 Determine the closed-loop transfer function T(s) =


FIGURE E2.20 Op-amp circuit. Y(s)/R(s) for the system of Figure E2.23.
140 Chapter 2 Mathematical Models of Systems

E2.26 Determine the transfer function X2(s)/F(s) for the


system shown in Figure E2.26. Both masses slide on a
frictionless surface, and k = 1 N/m.
/?(.v) O X2(s) 1
Answer:
F(s) s2(s2 + 2)

FIGURE E2.23 Control system with three feedback loops.


/*(/) MA/W-
E2.24 The block diagram of a system is shown in
Figure E2.24. Determine the transfer function
T(s) = Y(s)/R(s).
FIGURE E2.26 Two connected masses on a frictionless
surface.
10
R(s) Y(s)
E2.27 Find the transfer function Y(s)/Td(s) for the sys-
s+ 1
tem shown in Figure E2.27.

Y(s) G^s)
Answer:
Td(s) 1 + G,(s)G2(s)H(s)

O TAs)

FIGURE E2.24 Multiloop feedback system.


C,(s) G2{s) Yis)
> * - & *
E2.25 An amplifier may have a region of deadband as
shown in Figure E2.25. Use an approximation that
uses a cubic equation y = ax3 in the approximately
linear region. Select a and determine a linear approxi- H(s)
mation for the amplifier when the operating point is
JC = 0.6.
FIGURE E2.27 System with disturbance.

FIGURE E2.25
An amplifier with a
deadband region.
Problems 141
E2.28 Determine the transfer function \&(s)/V(s) for the (b) Determine Y(s)/R(s) for Figure E2.29(b).
op-amp circuit shown in Figure E2.28 [1]. Let /?j = E2.30 A system is shown in Figure E2.30.
167 kfl, R2 = 240 kH, R3 = 1 kH, RA = 100 kH, and
C = 1 /iF. Assume an ideal op-amp. (a) Find the closed-loop transfer function Y(s)/R(s)
E2.29 A system is shown in Fig. E2.29(a). 10
when G(s) = - : .
(a) Determine G(s) and H(s) of the block diagram
s2 + 2s + 10
shown in Figure E2.29(b) that are equivalent to
(b) Determine Y(s) when the input R(s) is a unit step.
those of the block diagram of Figure E2.29(a).
(c) Compute y(t).

3r
o*WV"
+ R,
-o +

FIGURE E2.28
Op-amp circuit.
_L _n
ns)
1 /?(*)

-k> s+ 10

(a) FIGURE E2.30 Unity feedback control system.

E2.31 Determine the partial fraction expansion for V(s)


R(s) n.v) and compute the inverse Laplace transform. The
transfer function V(s) is given by:
400
V(s)
s2 + Ss + 400
(b)
FIGURE E2.29 Block diagram equivalence.

PROBLEMS
Problems require an extension of the concepts of the chap-
ter to new situations.
P2.1 An electric circuit is shown in Figure P2.1. Obtain a
set of simultaneous integrodifferential equations rep-
resenting the network.
P2.2 A dynamic vibration absorber is shown in Figure *0
P2.2. This system is representative of many situations
involving the vibration of machines containing unbal-
anced components. The parameters M2 and kl2 may
be chosen so that the main mass Mi does not vibrate
in the steady state when F(t) = a sin(a>0f)- Obtain the
differential equations describing the system. FIGURE P2.1 Electric circuit.
142 Chapter 2 Mathematical Models of Systems

for the fluid-flow equation, (b) What happens to the


.J approximation obtained in part (a) if the operating
point is Pi - P2 = 0?
Force <;*! L
P2.6 Using the Laplace transformation, obtain the current
m > J
" I2(s) of Problem P2.1. Assume that all the initial cur-
AT, | v .<') rents are zero, the initial voltage across capacitor C\ is
zero, v{t) is zero, and the initial voltage across C2 is 10
volts.
P2.7 Obtain the transfer function of the differentiating
circuit shown in Figure P2.7.
ftf2 T>'2<'>

FIGURE P2.2 Vibration absorber.

P2.3 A coupled spring-mass system is shown in Figure


1
P2.3. The masses and springs are assumed to be equal.
Obtain the differential equations describing the system. +
VAs) V2[s)

- I ' I I O

Force v,(r) -* A',(0


ru) FIGURE P2.7 A differentiating circuit.
I

k
M WWWH u Hi P2.8 A bridged-T network is often used in AC control
systems as a filter network [8]. The circuit of one
k nzn bridged-T network is shown in Figure P2.8. Show that
the transfer function of the network is
FIGURE P2.3 Two-mass system.
V&) 1 + IR^Cs + RiRjpV
P2.4 A nonlinear amplifier can be described by the fol- Kn(i') 1 + (2, + R2)Cs + i ? , i ? 2 c V
lowing characteristic:
Sketch the pole-zero diagram when Rx = 0.5, = 1,
"o(') = J 4
and C = 0.5.
m < 0'
I 4
Tlie amplifier will be operated over a range of 0.5
volts around the operating point for vin. Describe the wv
amplifier by a linear approximation (a) when the op-
erating point is sjj,, = 0 and (b) when the operating
point is win = 1 volt. Obtain a sketch of the nonlinear
function and the approximation for each case.
P2.5 Fluid flowing through an orifice can be represented
by the nonlinear equation
Q = K(P, - A)" 2 , FIGURE P2.8 Bridged-T network.
where the variables are shown in Figure P2.5 and K is
a constant [2]. (a) Determine a linear approximation P2.9 Determine the transfer function Xi(s)/F(s) for the
coupled spring-mass system of Problem P2.3. Sketch
the s-plane pole-zero diagram for low damping when
M = l,b/k = l,and

4 0.1.
2-
P2.10 Determine the transfer function Yi{s)jF(s) for the
FIGURE P2.5 Flow through an orifice. vibration absorber system of Problem P2.2. Determine
Problems 143
the necessary parameters M2 and &12 so that the mass [8,19]. An amplidyne is a power amplifying rotary am-
Ml does not vibrate in the steady state when plifier. An amplidyne and a servomotor are shown in
F(t) a sin(&)o t). Figure P2.ll. Obtain the transfer function 9(s)/Vc(s),
P2.ll For electromechanical systems that require large and draw the block diagram of the system. Assume
power amplification, rotary amplifiers are often used vd = k2iq and vq = k{ic.

Control
field
it = Constant

lr/ \ 2 (
P \ Motor 1
1
V
f-
4 Load y, b

Am plidyne **

FIGURE P2.11 Amplidyne and armature-controlled motor.

P2.12 For the open-loop control system described by the the transfer function BL{s)fVf{s) and draw a block dia-
block diagram shown in Figure P2.12, determine the gram of the system. The generator voltage can be as-
value of K such that y(t) - * 1 as t oo when r(r) is a sumed to be proportional to the field current if.
unit step input. Assume zero initial conditions. P2.14 A rotating load is connected to a field-controlled
DC electric motor through a gear system. The motor is
assumed to be linear. A test results in the output load
Controller Process reaching a speed of 1 rad/s within 0.5 s when a constant
1 80 V is applied to the motor terminals. The output
Ms) A:
K -- YU) steady-state speed is 2.4 rad/s. Determine the transfer
s+20
function 0{s)/Vf(s) of the motor, in rad/V. The induc-
tance of the field may be assumed to be negligible (see
FIGURE P2.12 Open-loop control system. Figure 2.18). Also, note that the application of 80 V to
the motor terminals is a step input of 80 V in magnitude.
P2.13 An electromechanical open-loop control system is P2.15 Consider the spring-mass system depicted in Figure
shown in Figure P2.13. The generator, driven at a con- P2.15. Determine a differential equation to describe
stant speed, provides the field voltage for the motor. The the motion of the mass m. Obtain the system response
motor has an inertia Jm and bearing friction />,. Obtain x(t) with the initial conditions A(0) = Xg and i(0) = 0.

+ o-vV\A-| Motor

N
01 i
Gear ratio n =
No

Generator

FIGURE P2.13 Motor and generator.


144 Chapter 2 Mathematical Models of Systems

Obtain the relationship 7[3(s) between X^(s) and


X3(s) by using Mason's signal-flow gain formula.
Compare the work necessary to obtain 7^0) by ma-
k, spring constant trix methods to that using Mason's signal-flow gain
formula.
P2.18 An LC ladder network is shown in Figure P2.18.
One may write the equations describing the network
as follows:
h = (Vi ~ Vu)Yh Va = (J, - /0)Z2,
I, = K - v2)y3, v2 = / a z 4 .
Construct a flow graph from the equations and deter-
mine the transfer function K(s)/Vi(.r).
FIGURE P2.15 Suspended spring-mass system.

P2.16 Obtain a signal-flow graph to represent the follow- h L V L


0
JTYYV
ing set of algebraic equations where x\ and x2 are to Y,1
be considered the dependent variables and 6 and 11
are the inputs:
VAs) Z2
=
Xi + 1.5 6, 2JC, 4A-, 11.

Determine the value of each dependent variable by FIGURE P2.18 LC ladder network.
using the gain formula. After solving for JCJ by Mason's
signal-flow gain formula, verify the solution by using
P2.19 A voltage follower (buffer amplifier) is shown in
Cramers rule.
Figure P2.19. Show that T = vQ/vin = 1. Assume an
P2.17 A mechanical system is shown in Figure P2.17, ideal op-amp.
which is subjected to a known displacement x$(t) with
respect to the reference, (a) Determine the two inde-
pendent equations of motion, (b) Obtain the equations
of motion in terms of the Laplace transform, assuming
that the initial conditions are zero, (c) Sketch a signal-
flow graph representing the system of equations, (d) + 0-

Friction

,r~ A/, *i
FIGURE P2.19 A buffer amplifier.

,r" i
f.
M3 6,
P2.20 The source follower amplifier provides lower out-
put impedance and essentially unity gain. The circuit
diagram is shown in Figure P2.20(a), and the small-sig-
nal model is shown in Figure P2.20(b).This circuit uses
an FET and provides a gain of approximately unity.

*4i i
<
r J ZJ
Assume that R2 R] for biasing purposes and that
Rg R2. (a) Solve for the amplifier gain, (b) Solve
for the gain when gm = 2000 (tl and Rs = 10 kil
where Rs = Ry + R2. (c) Sketch a block diagram that

1 r
represents the circuit equations.
P2.21 A hydraulic servomechanism with mechanical
feedback is shown in Figure P2.21 [18]. The power pis-
ton has an area equal to A. When the valve is moved a
small amount Az, the oil will flow through to the cylin-
FIGURE P2.17 Mechanical system. der at a rate p Az, where p is the port coefficient. The
Problems 145

input oil pressure is assumed to be constant. From the


geometry, we find that Az = &-(x - y) - y.
h h
(a) Determine the closed-loop signal-flow graph or
block diagram for this mechanical system, (b) Obtain
the closed-loop transfer function Y{s)/X(s).
P2.22 Figure P2.22 shows two pendulums suspended
from frictionless pivots and connected at their mid-
points by a spring [1]. Assume that each pendulum can
be represented by a mass M a t the end of a massless
bar of length L. Also assume that the displacement is
small and linear approximations can be used for sin 8
and cos 8. The spring located in the middle of the bars
is unstretched when fy = 82. The input force is repre-
sented by /(r), which influences the left-hand bar only,
(a)
(a) Obtain the equations of motion, and sketch a
block diagram for them, (b) Determine the transfer
function T(s) = 8i(s)/F(s). (c) Sketch the location of

v
(P*"."* the poles and zeros of T(s) on the s-plane.
'in G gs
->
>*a >xi
"it Q

02

n/WWW
(b)
FIGURE P2.20 The source follower or common drain
amplifier using an FET.

Power FIGURE P2.22 The bars are each of length L and the
spring is located at L/2.
cylinder

P2.23 The small-signal circuit equivalent to a common-


emitter transistor amplifier is shown in Figure P2.23.
Input
The transistor amplifier includes a feedback resistor
pressure
Rf. Determine the input-output ratio vcJv-m.

->WSr
'c
rWV-o-*-^vVv1 -4 6-
*(* - v)

*. v
bf ,,,, A

FIGURE P2.23 CE amplifier.


I Output, v

P2.24 A two-transistor series voltage feedback amplifier


FIGURE P2.21 Hydraulic servomechanism. is shown in Figure P2.24(a). This AC equivalent circuit
146 Chapter 2 Mathematical Models of Systems

neglects the bias resistors and the shunt capacitors. A sign technique known as feedforward correction [19],
block diagram representing the circuit is shown in Recent experiments have shown that this technique
Figure P2.24(b).This block diagram neglects the effect offers the potential for yielding excellent amplifier
of hn., which is usually an accurate approximation, and stabilization. Black's amplifier is shown in Figure
assumes that R2 + RL R\- (a) Determine the volt- P2.25(a) in the form recorded in 1924. The block dia-
age gain vjvin. (b) Determine the current gain ia/lbi- gram is shown in Figure P2.25(b). Determine the
(c) Determine the input impedance V\Jib\. transfer function between the output Y(s) and the
input R(s) and between the output and the distur-
P2.25 H. S. Black is noted for developing a negative feed-
bance Td(s). G(s) is used to denote the amplifier rep-
back amplifier in 1927. Often overlooked is the fact
resented by fi in Figure P2.25(a).
that three years earlier he had invented a circuit de-

'ft i R

+ R,
M "ir

(a; (bl

FIGURE P2.24 Feedback amplifier.

A'(v)
ns)

1
G(s)
J G(s)

H^\P- J
.

FIGURE P2.25 HS
(a) (h)
Black's amplifier.

P2.26 A robot includes significant flexibility in the arm


members with a heavy load in the gripper [6, 20]. A
two-mass model of the robot is shown in Figure. P2.26.
Find the transfer function Y(s)IF(s).
P2.27 Magnetic levitation trains provide a high-speed, Pit)'
very low friction alternative to steel wheels on steel
rails. The train floats on an air gap as shown in Figure VWWVWA
P2.27 [25]. The levitation force FL is controlled by the k
coil current i in the levitation coils and may be ap- FIGURE P2.26 The spring-mass-damper model of a
proximated by robot arm.

where z is the air gap. This force is opposed by the


V downward force F = mg. Determine the linearized
Problems 147
relationship between the air gap z and the controlling input current i controls the torque with negligible fric-
current near the equilibrium condition. tion. Assume the beam may be balanced near the hor-
izontal (<f> = 0); therefore, we have a small deviation
of <f>. Find the transfer function X(s)/I(s). and draw a
block diagram illustrating the transfer function show-
ing (5), X(s), and T(s).
P2.30 The measurement or sensor element in a feedback
system is important to the accuracy of the system [6].
The dynamic response of the sensor is important.
Most sensor elements possess a transfer function

T.S + 1

Suppose that a position-sensing photo detector has


T = 4,us and 0.999 < k < 1.001. Obtain the step re-
sponse of the system, and find the k resulting in the
fastest responsethat is, the fastest time to reach 98%
of the final value.
FIGURE P2.27 Cutaway view of train. P2.31 An interacting control system with two inputs and
two outputs is shown in Figure P2.31. Solve for
Yt(s)/Ri(s) and Y2(s)/R1(s) when R2 = 0.
P2.28 A multiple-loop model of an urban ecological sys-
tem might include the following variables: number of
people in the city (P), modernization (M), migration
into the city (C), sanitation facilities (S), number of
diseases (D), bacteria/area (B), and amount of
garbage/area (G), where the symbol for the variable is
given in parentheses. The following causal loops are
hypothesized: KM Yds)
1. P^G^B^D-^P
2. P-*M^C-^P
3. P-*-MS-*D-*P
4. P^>M-*S^B-*D^>P
Sketch a signal-flow graph for these causal relation- ffji*) . v i: i i :i r,(.v)
ships, using appropriate gain symbols. Indicate whether
you believe each gain transmission is positive or nega-
tive. For example, the causal link S to B is negative be-
cause improved sanitation facilities lead to reduced
bacteria/area. Which of the four loops are positive feed- FIGURE P2.31 Interacting System.
back loops and which are negative feedback loops?
P2.29 We desire to balance a rolling ball on a tilting beam
as shown in Figure P2.29. We will assume the motor P2.32 A system consists of two electric motors that are
coupled by a continuous flexible belt. The belt also
passes over a swinging arm that is instrumented to
Torque motor allow measurement of the belt speed and tension. The
basic control problem is to regulate the belt speed and
tension by varying the motor torques.
An example of a practical system similar to that
shown occurs in textile fiber manufacturing processes
when yarn is wound from one spool to another at high
speed. Between the two spools, the yarn is processed
in a way that may require the yarn speed and tension
to be controlled within defined limits. A model of the
system is shown in Figure P2.32. Find J5(s)/i?j{5), De-
termine a relationship for the system that will make K
FIGURE P2.29 Tilting beam and ball. independent of jRj.
148 Chapter 2 Mathematical Models of Systems

-H2(s)

Speed Speed
control
input

R2(S) O r2[s)
Tension Tension
FIGURE P2.32 control
A model of the input
coupled motor
-HM
drives.

FIGURE P2.33 Idle speed control system.

P233 Find the transfer function for Y(s)/R(s) for the idle-
speed control system for a fuel-injected engine as
shown in Figure P2.33.
P2.34 The suspension system for one wheel of an old-
fashioned pickup truck is illustrated in Figure P2.34.
The mass of the vehicle is m% and the mass of the wheel
is m2-The suspension spring has a spring constant k^ and
the tire has a spring constant k2. The damping con-
stant of the shock absorber is b. Obtain the transfer
function Y\(s)j'X(s), which represents the vehicle re-
sponse to bumps in the road.
P2.35 A feedback control system has the structure shown FIGURE P2.34 Pickup truck suspension.
in Figure P2.35. Determine the closed-loop transfer
function Y(s)/R(s) (a) by block diagram manipulation so that the closed-loop response to a step input is crit-
and (b) by using a signal-flow graph and Mason's sig- ically damped with two equal roots at s = -10. (d)
nal-flow gain formula, (c) Select the gains /C, and K2 Plot the critically damped response for a unit step
Problems 149

1
tf(.v) ' YU)
s +1 s
K < *
1

K2
'
l
X

FIGURE P2.35 Multiloop feedback system.

input. What is the time required for the step response


to reach 90% of its final value?
P2.36 A system is represented by Figure P2.36. (a) Deter- FIGURE P2.37 Two-mass system.
mine the partial fraction expansion and y{t) for a ramp
input, /(f) = t, t > 0. (b) Obtain a plot of y(t) for part
(a), and find y(t) for l = 1.0 s. (c) Determine the im-
pulse response of the system v(/) for ( 2 0. (d) Obtain
a plot of y(t) for part (c) and find y(i) for ( = 1.0 s.

-0.5 m-
24
RU) * n.v)

4
s* + 9s2 + 26s + 24

FIGURE P2.36 A third-order system.


Q-
FIGURE P2.38 Winding oscillator.
P237 A two-mass system is shown in Figure P2.37 with an
input force u(t). When m | = m2 = l a n d ^ i = K2 = 1,
find the set of differential equations describing the viscous friction coefficient for the sphere in air is
system. 2 X 10~4 N m s/rad. The sphere has a mass of 1 kg.
P238 A winding oscillator consists of two steel spheres P2.39 For the circuit of Figure P2.39, determine the trans-
on each end of a long slender rod, as shown in form of the output voltage V0(s). Assume that the cir-
Figure P2.38. The rod is hung on a thin wire that can cuit is in steady state when t < 0. Assume that the
be twisted many revolutions without breaking. The switch moves instantaneously from contact 1 to con-
device will be wound up 4000 degrees. How long will tact 2 at t = 0.
it take until the motion decays to a swing of only 10 P2.40 A damping device is used to reduce the undesired
degrees? Assume that the thin wire has a rotational vibrations of machines. A viscous fluid, such as a
spring constant of 2 X 10~ 4 Nm/rad and that the heavy oil, is placed between the wheels, as shown in

2H 1 P
1
~^^77Q
/YYY\
-If
?2fi < > 0.5

6V lOe-aV

FIGURE P2.39 4ft


Model of an
electronic circuit.
150 Chapter 2 Mathematical Models of Systems

. Outer wheel shown in Figure P2.42. As the mirror rotates, a friction


force is developed that is proportional to its angular
speed. The friction constant is equal to 0.06 N s/rad,
. Inner wheel and the moment of inertia is equal to 0.1 kg m2. The
Shaft / ./,,0, output variable is the velocity cu(r). (a) Obtain the dif-
% ferential equation for the motor, (b) Find the response
of the system when the input motor torque is a unit
step and the initial velocity at J = 0 is equal to 0.7.
Fluid, b
Mirror
FIGURE P2.40 Cutaway view of damping device. Bar code

Figure P2.40. When vibration becomes excessive, the


relative motion of the two wheels creates damping.
When the device is rotating without vibration, there is
no relative motion and no damping occurs. Find B^s) Reflected light
and 02(s). Assume that the shaft has a spring constant
K and that b is the damping constant of the fluid. The
load torque is T.
P2.41 The lateral control of a rocket with a gimbaled en-
gine is shown in Figure P2.41. The lateral deviation Microcomputer
from the desired trajectory is h and the forward rock-
et speed is V. The control torque of the engine is . FIGURE P2.42 Optical scanner.
and the disturbance torque is Ttf. Derive the describ-
ing equations of a linear model of the system, and P2.43 An ideal set of gears is shown in Table 2.5, item 10.
draw the block diagram with the appropriate transfer Neglect the inertia and friction of the gears and as-
functions. sume that the work done by one gear is equal to that
of the other. Derive the relationships given in item 10
Desired Aclua] of Table 2.5. Also, determine the relationship between
trajectory trajectory the torques Tm and TL.
P2.44 An ideal set of gears is connected to a solid cylin-
der load as shown in Figure P2.44. The inertia of the
motor shaft and gear G2 is Jm. Determine (a) the iner-
tia of the load JL and (b) the torque T at the motor
shaft. Assume the friction at the load is bL and the fric-
tion at the motor shaft is bm. Also assume the density
of the load disk is p and the gear ratio is n. Hint: The
torque at the motorshaft is given by T = T\ + Tm.

Engine

FIGURE P2.44 Motor, gears, and load.


FIGURE P2.41 Rocket with gimbaled engine.

P2.42 In many applications, such as reading product P2.45 To exploit the strength advantage of robot manipu-
codes in supermarkets and in printing and manufac- lators and the intellectual advantage of humans, a class
turing, an optical scanner is utilized to read codes, as of manipulators called extenders has been examined
Problems 151
[22]. The extender is defined as an active manipulator P2.47 The water level h{t) in a tank is controlled by an
worn by a human to augment the human's strength. The open-loop system, as shown in Figure P2.47. A DC
human provides an input U(s), as shown in Figure motor controlled by an armature current ;' turns a
P2.45. The endpoint of the extender is P(s). Determine shaft, opening a valve. The inductance of the DC
the output P(s) for both U(s) and F(s) in the form motor is negligible, that is, La = 0. Also, the rota-
tional friction of the motor shaft and valve is negli-
P(s) = T^Uis) + T2(s)F(s). gible, that is, b = 0. The height of the water in the
tank is

h(t)
J [1.60(f) ~ h(t)]dt.

Human
w
v.

H(s)
r '
G(s)
B . Performance ,,
filter f
| B(s)
the motor constant is K, = 10, and the inertia of the
motor shaft and valve is J - 6 X KT3 kgm 2 . Deter-
mine (a) the differential equation for h(t) and v(t) and
i
,
,+ L (b) the transfer function H(s)IV(s).
Pis)
r
K. >r GiW * K(s) *-n P2.48 The circuit shown in Figure P2.48 is called a lead-
lag filter.
1

-V
stability t
V+
i
(a) Find the transfer function V2(s)/\{(s). Assume an
controller ideal op-amp.
(*) Gc{s)
(b) Determine V2(s)/V^s) when l?j = 100Hl,
Load i

I R2 = 200 kl, Q = 1 /JLF, and C2 = 0.1 fiF.


j

J J
(c) Determine the partial fraction expansion for

P2.49 A closed-loop control system is shown in Figure


P2.49.
FIGURE P2.45 Model of extender.
(a) Determine the transfer function
T{s) = Y(s)/R(s).
P2.46 A load added to a truck results in a force F on the
support spring, and the tire flexes as shown in Figure (b) Determine the poles and zeros of T(s).
P2.46(a).The model for the tire movement is shown in (c) Use a unit step input, .SKY) = 1/s, and obtain the
Figure P2.46(b). Determine the transfer function partial fraction expansion for Y(s) and the value
X,(s)/F(s). of the residues.

Force of material
placed in truck bed
Truck vehicle mass

*i r Shock absorber

FIGURE P2.46
Truck support
model. la) (h)
152 Chapter 2 Mathematical Models of Systems

Amplifier

Valve

FIGURE P2.47
Open-loop control
system for the
water level of a
tank.

(e) Predict the final value of y(t) for the unit step

)li r^Mh input.

-AAAr 14,000
R(s) * i'(-v)
,! 3 + 4 5 ^ + 31005 + 500)
ViM V2(s)

FIGURE P2.50 Third-order feedback system.


FIGURE P2.48 Lead-lag filter.
P2.51 Consider the two-mass system in Figure P2.51.
Find the set of differential equations describing the
6205 system.
tf(.v) ' * Y(s)
s(s-+ 13*+ 1281)

FIGURE P2.49 Unity feedback control system.

(d) Plot y(t) and discuss the effect of the real and
complex poles of T(s). Do the complex poles or
the real poles dominate the response?
P2.50 A closed-loop control system is shown in Figure
P2.50.
(a) Determine the transfer function T(s) = Y(s)/R(s).
(b) Determine the poles and zeros of T(s).
(c) Use a unit step input, R(s) = l/s, and obtain the
partial fraction expansion for Y(s) and the value
of the residues.
(d) Plot y(() and discuss the effect of the real and
complex poles of T(s). Do the complex poles or FIGURE P2.51 Two-mass system with two springs and
the real poles dominate the response? one damper.
Advanced Problems 153

ADVANCED PROBLEMS

AP2.1 An armature-controlled DC motor is driving a load.


The input voltage is 5 V. The speed at ( = 2 seconds is
30 rad/s, and the steady speed is 70 rad/s when t*oo.
Determine the transfer function <o(s)/V(s).
I
GdU)
AP2.2 A system has a block diagram as shown in Figure
AP2.2. Determine the transfer function L
T(s) =
g(f) A'(.v) Gc(s)
-o- G(0 *- n-o

It is desired to decouple Y(s) from R\(s) by obtaining H(s)


T(s) = 0. Select C 5 ( J ) in terms of the other Gj(s) to
achieve decoupling.
FIGURE AP2.3 Feedback system with a disturbance
input.
Hi(s)
heat flow of the heating element. The system parame-
ters are C ?, S, and Rr The thermal heating system is
illustrated in Table 2.5. (a) Determine the response of
R,(s) K_}~*" G'W G,W /,(.0
the system to a unit step q(s) = 1/s. (b) As t*oo.
what value does the step response determined in part
(a) approach? This is known as the steady-state re-
C5(.v)
sponse, (c) Describe how you would select the system
parameters C Q, 5, and R, to increase the speed of
JV GeC) response of the system to a step input.
AP2.5 For the three-cart system illustrated in Figure
AP2.5, obtain the equations of motion.The system has
-.(.0 ^o .,+
GiW K,(.0
three inputs j, 1. and u3 and three outputs JC-,, JC2-
and v3. Obtain three second-order ordinary differen-
tial equations with constant coefficients. If possible,
write the equations of motion in matrix form.
sy*) 4
:
l ". *i >-.v. .V,
FIGURE AP2.2 Interacting control system.
' - *

AP2.3 Consider the feedback control system in Figure M, M2 *3


M3
AP2.3. Define the tracking error as . VvW- VvVA V\AAA
E(t) = R(s) - Y(s).
h ()() h OO h ()()
(a) Determine a suitable H(s) such that the tracking
error is zero for any input R(s) in the absence of a FIGURE AP2.5 Three-cart system with three inputs and
disturbance input (that is, when Tlt(s) = 0). (b) Using three outputs.
H{s) determined in part (a), determine the response
Y(s) for a disturbance T,j(s) when the input R(s) = 0. AP2.6 Consider the hanging crane structure in Figure
(c) Is it possible to obtain Y(s) = 0 for an arbitrary AP2.6. Write the equations of motion describing the
disturbance T^(s) when G,i(s) > 0? Explain your motion of the cart and the payload. The mass of the
answer. cart is M, the mass of the payload is m, the massless
AP2.4 Consider a thermal heating system given by rigid connector has length L, and the friction is mod-
eled as Ft, = b'x where x is the distance traveled by
g(') _ 1 the cart.
q(s) C,s + (QS + l/R.Y AP2.7 Consider the unity feedback system described in the
where the output 3"(.?) is the temperature difference block diagram in Figure AP2.7. Compute analytically
due to the thermal process, the input q(s) is the rate of the response of the system to an impulse disturbance.
154 Chapter 2 Mathematical Models of Systems

FIGURE AP2.6
(a) Hanging crane
supporting the
Space Shuttle
Atlantis (Image
Credit: NASA/Jack
Pfaller) and
(b) schematic
representation
of the hanging
crane structure. (a) (b)

Td(s)

Controller Plant
FIGURE AP2.7
Unity feedback
control system with
, EM
K
+
o
+(
1
s + 20
+ Y(s)

controller
Gc(s) - K.

Determine a relationship between the gain K and AP2.9 Consider the inverting operational amplifier in
the minimum time it takes the impulse disturbance Figure AP2.9. Find the transfer function VJ,s){Vls),
response of the system to reach y(r) < 0.1. Assume Show that the transfer function can be expressed as
that K > 0. For what value of K does the disturbance
response first reach at y{t) = 0.1 at r = 0.05? G(s) = K, + + K&,
AP2.8 Consider the cable reel control system given in V,(s)
Figure AP2.8. Find the value of A and K such that the where the gains KP, Kh and KD are functions of
percent overshoot is P.O. 10% and a desired ve- Cj, C2, JRI, and R2, This circuit is a proportional-inte-
locity of 50 m/s in the steady state is achieved. Com- gral-derivative (PID) controller (more on PID con-
pute the closed-loop response v(f) analytically and trollers in Chapter 7).
confirm that the steady-state response and P. O. meet
the specifications.

Reel
Amplifier Motor dynamics Actual cable
Desired
, velocity + 200 Torque 1 velocity
/?(,)= - ^ K
>
K V{s)
J * s+ 1 j+8

Measured Tachometer
FIGURE AP2.8 velocity 1
Cable reel control
0.25i + 1
system.
Design Problems 155
c-,
R2
l-WV Hf
v,w<
- K,M

FIGURE AP2.9 An inverting operational amplifier circuit


representing a PID controller.

DESIGN PROBLEMS
CDP2.1 We want to accurately position a table for a ma- Table CDP2.1 Typical Parameters
chine as shown in Figure CDP2.1. A traction-drive
motor with a capstan roller possesses several desirable for the Armature-Controlled DC Motor
characteristics compared to the more popular ball and the Capstan and Slide
screw. The traction drive exhibits low friction and no
Ms Mass of slide 5.693 kg
backlash. However, it is susceptible to disturbances. De-
velop a model of the traction drive shown in Figure M,, Mass of drive bar 6.96 kg
CDP2.1(a) for the parameters given in Table CDP2.1. 'm Inertia of 10.91 lfT3 kg m2
The drive uses a DC armature-controlled motor with a roller, shaft, motor
capstan roller attached to the shaft.The drive bar moves and tachometer
the linear slide-table. The slide uses an air bearing, so its
friction is negligible. We are considering the open-loop r Roller radius 31.75-10 -
model, Figure CDP2.1(b), and its transfer function in bm Motor damping 0.268 N ms/rad
this problem. Feedback will be introduced later. Kn Torque constant 0.8379 N m/amp
K Back emf constant 0.838 Vs/rad
Traction drive motor Rm Motor resistance 1.36 Q,
and capstan roller L'm Motor inductance 3.6 mH

the closed-loop transfer function Y(s)!R($) is exactly


equal to 1.
DP2.2 The television beam circuit of a television is repre-
sented by the model in Figure DP2.2. Select the un-
Linear slide known conductance G so that the voltage v is 24 V.
Each conductance is given in Siemens (S).
(a)
DP2.3 An input r(t) = t, t a 0, is applied to a black box
with a transfer function G(s). The resulting output
response, when the initial conditions are zero, is
V(-v) G(s) X(s)
y(0 = e - ' - ^ - 2 ' - ^ + | / , / ^ 0 .
(b)

FIGURE CDP2.1 (a) Traction drive, capstan roller, and Determine G(s) for this system.
linear slide, (b) The block diagram model. DP2.4 An operational amplifier circuit that can serve as
a filter circuit is shown in Figure DP2.4. Determine
DP2.1 A control system is shown in Figure DP2.1. The the transfer function of the circuit, assuming an ideal
transfer functions G2(s) and H2(s) are fixed. Deter- op-amp. Find vt)(t) when the input is Uj(f) = At,
mine the transfer functions G{(s) and //](.?) so that t >0.
156 Chapter 2 Mathematical Models of Systems

W.v) -* Y(s)

FIGURE DP2.1
Selection of transfer ff.
functions.

Reference

JT A/W
1 a
<t>M. 20A f:
R2 b v> -o +

"<
"1 C
FIGURE DP2.2 Television beam circuit.

DP2.5 Consider the clock shown in Figure DP2.5. The


pendulum rod of length L supports a pendulum disk.
Assume that the pendulum rod is a massless rigid thin
rod and the pendulum disc has mass m. Design the FIGURE DP2.4 Operational amplifier circuit.
length of the pendulum, L, so that the period of mo-
tion is 2 seconds. Note that with a period of 2 seconds
each "tick" and each "tock" of the clock represents 1 analysis so that sin <p w (p. Can you explain why most
second, as desired. Assume small angles, <p, in the grandfather clocks are about 1.5 m or taller?

Pendulum rod

Pendulum disk ^.
FIGURE DP2.5
(a) Typical clock
(photo courtesy
of SuperStock)
and (b) schematic
representation
of the pendulum.
Computer Problems 157
JM COMPUTER PROBLEMS
CP2.1 Consider the two polynomials
/?(X = s.21 + 7s + 10
and Spring
Forcing
< * constant
q(s) = s + 2.
Compute the following
(a) p(s)q(s)
function

.I'
Mass Mass
in displacement
(b) poles and zeros of G(s) =
(C) /7(-1)
Pis)

CP2.2 Consider the feedback system depicted in Figure


Friction
constant
J, y(?)

CP2.2. b
(a) Compute the closed-loop transfer function using
the series and feedback functions.
FIGURE CP2.4 A mechanical spring-mass-damper
(b) Obtain the closed-loop system unit step response system.
with the step function, and verify that final value
of the output is 2/5.
unit step input. Let in = 10, k = 1, and b = 0.5. Show
that the peak amplitude of the output is about 1.8.
Controller Plant
CP2.5 A satellite single-axis attitude control system can
s +2 be represented by the block diagram in Figure CP2.5.
A'f v i
,v+ 1 j + 3 m The variables k, a, and b are controller parameters,
and J is the spacecraft moment of inertia. Suppose the
nominal moment of inertia is J = 10.8E8 (slug ft2),
and the controller parameters are k = 10.8E8, a = 1,
and 6 = 8.
FIGURE CP2.2 A negative feedback control system. (a) Develop an m-file script to compute the closed-
loop transfer function T(s) = 0(s)/0,i(s).
(b) Compute and plot the step response to a 10 step
CP2.3 Consider the differential equation input.
y + 4y + 3y = u, (c) The exact moment of inertia is generally unknown
and may change slowly with time. Compare the
where y(0) = y(0) = 0 and u(t) is a unit step. Deter- step response performance of the spacecraft when
mine the solution y(t) analytically and verify by co- / i s reduced by 20% and 50%. Use the controller
plotting the analytic solution and the step response parameters k = 10.8E8, a = 1, and b = 8 and a
obtained with the step function. 10 step input. Discuss your results.
CP2.4 Consider the mechanical system depicted in CP2.6 Consider the block diagram in Figure CP2.6.
Figure CP2.4.The input is given by/(i). and the output (a) Use an m-file to reduce the block diagram in
is y(t). Determine the transfer function from f(t) to Figure CP2.6, and compute the closed-loop trans-
y(t) and, using an m-file, plot the system response to a fer function.

Controller Spacecraft
0dU) 0(1)
Desired "> , k(s + a) 1
s+b J*2
attitude
altitude

FIGURE CP2.5 A spacecraft single-axis attitude control block diagram.


158 Chapter 2 Mathematical Models of Systems

+
Ris)
r~\ 1 s
\r * (s)
~i\. s+ 1 s2 + 2 s2
+

4s+ 2
50
s2 + 2s + 1

s2 + 2
,93 + 14

FIGURE CP2.6 A multiple-loop feedback control system block diagram.

(b) Generate a pole-zero map of the closed-loop CP2.8 A system has a transfer function
transfer function in graphical form using the
pzmap function. X(s) (20/z)(s + z)
(c) Determine explicitly the poles and zeros of the B{s) ~ s2 + 3s + 20"
closed-loop transfer function using the pole and
zero functions and correlate the results with the Plot the response of the system when R(s) is a unit
pole-zero map in part (b). step for the parameter z = 5,10, and 15.
CP2.7 For the simple pendulum shown in Figure CP2.7, CP2.9 Consider the feedback control system in Figure
the nonlinear equation of motion is given by CP2.9, where

0(0 + sin 6 0, G(s) = ^ ^ and H(s)


s +2 s+ 1
where L = 0.5 m, m = 1 kg, and g = 9.8 m/s~. When (a) Using an m-file, determine the closed-loop trans-
the nonlinear equation is linearized about the equi- fer function.
librium point 6 = 0, we obtain the linear time-invariant (b) Obtain the pole-zero map using the pzmap func-
model, tion. Where are the closed-loop system poles and
zeros?
0 + j6 0. (c) Are there any pole-zero cancellations? If so, use
the minreal function to cancel common poles and
zeros in the closed-loop transfer function.
Create an m-file to plot both the nonlinear and the lin- (d) Why is it important to cancel common poles and
ear response of the simple pendulum when the initial zeros in the transfer function?
angle of the pendulum is 0(0) = 30 and explain any
differences.

lite) G(s) * Y(s)


^ ^ ^ ^ . ^ ^ ^

H(s) 4

FIGURE CP2.9 Control system with nonunity feedback.

CP2.10 Consider the block diagram in Figure CP2.10.


Create an m-file to complete the following tasks:
(a) Compute the step response of the closed-loop
system (that is, R(s) = Vs and 7",,(i) = 0) and
plot the steady-state value of the output Y(s) as a
function of the controller gain 0 < K s 10.
(b) Compute the disturbance step response of the
FIGURE CP2.7 Simple pendulum. closed-loop system (that is, R(s) = 0 and
Terms and Concepts 159
Td(s) lis) and co-plot the steady-state value of (c) Determine the value of K such that the steady-
the output Y (s) as a function of the controller gain state value of the output is equal for both the
0 < K < 10 on the same plot as in (a) above. input response and the disturbance response.

''
FIGURE CP2.10
Block diagram of
a unity feedback R(s) *> Y(s)
system with a
reference input R[s)
and a disturbance
input Td(s).

m ANSWERS TO SKILLS CHECK


True or False: (1) False; (2) True; (3) False; (4) True;
(5) True
Multiple Choice: (6) b; (7) a; (8) b; (9) b; (10) c;
Word Match (in order, top to bottom): e, j , d, h, a, f,
c, b, k, g, o, 1, n, m, i

(11) a; (12) a; (13) c; (14) a; (15) a

TERMS AND CONCEPTS


Across-Variable A variable determined by measuring the closed or otherwise accounted for. Generally obtained
difference of the values at the two ends of an element. by block diagram or signal-flow graph reduction.
Actuator The device that causes the process to provide Coulomb damper A type of mechanical damper where the
the output. The device that provides the motive power model of the friction force is a nonlinear function of
to the process. the mass velocity and possesses a discontinuity around
zero velocity. Also know as dry friction.
Analogous variables Variables associated with electrical,
mechanical, thermal, and fluid systems possessing Critical damping The case where damping is on the
similar solutions providing the analyst with the ability boundary between underdamped and overdamped.
to extend the solution of one system to all analogous
systems with the same describing differential equations. Damped oscillation An oscillation in which the ampli-
tude decreases with time.
Assumptions Statements that reflect situations and con-
ditions that are taken for granted and without proof. Damping ratio A measure of damping. A dimensionless
In control systems, assumptions are often employed to number for the second-order characteristic equation.
simplify the physical dynamical models of systems DC motor An electric actuator that uses an input voltage
under consideration to make the control design as a control variable.
problem more tractable.
Differential equation An equation including differentials
Block diagrams Unidirectional, operational blocks that of a function.
represent the transfer functions of the elements of the
system. Error signal The difference between the desired out-
put R{s) and the actual output Y(s); therefore
Branch A unidirectional path segment in a signal-flow
E{s) = R(s) - Y(s).
graph that relates the dependency of an input and an
output variable. Final value The value that the output achieves after all
Characteristic equation The relation formed by equating the transient constituents of the response have faded.
to zero the denominator of a transfer function. Also referred to as the steady-state value.

Closed-loop transfer function A ratio of the output signal Final value theorem The theorem that states that
to the input signal for an interconnection of systems lim y(t) = lim .sY(.y), where Y(s) is the Laplace
when all the feedback or feedfoward loops have been transform of y(t).
160 Chapter 2 Mathematical Models of Systems

Homogeneity The property of a linear system in which Principle of superposition The law that states that if two
the system response, y(t), to an input (0 leads to the inputs are scaled and summed and routed through a
response /3y(f) when the input is (3u(t). linear, time-invariant system, then the output will be
identical to the sum of outputs due to the individual
Inverse Laplace transform A transformation of a function scaled inputs when routed through the same system.
F(s) from the complex frequency domain into the
time domain yielding f(t). Reference input The input to a control system often
representing the desired output, denoted by R(s).
Laplace transform A transformation of a function /(t)
from the time domain into the complex frequency Residues The constants k\ associated with the partial
domain yielding F(s). fraction expansion of the output Y(s), when the out-
put is written in a residue-pole format.
Linear approximation An approximate model that re-
sults in a linear relationship between the output and Signal-flow graph A diagram that consists of nodes con-
the input of the device. nected by several directed branches and that is a
graphical representation of a set of linear relations.
Linear system A system that satisfies the properties of
Simulation A model of a system that is used to investigate
superposition and homogeneity.
the behavior of a system by utilizing actual input signals.
Linearized Made linear or placed in a linear form. Taylor
Steady state The value that the output achieves after all
series approximations are commonly employed to
the transient constituents of the response have faded.
obtain linear models of physical systems.
Also referred to as the final value.
Loop A closed path that originates and terminates on the 5-plane The complex plane where, given the complex
same node of a signal-flow graph with no node being number s = s + jw, the x-axis (or horizontal axis) is
met twice along the path. the s-axis, and the y-axis (or vertical axis) is the ;w-axis.
Mason loop rule A rule that enables the user to obtain
a transfer function by tracing paths and loops with- Taylor series A power series defined by g(x) =
in a system.
^ :(x x0)m. For m < 00, the series is an
Mathematical models Descriptions of the behavior of a
system using mathematics. approximation which is used to linearize functions
Natural frequency The frequency of natural oscillation and system models.
that would occur for two complex poles if the damp- Through-variable A variable that has the same value at
ing were equal to zero. both ends of an element.
Necessary condition A condition or statement that must
Time constant The time interval necessary for a system to
be satisfied to achieve a desired effect or result. For ex-
change from one state to another by a specified per-
ample, for a linear system it is necessary that the input
centage. For a first order system, the time constant is
m(t) + M 2 (0 results in the response y^t) + frit),
the time it takes the output to manifest a 63.2%
where the input u^(t) results in the response y^t) and
change due to a step input.
the input 112(() results in the response yi(t).
Transfer function The ratio of the Laplace transform of
Node The input and output points or junctions in a
the output variable to the Laplace transform of the
signal-flow graph. input variable.
Nontouching Two loops in a signal-flow graph that do not
Underdamped The case where the damping ratio is f < 1.
have a common node.
Unity feedback A feedback control system wherein the
Overdamped The case where the damping ratio is > 1.
gain of the feedback loop is one.
Path A branch or a continuous sequence of branches
Viscous damper A type of mechanical damper where the
that can be traversed from one signal (node) to
model of the friction force is linearly proportional to
another signal (node) in a signal-flow graph.
the velocity of the mass.
Poles The roots of the denominator polynomial (i.e.,
Zeros The roots of the numerator polynomial of the
the roots of the characteristic equation) of the trans-
transfer function.
fer function.
Positive feedback loop Feedback loop wherein the output
signal is fed back so that it adds to the input signal.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

C H A P T E R 2

Mathematical Models of Systems

Exercises
E2.1 We have for the open-loop

y = r2

and for the closed-loop

e = r y and y = e2 .

So, e = r e2 and e2 + e r = 0 .

16

14

12

10

8
y

6 open-loop

2 closed-loop

0
0 0.5 1 1.5 2 2.5 3 3.5 4
r

FIGURE E2.1
Plot of open-loop versus closed-loop.

For example, if r = 1, then e2 + e 1 = 0 implies that e = 0.618. Thus,


y = 0.382. A plot y versus r is shown in Figure E2.1.

22
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Exercises 23

E2.2 Define

f (T ) = R = R0 e0.1T

and

R = f (T ) f (T0 ) , T = T T0 .

Then,
f

R = f (T ) f (T0 ) = T +
T T =T0 =20

where
f

= 0.1R0 e0.1T0 = 135,
T T =T0 =20

when R0 = 10, 000. Thus, the linear approximation is computed by


considering only the first-order terms in the Taylor series expansion, and
is given by

R = 135T .

E2.3 The spring constant for the equilibrium point is found graphically by
estimating the slope of a line tangent to the force versus displacement
curve at the point y = 0.5cm, see Figure E2.3. The slope of the line is
K 1.
2

1.5

1 Spring breaks

0.5

0
Force (n)

-0.5

-1

-1.5

-2

-2.5 Spring compresses


-3
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3

y=Displacement (cm)

FIGURE E2.3
Spring force as a function of displacement.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

24 CHAPTER 2 Mathematical Models of Systems

E2.4 Since
1
R(s) =
s
we have
4(s + 50)
Y (s) = .
s(s + 20)(s + 10)

The partial fraction expansion of Y (s) is given by

A1 A2 A3
Y (s) = + +
s s + 20 s + 10
where

A1 = 1 , A2 = 0.6 and A3 = 1.6 .

Using the Laplace transform table, we find that

y(t) = 1 + 0.6e20t 1.6e10t .

The final value is computed using the final value theorem:

4(s + 50)
 
lim y(t) = lim s 2
=1.
t s0 s(s + 30s + 200)

E2.5 The circuit diagram is shown in Figure E2.5.

R2

v-
-
A R1
+ +
+
vin v0
- -

FIGURE E2.5
Noninverting op-amp circuit.

With an ideal op-amp, we have

vo = A(vin v ),
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Exercises 25

where A is very large. We have the relationship

R1
v = vo .
R1 + R2
Therefore,

R1
vo = A(vin vo ),
R1 + R2
and solving for vo yields

A
vo = AR1
vin .
1+ R1 +R2

Since A 1, it follows that 1 + RAR 1


1 +R2
AR1
R1 +R2 . Then the expression for
vo simplifies to

R1 + R2
vo = vin .
R1
E2.6 Given

y = f (x) = ex

and the operating point xo = 1, we have the linear approximation

f

y = f (x) = f (xo ) + (x xo ) +
x x=xo

where
df

f (xo ) = e, = e, and x xo = x 1.
dx x=xo =1

Therefore, we obtain the linear approximation y = ex.


E2.7 The block diagram is shown in Figure E2.7.

+ Ea(s)
R(s) G1(s) G2(s) I(s)
-

H(s)

FIGURE E2.7
Block diagram model.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

26 CHAPTER 2 Mathematical Models of Systems

Starting at the output we obtain

I(s) = G1 (s)G2 (s)E(s).

But E(s) = R(s) H(s)I(s), so

I(s) = G1 (s)G2 (s) [R(s) H(s)I(s)] .

Solving for I(s) yields the closed-loop transfer function


I(s) G1 (s)G2 (s)
= .
R(s) 1 + G1 (s)G2 (s)H(s)
E2.8 The block diagram is shown in Figure E2.8.

H2(s)

- A(s) Z(s) 1
R(s) K G1(s) G2(s) Y(s)
- W(s) - s
- E(s)

H3(s)

H1(s)

FIGURE E2.8
Block diagram model.

Starting at the output we obtain


1 1
Y (s) = Z(s) = G2 (s)A(s).
s s
But A(s) = G1 (s) [H2 (s)Z(s) H3 (s)A(s) + W (s)] and Z(s) = sY (s),
so
1
Y (s) = G1 (s)G2 (s)H2 (s)Y (s) G1 (s)H3 (s)Y (s) + G1 (s)G2 (s)W (s).
s
Substituting W (s) = KE(s) H1 (s)Z(s) into the above equation yields

Y (s) = G1 (s)G2 (s)H2 (s)Y (s) G1 (s)H3 (s)Y (s)


1
+ G1 (s)G2 (s) [KE(s) H1 (s)Z(s)]
s
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Exercises 27

and with E(s) = R(s) Y (s) and Z(s) = sY (s) this reduces to

Y (s) = [G1 (s)G2 (s) (H2 (s) + H1 (s)) G1 (s)H3 (s)


1 1
G1 (s)G2 (s)K]Y (s) + G1 (s)G2 (s)KR(s).
s s
Solving for Y (s) yields the transfer function

Y (s) = T (s)R(s),

where
KG1 (s)G2 (s)/s
T (s) = .
1 + G1 (s)G2 (s) [(H2 (s) + H1 (s)] + G1 (s)H3 (s) + KG1 (s)G2 (s)/s
E2.9 From Figure E2.9, we observe that

Ff (s) = G2 (s)U (s)

and

FR (s) = G3 (s)U (s) .

Then, solving for U (s) yields


1
U (s) = Ff (s)
G2 (s)
and it follows that
G3 (s)
FR (s) = U (s) .
G2 (s)
Again, considering the block diagram in Figure E2.9 we determine

Ff (s) = G1 (s)G2 (s)[R(s) H2 (s)Ff (s) H2 (s)FR (s)] .

But, from the previous result, we substitute for FR (s) resulting in

Ff (s) = G1 (s)G2 (s)R(s)G1 (s)G2 (s)H2 (s)Ff (s)G1 (s)H2 (s)G3 (s)Ff (s) .

Solving for Ff (s) yields


G1 (s)G2 (s)
 
Ff (s) = R(s) .
1 + G1 (s)G2 (s)H2 (s) + G1 (s)G3 (s)H2 (s)
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

28 CHAPTER 2 Mathematical Models of Systems

H2(s)

U(s) G2(s) Ff (s)


+ -
R(s) G1(s)
-
U(s) G3(s) FR(s)

H2(s)

FIGURE E2.9
Block diagram model.

E2.10 The shock absorber block diagram is shown in Figure E2.10. The closed-
loop transfer function model is
Gc (s)Gp (s)G(s)
T (s) = .
1 + H(s)Gc (s)Gp (s)G(s)

Plunger and
Controller Gear Motor
Piston System
+
R(s) Gc(s) Gp(s) G(s) Y(s)
Desired piston - Piston
travel travel

Sensor

H(s)
Piston travel
measurement

FIGURE E2.10
Shock absorber block diagram.

E2.11 Let f denote the spring force (n) and x denote the deflection (m). Then
f
K= .
x
Computing the slope from the graph yields:
(a) xo = 0.14m K = f /x = 10 n / 0.04 m = 250 n/m
(b) xo = 0m K = f /x = 10 n / 0.05 m = 200 n/m
(c) xo = 0.35m K = f /x = 3n / 0.05 m = 60 n/m
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Exercises 29

E2.12 The signal flow graph is shown in Fig. E2.12. Find Y (s) when R(s) = 0.

Td(s)
-K
1

1
K2 G(s)
Y (s)

-1

FIGURE E2.12
Signal flow graph.

The transfer function from Td (s) to Y (s) is

G(s)Td (s) K1 K2 G(s)Td (s) G(s)(1 K1 K2 )Td (s)


Y (s) = = .
1 (K2 G(s)) 1 + K2 G(s)
If we set

K1 K2 = 1 ,

then Y (s) = 0 for any Td (s).


E2.13 The transfer function from R(s), Td (s), and N (s) to Y (s) is
K 1 K
     
Y (s) = 2 R(s)+ 2 Td (s) 2 N (s)
s + 10s + K s + 10s + K s + 10s + K
Therefore, we find that
1 K
Y (s)/Td (s) = and Y (s)/N (s) =
s2 + 10s + K s2 + 10s + K
E2.14 Since we want to compute the transfer function from R2 (s) to Y1 (s), we
can assume that R1 = 0 (application of the principle of superposition).
Then, starting at the output Y1 (s) we obtain

Y1 (s) = G3 (s) [H1 (s)Y1 (s) + G2 (s)G8 (s)W (s) + G9 (s)W (s)] ,

or

[1 + G3 (s)H1 (s)] Y1 (s) = [G3 (s)G2 (s)G8 (s)W (s) + G3 (s)G9 (s)] W (s).

Considering the signal W (s) (see Figure E2.14), we determine that

W (s) = G5 (s) [G4 (s)R2 (s) H2 (s)W (s)] ,


2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

30 CHAPTER 2 Mathematical Models of Systems

H1(s)

-
+ + G3(s)
R1(s) G1(s) G2(s) Y1(s)
+ +

G7(s) G8(s) G9(s)

+
+ G6(s)
R2(s) G4(s) G5(s) Y2(s)
- W(s)

H2(s)

FIGURE E2.14
Block diagram model.

or

[1 + G5 (s)H2 (s)] W (s) = G5 (s)G4 (s)R2 (s).

Substituting the expression for W (s) into the above equation for Y1 (s)
yields
Y1 (s) G2 (s)G3 (s)G4 (s)G5 (s)G8 (s) + G3 (s)G4 (s)G5 (s)G9 (s)
= .
R2 (s) 1 + G3 (s)H1 (s) + G5 (s)H2 (s) + G3 (s)G5 (s)H1 (s)H2 (s)
E2.15 For loop 1, we have
di1 1
Z
R1 i1 + L1 + (i1 i2 )dt + R2 (i1 i2 ) = v(t) .
dt C1
And for loop 2, we have
1 di2 1
Z Z
i2 dt + L2 + R2 (i2 i1 ) + (i2 i1 )dt = 0 .
C2 dt C1
E2.16 The transfer function from R(s) to P (s) is
P (s) 4.2
= 3 2
.
R(s) s + 2s + 4s + 4.2
The block diagram is shown in Figure E2.16a. The corresponding signal
flow graph is shown in Figure E2.16b for
4.2
P (s)/R(s) = .
s3 + 2s2 + 4s + 4.2
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Exercises 31

v1(s) v2(s) q(s) 1


R(s) 0.6 P(s)
7
- s s2+2s+4

(a)
1
V1 V2 0.6
s s2 + 2 s + 4
1 7
R(s ) P (s)

-1

(b)
FIGURE E2.16
(a) Block diagram, (b) Signal flow graph.

E2.17 A linear approximation for f is given by


f

f = x = 2kxo x = kx
x x=xo

where xo = 1/2, f = f (x) f (xo ), and x = x xo .


E2.18 The linear approximation is given by

y = mx

where
y

m= .
x x=xo

(a) When xo = 1, we find that yo = 2.4, and yo = 13.2 when xo = 2.

(b) The slope m is computed as follows:


y

m= = 1 + 4.2x2o .
x x=xo

Therefore, m = 5.2 at xo = 1, and m = 18.8 at xo = 2.


2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

32 CHAPTER 2 Mathematical Models of Systems

E2.19 The output (with a step input) is


15(s + 1)
Y (s) = .
s(s + 7)(s + 2)
The partial fraction expansion is
15 18 1 3 1
Y (s) = + .
14s 7 s+7 2s+2
Taking the inverse Laplace transform yields
15 18 7t 3 2t
y(t) = e + e .
14 7 2
E2.20 The input-output relationship is
Vo A(K 1)
=
V 1 + AK
where
Z1
K= .
Z1 + Z2
Assume A 1. Then,
Vo K1 Z2
= =
V K Z1
where
R1 R2
Z1 = and Z2 = .
R1 C 1 s + 1 R2 C 2 s + 1
Therefore,
Vo (s) R2 (R1 C1 s + 1) 2(s + 1)
= = .
V (s) R1 (R2 C2 s + 1) s+2
E2.21 The equation of motion of the mass mc is

mc xp + (bd + bs )xp + kd xp = bd xin + kd xin .

Taking the Laplace transform with zero initial conditions yields

[mc s2 + (bd + bs )s + kd ]Xp (s) = [bd s + kd ]Xin (s) .

So, the transfer function is


Xp (s) bd s + kd 0.7s + 2
= = 2 .
Xin (s) mc s2 + (bd + bs )s + kd s + 2.8s + 2
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Exercises 33

E2.22 The rotational velocity is


2(s + 4) 1
(s) = 2
.
(s + 5)(s + 1) s
Expanding in a partial fraction expansion yields
81 1 1 3 1 13 1
(s) = + 2
.
5 s 40 s + 5 2 (s + 1) 8 s+1
Taking the inverse Laplace transform yields
8 1 3 13
(t) = + e5t tet et .
5 40 2 8
E2.23 The closed-loop transfer function is
Y (s) K1 K2
= T (s) = 2 .
R(s) s + (K1 + K2 K3 + K1 K2 )s + K1 K2 K3
E2.24 The closed-loop tranfser function is
Y (s) 10
= T (s) = 2 .
R(s) s + 21s + 10

E2.25 Let x = 0.6 and y = 0.8. Then, with y = ax3 , we have

0.8 = a(0.6)3 .

Solving for a yields a = 3.704. A linear approximation is

y yo = 3ax2o (x xo )

or y = 4x 1.6, where yo = 0.8 and xo = 0.6.


E2.26 The equations of motion are

m1 x1 + k(x1 x2 ) = F
m2 x2 + k(x2 x1 ) = 0 .

Taking the Laplace transform (with zero initial conditions) and solving
for X2 (s) yields
k
X2 (s) = F (s) .
(m2 s2 + k)(m1 s2 + k) k 2
Then, with m1 = m2 = k = 1, we have
1
X2 (s)/F (s) = .
s2 (s2 + 2)
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

34 CHAPTER 2 Mathematical Models of Systems

E2.27 The transfer function from Td (s) to Y (s) is

G2 (s)
Y (s)/Td (s) = .
1 + G1 G2 H(s)

E2.28 The transfer function is


Vo (s) R2 R4 C R2 R4
= s+ = 24s + 144 .
V (s) R3 R1 R3

E2.29 (a) If

1
G(s) = and H(s) = 2s + 15 ,
s2 + 15s + 50

then the closed-loop transfer function of Figure E2.28(a) and (b) (in
Dorf & Bishop) are equivalent.
(b) The closed-loop transfer function is

1
T (s) = .
s2 + 17s + 65

E2.30 (a) The closed-loop transfer function is

G(s) 1 10 10
T (s) = = 2
where G(s) = .
1 + G(s) s s(s + 2s + 20) s2 + 2s + 10

0.8

0.7

0.6

0.5
Amplitude

0.4

0.3

0.2

0.1

0
0 1 2 3 4 5 6
Time sec

FIGURE E2.30
Step response.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Exercises 35

(b) The output Y (s) (when R(s) = 1/s) is


0.5 0.25 + 0.0573j 0.25 0.0573j
Y (s) = + .
s s + 1 4.3589j s + 1 + 4.3589j
(c) The plot of y(t) is shown in Figure E2.30. The output is given by
1
 
1 
y(t) = 1 et cos 19t sin 19t
2 19
E2.31 The partial fraction expansion is
a b
V (s) = +
s + p1 s + p2
where p1 = 4 19.6j and p2 = 4 + 19.6j. Then, the residues are

a = 10.2j b = 10.2j .

The inverse Laplace transform is

v(t) = 10.2je(4+19.6j)t + 10.2je(419.6j)t = 20.4e4t sin 19.6t .


2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

36 CHAPTER 2 Mathematical Models of Systems

Problems
P2.1 The integrodifferential equations, obtained by Kirchoffs voltage law to
each loop, are as follows:
1 d(i1 i2 )
Z
R1 i1 + i1 dt + L1 + R2 (i1 i2 ) = v(t) (loop 1)
C1 dt
and
1 d(i2 i1 )
Z
R3 i2 + i2 dt + R2 (i2 i1 ) + L1 =0 (loop 2) .
C2 dt
P2.2 The differential equations describing the system can be obtained by using
a free-body diagram analysis of each mass. For mass 1 and 2 we have

M1 y1 + k12 (y1 y2 ) + by1 + k1 y1 = F (t)


M2 y2 + k12 (y2 y1 ) = 0 .

Using a force-current analogy, the analagous electric circuit is shown in


Figure P2.2, where Ci Mi , L1 1/k1 , L12 1/k12 , and R 1/b .

FIGURE P2.2
Analagous electric circuit.

P2.3 The differential equations describing the system can be obtained by using
a free-body diagram analysis of each mass. For mass 1 and 2 we have

M x1 + kx1 + k(x1 x2 ) = F (t)


M x2 + k(x2 x1 ) + bx2 = 0 .

Using a force-current analogy, the analagous electric circuit is shown in


Figure P2.3, where

CM L 1/k R 1/b .
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 37

FIGURE P2.3
Analagous electric circuit.

P2.4 (a) The linear approximation around vin = 0 is vo = 0vin , see Fig-
ure P2.4(a).
(b) The linear approximation around vin = 1 is vo = 2vin 1, see Fig-
ure P2.4(b).

(a) (b)
0.4 4

3.5
0.3

3
0.2
2.5

0.1
2
vo

vo

0 1.5
linear approximation

1
-0.1

0.5
-0.2
0

-0.3
-0.5 linear approximation

-0.4 -1
-1 -0.5 0 0.5 1 -1 0 1 2
vin vin

FIGURE P2.4
Nonlinear functions and approximations.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

38 CHAPTER 2 Mathematical Models of Systems

P2.5 Given

Q = K(P1 P2 )1/2 .

Let P = P1 P2 and Po = operating point. Using a Taylor series


expansion of Q, we have
Q

Q = Qo + (P Po ) +
P P =Po
where
Q K

Qo = KPo1/2 and = Po1/2 .
P P =Po
2
Define Q = Q Qo and P = P Po . Then, dropping higher-order
terms in the Taylor series expansion yields

Q = mP

where
K
m= 1/2
.
2Po
P2.6 From P2.1 we have
1 d(i1 i2 )
Z
R1 i1 + i1 dt + L1 + R2 (i1 i2 ) = v(t)
C1 dt
and
1 d(i2 i1 )
Z
R3 i2 + i2 dt + R2 (i2 i1 ) + L1 =0.
C2 dt
Taking the Laplace transform and using the fact that the initial voltage
across C2 is 10v yields
1
[R1 + + L1 s + R2 ]I1 (s) + [R2 L1 s]I2 (s) = 0
C1 s
and
1 10
[R2 L1 s]I1 (s) + [L1 s + R3 + + R2 ]I2 (s) = .
C2 s s
Rewriting in matrix form we have

1
R1 + C1 s + L 1 s + R2 R2 L1 s I1 (s) 0
= .
1
R2 L1 s L 1 s + R3 + C2 s + R2 I2 (s) 10/s
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 39

Solving for I2 yields



1
I1 (s) 1 L1 s + R3 + C2 s + R2 R2 + L 1 s 0
= .
I2 (s) R2 + L 1 s 1
R1 + C1 s + L1 s + R2 10/s

or

10(R1 + 1/C1 s + L1 s + R2 )
I2 (s) =
s

where

1 1
= (R1 + + L1 s + R2 )(L1 s + R3 + + R2 ) (R2 + L1 s)2 .
C1 s C2 s

P2.7 Consider the differentiating op-amp circuit in Figure P2.7. For an ideal
op-amp, the voltage gain (as a function of frequency) is

Z2 (s)
V2 (s) = V1 (s),
Z1 (s)

where

R1
Z1 =
1 + R1 Cs

and Z2 = R2 are the respective circuit impedances. Therefore, we obtain

R2 (1 + R1 Cs)
 
V2 (s) = V1 (s).
R1

Z
1 Z
C
2
R2

+ -
R1 +
+
V1(s) V2(s)

- -

FIGURE P2.7
Differentiating op-amp circuit.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

40 CHAPTER 2 Mathematical Models of Systems

P2.8 Let


G2 + Cs Cs G2


= Cs G1 + 2Cs Cs .

G2 Cs Cs + G2

Then,

ij V3 13 I1 /
Vj = I1 or or = .
V1 11 I1 /

Therefore, the transfer function is



Cs 2Cs + G1




V3 13 G2 Cs
T (s) = = =
V1 11

2Cs + G1 Cs




Cs Cs + G2

Pole-zero map (x:poles and o:zeros)


3

2 o

1
Imag Axis

0 x x

-1

-2 o

-3
-8 -7 -6 -5 -4 -3 -2 -1 0

Real Axis

FIGURE P2.8
Pole-zero map.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 41

C 2 R1 R2 s2 + 2CR1 s + 1
= .
C 2 R1 R2 s2 + (2R1 + R2 )Cs + 1

Using R1 = 0.5, R2 = 1, and C = 0.5, we have

s2 + 4s + 8 (s + 2 + 2j)(s + 2 2j)
T (s) = 2
= .
s + 8s + 8 (s + 4 + 8)(s + 4 8)

The pole-zero map is shown in Figure P2.8.


P2.9 From P2.3 we have

M x1 + kx1 + k(x1 x2 ) = F (t)


M x2 + k(x2 x1 ) + bx2 = 0 .

Taking the Laplace transform of both equations and writing the result in
matrix form, it follows that

M s2 + 2k k X1 (s) F (s)
= ,
k M s2 + bs + k X2 (s) 0

Pole zero map


0.4

0.3

0.2

0.1
Imag Axis

- 0.1

-0.2

-0.3

-0.4
-0.03 -0.025 -0.02 -0.015 -0.01 -0.005 0
Real Axis

FIGURE P2.9
Pole-zero map.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

42 CHAPTER 2 Mathematical Models of Systems

or


X1 (s)
=
1 M s2 + bs + k k

F (s)

X2 (s) k 2
M s + 2k 0

where = (M s2 + bs + k)(M s2 + 2k) k 2 . So,


X1 (s) M s2 + bs + k
G(s) = = .
F (s)

When b/k = 1, M = 1 , b2 /M k = 0.04, we have


s2 + 0.04s + 0.04
G(s) = .
s4 + 0.04s3 + 0.12s2 + 0.0032s + 0.0016
The pole-zero map is shown in Figure P2.9.
P2.10 From P2.2 we have

M1 y1 + k12 (y1 y2 ) + by1 + k1 y1 = F (t)


M2 y2 + k12 (y2 y1 ) = 0 .

Taking the Laplace transform of both equations and writing the result in
matrix form, it follows that

M1 s2 + bs + k1 + k12 k12 Y1 (s) F (s)
=
k12 M2 s2 + k12 Y2 (s) 0
or


Y1 (s)
=
1 M2 s2 + k12 k12

F (s)

Y2 (s) k12 M1 s2 + bs + k1 + k12 0

where

= (M2 s2 + k12 )(M1 s2 + bs + k1 + k12 ) k12


2
.

So, when f (t) = a sin o t, we have that Y1 (s) is given by


aM2 o (s2 + k12 /M2 )
Y1 (s) = .
(s2 + o2 )(s)
For motionless response (in the steady-state), set the zero of the transfer
function so that
k12 k12
(s2 + ) = s2 + o2 or o2 = .
M2 M2
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 43

P2.11 The transfer functions from Vc (s) to Vd (s) and from Vd (s) to (s) are:
K1 K2
Vd (s)/Vc (s) = , and
(Lq s + Rq )(Lc s + Rc )
Km
(s)/Vd (s) = 2
.
(Js + f s)((Ld + La )s + Rd + Ra ) + K3 Km s
The block diagram for (s)/Vc (s) is shown in Figure P2.11, where
(s) Vd (s) K1 K2 Km
(s)/Vc (s) = = ,
Vd (s) Vc (s) (s)
where

(s) = s(Lc s + Rc )(Lq s + Rq )((Js + b)((Ld + La )s + Rd + Ra ) + Km K3 ) .

Ic Vq Iq Vd + Id Tm w
Vc 1 K1 1 K2 1 1 1
L cs+R c L qs+R q (L d+L a)s+R d+R a Km q
Js+f s
-
Vb
K3

FIGURE P2.11
Block diagram.

P2.12 The open-loop transfer function is


Y (s) K
= .
R(s) s + 20
With R(s) = 1/s, we have
K
Y (s) = .
s(s + 20)
The partial fraction expansion is
K 1 1
 
Y (s) = ,
20 s s + 20
and the inverse Laplace transform is
K 
y(t) = 1 e20t ,
20
As t , it follows that y(t) K/20. So we choose K = 20 so that y(t)
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

44 CHAPTER 2 Mathematical Models of Systems

approaches 1. Alternatively we can use the final value theorem to obtain

K
y(t)t = lim sY (s) = =1.
s0 20
It follows that choosing K = 20 leads to y(t) 1 as t .
P2.13 The motor torque is given by

Tm (s) = (Jm s2 + bm s)m (s) + (JL s2 + bL s)nL (s)


= n((Jm s2 + bm s)/n2 + JL s2 + bL s)L (s)

where

n = L (s)/m (s) = gear ratio .

But

Tm (s) = Km Ig (s)

and
1
Ig (s) = Vg (s) ,
(Lg + Lf )s + Rg + Rf

and
Kg
Vg (s) = Kg If (s) = Vf (s) .
Rf + L f s

Combining the above expressions yields

L (s) Kg Km
= .
Vf (s) n1 (s)2 (s)

where
Jm s2 + bm s
1 (s) = JL s2 + bL s +
n2
and

2 (s) = (Lg s + Lf s + Rg + Rf )(Rf + Lf s) .

P2.14 For a field-controlled dc electric motor we have


Km /Rf
(s)/Vf (s) = .
Js + b
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 45

With a step input of Vf (s) = 80/s, the final value of (t) is

80Km Km
(t)t = lim s(s) = = 2.4 or = 0.03 .
s0 Rf b Rf b

Solving for (t) yields

80Km 1 1 80Km
 
(t) = L = (1e(b/J)t ) = 2.4(1e(b/J)t ) .
Rf J s(s + b/J) Rf b

At t = 1/2, (t) = 1, so

(1/2) = 2.4(1 e(b/J)t ) = 1 implies b/J = 1.08 sec .

Therefore,
0.0324
(s)/Vf (s) = .
s + 1.08
P2.15 Summing the forces in the vertical direction and using Newtons Second
Law we obtain
k
x + x=0.
m
The system has no damping and no external inputs. Taking the Laplace
transform yields
x0 s
X(s) = ,
s2 + k/m

where we used the fact that x(0) = x0 and x(0) = 0. Then taking the
inverse Laplace transform yields
s
k
x(t) = x0 cos t.
m

P2.16 Using Cramers rule, we have



1 1.5 x1 6
=
2 4 x2 11

or

x1 1 4 1.5 6
=
x2 2 1 11
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

46 CHAPTER 2 Mathematical Models of Systems

where = 4(1) 2(1.5) = 1 . Therefore,


4(6) 1.5(11) 2(6) + 1(11)
x1 = = 7.5 and x2 = = 1 .
1 1
The signal flow graph is shown in Figure P2.16.

11
1/4

1 -1/2
6 X2
X1
-1.5

FIGURE P2.16
Signal flow graph.

So,

6(1) 1.5( 11
4 ) 11( 41 ) + 1
2 (6)
x1 = 3 = 7.5 and x2 = 3 = 1 .
1 4 1 4

P2.17 (a) For mass 1 and 2, we have

M1 x1 + K1 (x1 x2 ) + b1 (x3 x1 ) = 0
M2 x2 + K2 (x2 x3 ) + b2 (x3 x2 ) + K1 (x2 x1 ) = 0 .

(b) Taking the Laplace transform yields

(M1 s2 + b1 s + K1 )X1 (s) K1 X2 (s) = b1 sX3 (s)


K1 X1 (s) + (M2 s2 + b2 s + K1 + K2 )X2 (s) = (b2 s + K2 )X3 (s) .

(c) Let

G1 (s) = K2 + b2 s
G2 (s) = 1/p(s)
G3 (s) = 1/q(s)
G4 (s) = sb1 ,

where

p(s) = s2 M2 + sf2 + K1 + K2

and

q(s) = s2 M1 + sf1 + K1 .
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 47

The signal flow graph is shown in Figure P2.17.

G4

G3
X3 X1
G1 G2 K1

K1

FIGURE P2.17
Signal flow graph.

(d) The transfer function from X3 (s) to X1 (s) is

X1 (s) K1 G1 (s)G2 (s)G3 (s) + G4 (s)G3 (s)


= .
X3 (s) 1 K12 G2 (s)G3 (s)

P2.18 The signal flow graph is shown in Figure P2.18.

I1 Va Ia
Z2 Y3 Z4
V1 V2
Y1

-Y 1 -Z 2 -Y 3

FIGURE P2.18
Signal flow graph.

The transfer function is


V2 (s) Y 1 Z2 Y 3 Z4
= .
V1 (s) 1 + Y 1 Z2 + Y 3 Z2 + Y 3 Z4 + Y 1 Z2 Z4 Y 3

P2.19 For a noninerting op-amp circuit, depicted in Figure P2.19a, the voltage
gain (as a function of frequency) is

Z1 (s) + Z2 (s)
Vo (s) = Vin (s),
Z1 (s)

where Z1 (s) and Z2 (s) are the impedances of the respective circuits. In
the case of the voltage follower circuit, shown in Figure P2.19b, we have
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

48 CHAPTER 2 Mathematical Models of Systems

Z2

Z1 - -
v0 v0
vin + vin +

(a) (b)

FIGURE P2.19
(a) Noninverting op-amp circuit. (b) Voltage follower circuit.

Z1 = (open circuit) and Z2 = 0. Therefore, the transfer function is

Vo (s) Z1
= = 1.
Vin (s) Z1

P2.20 (a) Assume Rg Rs and Rs R1 . Then Rs = R1 + R2 R2 , and

vgs = vin vo ,

where we neglect iin , since Rg Rs . At node S, we have

vo vo gm Rs
= gm vgs = gm (vin vo ) or = .
Rs vin 1 + gm Rs

(b) With gm Rs = 20, we have

vo 20
= = 0.95 .
vin 21

(c) The block diagram is shown in Figure P2.20.

vin(s) gmRs vo(s)


-

FIGURE P2.20
Block diagram model.

P2.21 From the geometry we find that

l1 l2 l2
z = k (x y) y .
l1 l1
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 49

The flow rate balance yields


dy pZ(s)
A = pz which implies Y (s) = .
dt As
By combining the above results it follows that
p l1 l2 l2
   
Y (s) = k (X(s) Y (s)) Y (s) .
As l1 l1
Therefore, the signal flow graph is shown in Figure P2.21. Using Masons

-1

k (l 1 - l 2)/l 1 DZ
p/As
X Y
1

-l 2 / l 1

FIGURE P2.21
Signal flow graph.

gain formula we find that the transfer function is given by


k(l1 l2 )p
Y (s) l1 As K1
= l2 p k(l1 l2 )p
= ,
X(s) 1+ s + K2 + K1
l1 As + l1 As

where
k(l1 l2 )p l2 p
K1 = p and K2 = .
l1 A l1 A
P2.22 (a) The equations of motion for the two masses are

L 2 L
 
M L 1 + M gL1 + k
2
(1 2 ) = f (t)
2 2
 2
L
M L2 2 + M gL2 + k (2 1 ) = 0 .
2

With 1 = 1 and 2 = 2 , we have


g k k f (t)
 
1 = + 1 + 2 +
L 4M 4M 2M L
k g k
 
2 = 1 + 2 .
4M L 4M
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

50 CHAPTER 2 Mathematical Models of Systems

F (t) - w1 q1
1/2ML 1/s 1/s

(a) b

w2 q2
1/s 1/s

Imag(s)
g k
+ j
L + 2M

g
X
+ j k
(b) L + 4M

g
X + j
L

Re(s)

FIGURE P2.22
(a) Block diagram. (b) Pole-zero map.

(b) Define a = g/L + k/4M and b = k/4M . Then

1 (s) 1 s2 + a
= .
F (s) 2M L (s2 + a)2 b2
(c) The block diagram and pole-zero map are shown in Figure P2.22.
P2.23 The input-output ratio, Vce /Vin , is found to be
Vce (R 1) + hie Rf
= .
Vin hre + hie (hoe + Rf )

P2.24 (a) The voltage gain is given by


vo RL 1 2 (R1 + R2 )
= .
vin (R1 + R2 )(Rg + hie1 ) + R1 (R1 + R2 )(1 + 1 ) + R1 RL 1 2
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 51

(b) The current gain is found to be


ic2
= 1 2 .
ib1
(c) The input impedance is

vin (R1 + R2 )(Rg + hie1 ) + R1 (R1 + R2 )(1 + 1 ) + R1 RL 1 2


= ,
ib1 R1 + R2
and when 1 2 is very large, we have the approximation
vin RL R1 1 2
.
ib1 R1 + R2
P2.25 The transfer function from R(s) and Td (s) to Y (s) is given by
1
 
Y (s) = G(s) R(s) (G(s)R(s) + Td (s)) + Td (s) + G(s)R(s)
G(s)
= G(s)R(s) .

Thus,

Y (s)/R(s) = G(s) .

Also, we have that

Y (s) = 0 .

when R(s) = 0. Therefore, the effect of the disturbance, Td (s), is elimi-


nated.
P2.26 The equations of motion for the two mass model of the robot are

M x + b(x y) + k(x y) = F (t)


my + b(y x) + k(y x) = 0 .

Taking the Laplace transform and writing the result in matrix form yields

M s2 + bs + k (bs + k) X(s) F (s)
= .
(bs + k) ms2 + bs + k Y (s) 0

Solving for Y (s) we find that


1
Y (s) mM (bs +k)
=  .
F (s) m b
s2 [s2 + 1 + M k
ms + m ]
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

52 CHAPTER 2 Mathematical Models of Systems

P2.27 The describing equation of motion is


i2
mz = mg k .
z2
Defining
ki2
f (z, i) = g
mz 2
leads to

z = f (z, i) .

The equilibrium condition for io and zo , found by solving the equation of


motion when

z = z = 0 ,

is
ki2o
= zo2 .
mg
We linearize the equation of motion using a Taylor series approximation.
With the definitions

z = z zo and i = i io ,
= z and z
we have z = z. Therefore,

= f (z, i) = f (zo , io ) + f z=z z + f z=z i +




z o
z i=io
i i=ioo
But f (zo , io ) = 0, and neglecting higher-order terms in the expansion
yields
2
= 2kio z 2kio i .
z
mzo3 mzo2
Using the equilibrium condition which relates zo to io , we determine that

= 2g z g i .
z
zo io
Taking the Laplace transform yields the transfer function (valid around
the equilibrium point)
Z(s) g/io
= 2 .
I(s) s 2g/zo
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 57

K1
R(s ) Y (s)
+ s (s+1)
-

1 +K 2s

0.9

0.8

0.7

0.6 <---- time to 90% = 0.39 sec


y(t)

0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

time(sec)

FIGURE P2.35
The equivalent block diagram and the system step response.

P2.36 (a) Given R(s) = 1/s2 , the partial fraction expansion is

24 3 8/3 3/4 1 13/12


Y (s) = = + + 2 .
s2 (s + 2)(s + 3)(s + 4) s+2 s+3 s+4 s s

Therefore, using the Laplace transform table, we determine that the


ramp response is
8 3 13
y(t) = 3e2t e3t + e4t + t , t0.
3 4 12
(b) For the ramp input, y(t) 0.21 at t = 1. second (see Figure P2.36a).
(c) Given R(s) = 1, the partial fraction expansion is
24 12 24 12
Y (s) = = + .
(s + 2)(s + 3)(s + 4) s+2 s+3 s+4

Therefore, using the Laplace transform table, we determine that the


impulse response is

y(t) = 12e2t 24e3t + 412e4t , t0.


2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 53

P2.28 The signal flow graph is shown in Figure P2.28.

-d

G B
+b +c
P D
+a

-m
+e -k
M

+g +h
+f S

FIGURE P2.28
Signal flow graph.

(a) The PGBDP loop gain is equal to -abcd. This is a negative transmis-
sion since the population produces garbage which increases bacteria
and leads to diseases, thus reducing the population.
(b) The PMCP loop gain is equal to +efg. This is a positive transmis-
sion since the population leads to modernization which encourages
immigration, thus increasing the population.
(c) The PMSDP loop gain is equal to +ehkd. This is a positive trans-
mission since the population leads to modernization and an increase
in sanitation facilities which reduces diseases, thus reducing the rate
of decreasing population.
(d) The PMSBDP loop gain is equal to +ehmcd. This is a positive
transmission by similar argument as in (3).
P2.29 Assume the motor torque is proportional to the input current

Tm = ki .

Then, the equation of motion of the beam is

J = ki ,

where J is the moment of inertia of the beam and shaft (neglecting the
inertia of the ball). We assume that forces acting on the ball are due to
gravity and friction. Hence, the motion of the ball is described by

mx = mg bx
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

54 CHAPTER 2 Mathematical Models of Systems

where m is the mass of the ball, b is the coefficient of friction, and we


have assumed small angles, so that sin . Taking the Laplace transfor
of both equations of motion and solving for X(s) yields

gk/J
X(s)/I(s) = .
s2 (s2+ b/m)
P2.30 Given
k
H(s) =
s + 1
where = 4s = 4 106 seconds and 0.999 k < 1.001. The step
response is
k 1 k k
Y (s) = = .
s + 1 s s s + 1/
Taking the inverse Laplace transform yields

y(t) = k ket/ = k(1 et/ ) .

The final value is k. The time it takes to reach 98% of the final value is
t = 15.6s independent of k.
P2.31 From the block diagram we have

Y1 (s) = G2 (s)[G1 (s)E1 (s) + G3 (s)E2 (s)]


= G2 (s)G1 (s)[R1 (s) H1 (s)Y1 (s)] + G2 (s)G3 (s)E2 (s) .

Therefore,
G1 (s)G2 (s) G2 (s)G3 (s)
Y1 (s) = R1 (s) + E2 (s) .
1 + G1 (s)G2 (s)H1 (s) 1 + G1 (s)G2 (s)H1 (s)

And, computing E2 (s) (with R2 (s) = 0) we find

G4 (s)
 
E2 (s) = H2 (s)Y2 (s) = H2 (s)G6 (s) Y1 (s) + G5 (s)E2 (s)
G2 (s)
or
G4 (s)G6 (s)H2 (s)
E2 (s) = Y1 (s) .
G2 (s)(1 G5 (s)G6 (s)H2 (s))

Substituting E2 (s) into equation for Y1 (s) yields

G1 (s)G2 (s)
Y1 (s) = R1 (s)
1 + G1 (s)G2 (s)H1 (s)
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 55

G3 (s)G4 (s)G6 (s)H2 (s)


+ Y1 (s) .
(1 + G1 (s)G2 (s)H1 (s))(1 G5 (s)G6 (s)H2 (s))

Finally, solving for Y1 (s) yields

Y1 (s) = T1 (s)R1 (s)

where

T1 (s) =
G1 (s)G2 (s)(1 G5 (s)G6 (s)H2 (s))
 
.
(1 + G1 (s)G2 (s)H1 (s))(1 G5 (s)G6 (s)H2 (s)) G3 (s)G4 (s)G6 (s)H2 (s)

Similarly, for Y2 (s) we obtain

Y2 (s) = T2 (s)R1 (s) .

where

T2 (s) =
G1 (s)G4 (s)G6 (s)
 
.
(1 + G1 (s)G2 (s)H1 (s))(1 G5 (s)G6 (s)H2 (s)) G3 (s)G4 (s)G6 (s)H2 (s)

P2.32 The signal flow graph shows three loops:

L1 = G1 G3 G4 H2
L2 = G2 G5 G6 H1
L3 = H1 G8 G6 G2 G7 G4 H2 G1 .

The transfer function Y2 /R1 is found to be

Y2 (s) G1 G8 G6 1 G2 G5 G6 2
= ,
R1 (s) 1 (L1 + L2 + L3 ) + (L1 L2 )

where for path 1

1 = 1

and for path 2

2 = 1 L1 .

Since we want Y2 to be independent of R1 , we need Y2 /R1 = 0. Therefore,


we require

G1 G8 G6 G2 G5 G6 (1 + G1 G3 G4 H2 ) = 0 .
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

56 CHAPTER 2 Mathematical Models of Systems

P2.33 The closed-loop transfer function is


Y (s) G3 (s)G1 (s)(G2 (s) + K5 K6 )
= .
R(s) 1 G3 (s)(H1 (s) + K6 ) + G3 (s)G1 (s)(G2 (s) + K5 K6 )(H2 (s) + K4 )
P2.34 The equations of motion are

m1 y1 + b(y1 y2 ) + k1 (y1 y2 ) = 0
m2 y2 + b(y2 y1 ) + k1 (y2 y1 ) + k2 y2 = k2 x

Taking the Laplace transform yields

(m1 s2 + bs + k1 )Y1 (s) (bs + k1 )Y2 (s) = 0


(m2 s2 + bs + k1 + k2 )Y2 (s) (bs + k1 )Y1 (s) = k2 X(s)

Therefore, after solving for Y1 (s)/X(s), we have


Y2 (s) k2 (bs + k1 )
= .
X(s) (m1 s + bs + k1 )(m2 s2 + bs + k1 + k2 ) (bs + k1 )2
2

P2.35 (a) We can redraw the block diagram as shown in Figure P2.35. Then,
K1 /s(s + 1) K1
T (s) = = 2 .
1 + K1 (1 + K2 s)/s(s + 1) s + (1 + K2 K1 )s + K2
(b) The signal flow graph reveals two loops (both touching):
K1 K1 K2
L1 = and L2 = .
s(s + 1) s+1
Therefore,
K1 /s(s + 1) K1
T (s) = = 2 .
1 + K1 /s(s + 1) + K1 K2 /(s + 1) s + (1 + K2 K1 )s + K1
(c) We want to choose K1 and K2 such that

s2 + (1 + K2 K1 )s + K1 = s2 + 20s + 100 = (s + 10)2 .

Therefore, K1 = 100 and 1 + K2 K1 = 20 or K2 = 0.19.


(d) The step response is shown in Figure P2.35.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

58 CHAPTER 2 Mathematical Models of Systems

(d) For the impulse input, y(t) 0.65 at t = 1 seconds (see Figure P2.36b).

(a) Ramp input (b) Impulse input


2 0.8

1.8
0.7

1.6
0.6
1.4

0.5
1.2
y(t)

y(t)
1 0.4

0.8
0.3

0.6
0.2
0.4

0.1
0.2

0 0
0 1 2 3 0 1 2 3
Time (sec) Time (sec)

FIGURE P2.36
(a) Ramp input response. (b) Impulse input response.

P2.37 The equations of motion are


d2 x d2 y
m1 = (k1 + k2 )x + k2 y and m2 = k2 (x y) + u .
dt2 dt2
When m1 = m2 = 1 and k1 = k2 = 1, we have
d2 x d2 y
= 2x + y and =xy+u .
dt2 dt2
P2.38 The equation of motion for the system is
d2 d
J + b + k = 0 ,
dt2 dt
where k is the rotational spring constant and b is the viscous friction
coefficient. The initial conditions are (0) = o and (0) = 0. Taking the
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 59

Laplace transform yields

J(s2 (s) so ) + b(s(s) o ) + k(s) = 0 .

Therefore,

(s + Jb o ) (s + 2n )o
(s) = b K
= .
(s2 + Js + J)
s2+ 2n s + n2

Neglecting the mass of the rod, the moment of inertia is detemined to be

J = 2M r 2 = 0.5 kg m2 .

Also,
s
k b
n = = 0.02 rad/s and = = 0.01 .
J 2Jn

Solving for (t), we find that

o q
(t) = p en t sin(n 1 2 t + ) ,
1 2
p
where tan = 1 2 /). Therefore, the envelope decay is

o
e = p en t .
1 2

So, with n = 2 104 , o = 4000o and f = 10o , the elapsed time is


computed as
1 o
t= ln p = 8.32 hours .
n 1 2 f

P2.39 When t < 0, we have the steady-state conditions

i1 (0) = 1A , va (0) = 2V and vc (0) = 5V ,

where vc (0) is associated with the 1F capacitor. After t 0, we have

di1
2 + 2i1 + 4(i1 i2 ) = 10e2t
dt
and
Z
i2 dt + 10i2 + 4(i2 i1 ) i1 = 0 .
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

60 CHAPTER 2 Mathematical Models of Systems

Taking the Laplace transform (using the initial conditions) yields


10 s+7
2(sI1 i1 (0)) + 2I1 + 4I1 4I2 = or (s + 3)I1 (s) 2I2 (s) =
s+2 s+2
and
1
[ I2 vc (0)]+10I2 +4(I2 I1 ) = I1 (s) or 5sI1 (s)+(14s+1)I2 (s) = 5s .
s
Solving for I2 (s) yields

5s(s2 + 6s + 13)
I2 = ,
14(s + 2)(s)
where

s+3 2

= 14s2 + 33s + 3 .

(s) =
5s 14s + 1

Then,

Vo (s) = 10I2 (s) .

P2.40 The equations of motion are

J1 1 = K(2 1 ) b(1 2 ) + T and J2 2 = b(1 2 ) .

Taking the Laplace transform yields

(J1 s2 + bs + K)1 (s) bs2 (s) = K2 (s) + T (s)

and

(J2 s2 + bs)2 (s) bs1 (s) = 0 .

Solving for 1 (s) and 2 (s), we find that


(K2 (s) + T (s))(J2 s + b) b(K2 (s) + T (s))
1 (s) = and 2 (s) = ,
(s) (s)
where

(s) = J1 J2 s3 + b(J1 + J2 )s2 + J2 Ks + bK .

P2.41 Assume that the only external torques acting on the rocket are control
torques, Tc and disturbance torques, Td , and assume small angles, (t).
Using the small angle approximation, we have

h = V
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 61

J = Tc + Td ,

where J is the moment of inertia of the rocket and V is the rocket velocity
(assumed constant). Now, suppose that the control torque is proportional
to the lateral displacement, as

Tc (s) = KH(s) ,

where the negative sign denotes a negative feedback system. The corre-
sponding block diagram is shown in Figure P2.41.

Td

Tc +
H desired=0 K 1 V
+ Js 2 s H( s)
+
-

FIGURE P2.41
Block diagram.

P2.42 (a) The equation of motion of the motor is


d
J = Tm b ,
dt
where J = 0.1, b = 0.06, and Tm is the motor input torque.
(b) Given Tm (s) = 1/s, and (0) = 0.7, we take the Laplace transform
of the equation of motion yielding

s(s) (0) + 0.6(s) = 10Tm

or
0.7s + 10
(s) = .
s(s + 0.6)
Then, computing the partial fraction expansion, we find that
A B 16.67 15.97
(s) = + = .
s s + 0.6 s s + 0.6
The step response, determined by taking the inverse Laplace trans-
form, is

(t) = 16.67 15.97e0.6t , t0.


2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

62 CHAPTER 2 Mathematical Models of Systems

P2.43 The work done by each gear is equal to that of the other, therefore

Tm m = TL L .

Also, the travel distance is the same for each gear, so

r1 m = r2 L .

The number of teeth on each gear is proportional to the radius, or

r1 N 2 = r2 N 1 .

So,
m r2 N2
= = ,
L r1 N1
and

N1 m = N2 L
N1
L = m = nm ,
N2
where

n = N1 /N2 .

Finally,
Tm L N1
= = =n.
TL m N2
P2.44 The inertia of the load is
Lr 4
JL = .
2
Also, from the dynamics we have

T2 = JL 2 + bL 2

and

T1 = nT2 = n(JL 2 + bL 2 ) .

So,

T1 = n2 (JL 1 + bL 1 ) ,
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 63

since

2 = n1 .

Therefore, the torque at the motor shaft is

T = T1 + Tm = n2 (JL 1 + bL 1 ) + Jm 1 + bm 1 .

P2.45 Let U (s) denote the human input and F (s) the load input. The transfer
function is
G(s) + KG1 (s) Gc (s) + KG1 (s)
P (s) = U (s) + F (s) ,
(s) (s)
where

= 1 + GH(s) + G1 KBH(s) + Gc E(s) + G1 KE(s) .

Consider the application of Newtons law (


P
P2.46 F = mx). From the mass
mv we obtain

mv x1 = F k1 (x1 x2 ) b1 (x1 x2 ).

Taking the Laplace transform, and solving for X1 (s) yields


1 b1 s + k1
X1 (s) = F (s) + X2 (s),
1 (s) 1 (s)
where

1 := mv s2 + b1 s + k1 .

From the mass mt we obtain

mt x2 = k2 x2 b2 x2 + k1 (x1 x2 ) + b1 (x1 x2 ).

Taking the Laplace transform, and solving for X2 (s) yields


b1 s + k1
X2 (s) = X1 (s),
2 (s)
where

2 := mt s2 + (b1 + b2 )s + k1 + k2 .

Substituting X2 (s) above into the relationship fpr X1 (s) yields the trans-
fer function
X1 (s) 2 (s)
= .
F (s) 1 (s)2 (s) (b1 s + k1 )2
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

64 CHAPTER 2 Mathematical Models of Systems

P2.47 Using the following relationships


Z
h(t) = (1.6(t) h(t))dt

(t) = (t)
J (t) = Km ia (t)
va (t) = 50vi (t) = 10ia (t) + vb (t)
= Kvb

we find the differential equation is


d3 h Km d2 h Km dh 8Km
 
3
+ 1+ 2
+ = vi .
dt 10JK dt 10JK dt J
P2.48 (a) The transfer function is
V2 (s) (1 + sR1 C1 )(1 + sR2 C2 )
= .
V1 (s) R1 C 2 s
(b) When R1 = 100 k, R2 = 200 k, C1 = 1 F and C2 = 0.1 F , we
have
V2 (s) 0.2(s + 10)(s + 50)
= .
V1 (s) s
P2.49 (a) The closed-loop transfer function is
G(s) 6205
T (s) = = 3 .
1 + G(s) s + 13s2 + 1281s + 6205

(b) The poles of T (s) are s1 = 5 and s2,3 = 4 j35.


(c) The partial fraction expansion (with a step input) is
1.0122 0.0061 + 0.0716j 0.0061 0.0716j
Y (s) = 1 + + .
s+5 s + 4 + j35 s + 4 j35
(d) The step response is shown in Figure P2.49. The real and complex
roots are close together and by looking at the poles in the s-plane we
have difficulty deciding which is dominant. However, the residue at
the real pole is much larger and thus dominates the response.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Problems 65

0.9

0.8

0.7

0.6

Amplitude 0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Time (secs)

FIGURE P2.49
Step response.

P2.50 (a) The closed-loop transfer function is


14000
T (s) = .
s3 + 45s2 + 3100s + 14500

(b) The poles of T (s) are

s1 = 5 and s2,3 = 20 j50.

(c) The partial fraction expansion (with a step input) is


0.9655 1.0275 0.0310 0.0390j 0.0310 + 0.0390j
Y (s) = + + .
s s+5 s + 20 + j50 s + 20 j50
(d) The step response is shown in Figure P2.50. The real root dominates
the response.
(e) The final value of y(t) is

yss = lim sY (s) = 0.9655 .


s0
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

66 CHAPTER 2 Mathematical Models of Systems

0.9

0.8

0.7

0.6
Amplitude
0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Time (secs)

FIGURE P2.50
Step response.

P2.51 Consider the free body diagram in Figure P2.51. Using Newtons Law
and summing the forces on the two masses yields

M1 x(t) + b1 x(t) + k1 x(t) = b1 y(t)


M2 y(t) + b1 y(t) + k2 y(t) = b1 x(t) + u(t)

k1x

M1
k1
x
. . M1 k2
b1(x - y)
. . x
b1(y - x) k2 y b1

M2 M2
y y
u(t) u(t)

FIGURE P2.51
Free body diagram.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Advanced Problems 67

Advanced Problems
AP2.1 The transfer function from V (s) to (s) has the form
(s) Km
= .
V (s) m s + 1
In the steady-state,
Km 5
 
ss = lim s = 5Km .
s0 m s + 1 s
So,

Km = 70/5 = 14 .

Also,

(t) = Vm Km (1 et/m )

where V (s) = Vm /s. Solving for m yields


t
m = .
ln(1 (t)/ss )
When t = 2, we have
2
m = = 3.57 .
ln(1 30/70)
Therefore, the transfer function is
(s) 14
= .
V (s) 3.57s + 1
AP2.2 The closed-loop transfer function form R1 (s) to Y2 (s) is
Y2 (s) G1 G4 G5 (s) + G1 G2 G3 G4 G6 (s)
=
R1 (s)
where

= [1 + G3 G4 H2 (s)][1 + G1 G2 H3 (s)] .

If we select

G5 (s) = G2 G3 G6 (s)

then the numerator is zero, and Y2 (s)/R1 (s) = 0. The system is now
decoupled.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

68 CHAPTER 2 Mathematical Models of Systems

AP2.3 (a) Computing the closed-loop transfer function:


G(s)Gc (s)
 
Y (s) = R(s) .
1 + Gc (s)G(s)H(s)
Then, with E(s) = R(s) Y (s) we obtain
1 + Gc (s)G(s)(H(s) 1)
 
E(s) = R(s) .
1 + Gc (s)G(s)H(s)
If we require that E(s) 0 for any input, we need 1 + Gc (s)G(s)(H(s)
1) = 0 or
Gc (s)G(s) 1 n(s)
H(s) = = .
Gc (s)G(s) d(s)
Since we require H(s) to be a causal system, the order of the numerator
polynomial, n(s), must be less than or equal to the order of the denom-
inator polynomial, d(s). This will be true, in general, only if both Gc (s)
and G(s) are proper rational functions (that is, the numerator and de-
nominator polynomials have the same order). Therefore, making E 0
for any input R(s) is possible only in certain circumstances.
(b) The transfer function from Td (s) to Y (s) is
Gd (s)G(s)
 
Y (s) = Td (s) .
1 + Gc (s)G(s)H(s)
With H(s) as in part (a) we have
Gd (s)
 
Y (s) = Td (s) .
Gc (s)
(c) No. Since
Gd (s)G(s)
 
Y (s) = Td (s) = T (s)Td (s) ,
1 + Gc (s)G(s)H(s)
the only way to have Y (s) 0 for any Td (s) is for the transfer function
T (s) 0 which is not possible in general (since G(s) 6= 0).
AP2.4 (a) With q(s) = 1/s we obtain
1/Ct 1
(s) = QS+1/R
.
s+ s
Ct

Define
QS + 1/R
:= and := 1/Ct .
Ct
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Advanced Problems 69

Then, it follows that


1 / /
(s) = = + .
s+ s s+ s
Taking the inverse Laplace transform yields
t
(t) = e + = [1 et ] .

(b) As t , (t) = Qs+1/R
1
.
(c) To increase the speed of response, you want to choose Ct , Q, S and
R such that
Qs + 1/R
:=
Ct
is large.
AP2.5 Considering the motion of each mass, we have

M3 x3 + b3 x3 + k3 x3 = u3 + b3 x2 + k3 x2
M2 x2 + (b2 + b3 )x2 + (k2 + k3 )x2 = u2 + b3 x3 + k3 x3 + b2 x1 + k2 x1
M1 x1 + (b1 + b2 )x1 + (k1 + k2 )x1 = u1 + b2 x2 + k2 x2

In matrix form the three equations can be written as



M1 0 0 x1 b1 + b2 b2 0 x1

x2 + b2
0 M2 0 b2 + b3 b3 x
2

0 0 M3 x3 0 b3 b3 x3

k1 + k2 k2 0 x 1 u1

+
k2 k2 + k3 k3 x = u .
2 2

0 k3 k3 x3 u3

AP2.6 Considering the cart mass and using Newtons Law we obtain

M x = u bx F sin

where F is the reaction force between the cart and the pendulum. Con-
sidering the pendulum we obtain
d2 (x + L sin )
m = F sin
dt2
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

70 CHAPTER 2 Mathematical Models of Systems

d2 (L cos )
m = F cos + mg
dt2

Eliminating the reaction force F yields the two equations

(m + M )x + bx + mL cos mL2 sin = u


mL2 + mgL sin + mLx cos = 0

If we assume that the angle 0, then we have the linear model

(m + M )x + bx + mL = u
mL2 + mgL = mLx

AP2.7 The transfer function from the disturbance input to the output is
1
Y (s) = Td (s) .
s + 20 + K
When Td (s) = 1, we obtain

y(t) = e(20+K)t .

Solving for t when y(t) < 0.1 yields


2.3
t> .
20 + K
When t = 0.05 and y(0.05) = 0.1, we find K = 26.05.
AP2.8 The closed-loop transfer function is
200K(0.25s + 1)
T (s) =
(0.25s + 1)(s + 1)(s + 8) + 200K

The final value due to a step input of R(s) = A/s is


200K
v(t) A .
200K + 8
We need to select K so that v(t) 50. However, to keep the percent
overshoot to less than 10%, we need to limit the magnitude of K. Fig-
ure AP2.8a shows the percent overshoot as a function of K. Let K = 0.06
and select the magnitude of the input to be A = 83.3. The inverse Laplace
transform of the closed-loop response with R(s) = 83.3/s is

v(t) = 50 + 9.85e9.15t e1.93t (59.85 cos(2.24t) + 11.27 sin(2.24t))


2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Advanced Problems 71

The result is P.O. = 9.74% and the steady-state value of the output is
approximately 50 m/s, as shown in Figure AP2.8b.

25

20

Percent Overshoot (%)

15

10

0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1
K

Step Response

60

System: untitled1
50 Peak amplitude: 54.9
Overshoot (%): 9.74
At time (sec): 1.15

40
Amplitude

30

20

10

0
0 0.5 1 1.5 2 2.5
Time (sec)

FIGURE AP2.8
(a) Percent overshoot versus the gain K. (b) Step response.

AP2.9 The transfer function is

Vo (s) Z2 (s)
= ,
Vi (s) Z1 (s)
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

72 CHAPTER 2 Mathematical Models of Systems

where
R1 R2 C 2 s + 1
Z1 (s) = and Z2 (s) = .
R1 C 1 s + 1 C2 s
Then we can write
Vo (s) KI
= Kp + + KD s
Vi (s) s
where
R1 C 1 1
 
KP = +1 , KI = , KD = R2 C1 .
R2 C 2 R1 C 2
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Design Problems 73

Design Problems
CDP2.1 The model of the traction drive, capstan roller, and linear slide follows
closely the armature-controlled dc motor model depicted in Figure 2.18
in Dorf and Bishop. The transfer function is

rKm
T (s) = ,
s [(Lm s + Rm )(JT s + bm ) + Kb Km ]

where

JT = Jm + r 2 (Ms + Mb ) .

Km 1 w 1 q
Va(s) r X(s)
Lms+Rm JTs+bm s
-

Kb
Back EMF

DP2.1 The closed-loop transfer function is

Y (s) G1 (s)G2 (s)


= .
R(s) 1 + G1 (s)H1 (s) G2 (s)H2 (s)

When G1 H1 = G2 H2 and G1 G2 = 1, then Y (s)/R(s) = 1. Therefore,


select
1 G2 (s)H2 (s)
G1 (s) = and H1 (s) = = G22 (s)H2 (s) .
G2 (s) G1 (s)

DP2.2 At the lower node we have


1 1
 
v + + G + 2i2 20 = 0 .
4 3

Also, we have v = 24 and i2 = Gv . So

1 1
 
v + + G + 2Gv 20 = 0
4 3
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

74 CHAPTER 2 Mathematical Models of Systems

and
 
1 1
20 v 4 + 3 1
G= = S.
3v 12
DP2.3 Taking the Laplace transform of
1 3 1
y(t) = et e2t + t
4 4 2
yields
1 1 3 1
Y (s) = + 2 .
s + 1 4(s + 2) 4s 2s

Similarly, taking the Laplace transform of the ramp input yields


1
R(s) = .
s2
Therefore
Y (s) 1
G(s) = = .
R(s) (s + 1)(s + 2)

DP2.4 For an ideal op-amp, at node a we have


vin va vo va
+ =0,
R1 R1
and at node b
vin vb
= C vb ,
R2
from it follows that
1 1
 
+ Cs Vb = Vin .
R2 R2
Also, for an ideal op-amp, Vb Va = 0. Then solving for Vb in the above
equation and substituting the result into the node a equation for Va yields
1
" #
Vo 2 1 R2 + Cs
= 1
Vin R2 + Cs R 2 2

or
Vo (s) R2 Cs 1
= .
Vin (s) R2 Cs + 1
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Design Problems 75

For vin (t) = At, we have Vin (s) = A/s2 , therefore


2 2
 
vo (t) = A et + t

where = 1/R2 C.
DP2.5 The equation of motion describing the motion of the inverted pendulum
(assuming small angles) is
g
+ =0.
L
Assuming a solution of the form = k cos , taking the appropriate
derivatives and substituting the result into the equation of motion yields
the relationship
g
r
= .
L
If the period is T = 2 seconds, we compute = 2/T . Then solving for L
yields L = 0.99 meters when g = 9.81 m/s2 . So, to fit the pendulum into
the grandfather clock, the dimensions are generally about 1.5 meters or
more.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

76 CHAPTER 2 Mathematical Models of Systems

Computer Problems
CP2.1 The m-file script is shown in Figure CP2.1.

pq =
p=[1 7 10]; q=[1 2]; 1 9 24 20
% Part (a) P=
pq=conv(p,q) -5
% Part (b) -2
P=roots(p), Z=roots(q) Z=
% Part (c) -2
value=polyval(p,-1) value =
4

FIGURE CP2.1
Script for various polynomial evaluations.

CP2.2 The m-file script and step response is shown in Figure CP2.2.

numc = [1]; denc = [1 1]; sysc = tf(numc,denc)


numg = [1 2]; deng = [1 3]; sysg = tf(numg,deng)
% part (a)
sys_s = series(sysc,sysg);
sys_cl = feedback(sys_s,[1])
% part (b) Transfer function:
step(sys_cl); grid on s+2
-------------
s^2 + 5 s + 5
Step Response
From: U(1)
0.4

0.35

0.3

0.25
Amplitude

To: Y(1)

0.2

0.15

0.1

0.05

0
0 0.5 1 1.5 2 2.5 3 3.5 4

Time (sec.)

FIGURE CP2.2
Step response.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Computer Problems 77

CP2.3 Given

y + 4y + 3y = u

with y(0) = y = 0 and U (s) = 1/s, we obtain (via Laplace transform)


1 1
Y (s) = = .
s(s2 + 4s + 3) s(s + 3)(s + 1)
Expanding in a partial fraction expansion yields
1 1 1
Y (s) = .
3s 6(s + 3) 2(s + 1)
Taking the inverse Laplace transform we obtain the solution

y(t) = 0.3333 + 0.1667e3t 0.5et .

The m-file script and step response is shown in Figure CP2.3.

Step Response
0.35

0.3

0.25

0.2 n=[1]; d=[1 4 3]; sys = tf(n,d);


Amplitude

t=[0:0.1:5];
y = step(sys,t);
0.15 ya=0.3333+0.1667*exp(-3*t)-0.5*exp(-t);
plot(t,y,t,ya); grid;
0.1
title('Step Response');
xlabel('Time (sec)');
ylabel('Amplitude');
0.05

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (sec)

FIGURE CP2.3
Step response.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

78 CHAPTER 2 Mathematical Models of Systems

CP2.4 The mass-spring-damper system is represented by

mx + bx + kx = f .

Taking the Laplace transform (with zero initial conditions) yields the
transfer function
1/m
X(s)/F (s) = .
s2 + bs/m + k/m
The m-file script and step response is shown in Figure CP2.4.

m=10; k=1; b=0.5;


num=[1/m]; den=[1 b/m k/m];
sys = tf(num,den);
t=[0:0.1:150];
step(sys,t)
Step Response
From: U(1)
1.8

1.6

1.4

1.2
Amplitude

1
To: Y(1)

0.8

0.6

0.4

0.2

0
0 50 100 150

Time (sec.)

FIGURE CP2.4
Step response.

CP2.5 The spacecraft simulations are shown in Figure CP2.5. We see that as J
is decreased, the time to settle down decreases. Also, the overhoot from
10o decreases as J decreases. Thus, the performance seems to get better
(in some sense) as J decreases.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Computer Problems 79

Nominal (solid); Off-nominal 80% (dashed); Off-nominal 50% (dotted)


18

16

14

Spacecraft attitude (deg)


12

10

0
0 10 20 30 40 50 60 70 80 90 100

Time (sec)

%Part (a)
a=1; b=8; k=10.8e+08; J=10.8e+08;
num=k*[1 a];
den=J*[1 b 0 0]; sys=tf(num,den);
sys_cl=feedback(sys,[1]);
%
% Part (b) and (c)
t=[0:0.1:100];
%
% Nominal case
f=10*pi/180; sysf=sys_cl*f ;
y=step(sysf,t);
%
% Off-nominal case 80%
J=10.8e+08*0.8; den=J*[1 b 0 0];
sys=tf(num,den); sys_cl=feedback(sys,[1]);
sysf=sys_cl*f ;
y1=step(sysf,t);
%
% Off-nominal case 50%
J=10.8e+08*0.5; den=J*[1 b 0 0];
sys=tf(num,den); sys_cl=feedback(sys,[1]);
sysf=sys_cl*f ;
y2=step(sysf,t);
%
plot(t,y*180/pi,t,y1*180/pi,'--',t,y2*180/pi,':'),grid
xlabel('Time (sec)')
ylabel('Spacecraft attitude (deg)')
title('Nominal (solid); Off-nominal 80% (dashed); Off-nominal 50% (dotted)')

FIGURE CP2.5
Step responses for the nominal and off-nominal spacecraft parameters.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

80 CHAPTER 2 Mathematical Models of Systems

CP2.6 The closed-loop transfer function is

4s6 + 8s5 + 4s4 + 56s3 + 112s2 + 56s


T (s) = ,
(s)

p=

7.0709
num1=[4]; den1=[1]; sys1 = tf(num1,den1);
num2=[1]; den2=[1 1]; sys2 = tf(num2,den2); -7.0713
num3=[1 0]; den3=[1 0 2]; sys3 = tf(num3,den3); 1.2051 + 2.0863i
num4=[1]; den4=[1 0 0]; sys4 = tf(num4,den4); 1.2051 - 2.0863i
num5=[4 2]; den5=[1 2 1]; sys5 = tf(num5,den5); 0.1219 + 1.8374i
num6=[50]; den6=[1]; sys6 = tf(num6,den6); 0.1219 - 1.8374i
num7=[1 0 2]; den7=[1 0 0 14]; sys7 = tf(num7,den7); -2.3933
sysa = feedback(sys4,sys6,+1); -2.3333
sysb = series(sys2,sys3);
-0.4635 + 0.1997i
sysc = feedback(sysb,sys5);
-0.4635 - 0.1997i
sysd = series(sysc,sysa);
syse = feedback(sysd,sys7);
sys = series(sys1,syse) z=
% poles 0
pzmap(sys) 1.2051 + 2.0872i
% 1.2051 - 2.0872i
p=pole(sys) -2.4101
z=zero(sys) -1.0000 + 0.0000i
-1.0000 - 0.0000i
Polezero map

2.5

1.5

0.5
Imag Axis

-0.5

-1

-1.5

-2

-2.5
-8 -6 -4 -2 0 2 4 6 8

Real Axis

FIGURE CP2.6
Pole-zero map.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Computer Problems 81

where

(s) = s10 + 3s9 45s8 125s7 200s6 1177s5


2344s4 3485s3 7668s2 5598s 1400 .

CP2.7 The m-file script and plot of the pendulum angle is shown in Figure CP2.7.
With the initial conditions, the Laplace transform of the linear system is
0 s
(s) = .
s2 + g/L
To use the step function with the m-file, we can multiply the transfer
function as follows:
s2 0
(s) = 2
,
s + g/L s
which is equivalent to the original transfer function except that we can
use the step function input with magnitude 0 . The nonlinear response
is shown as the solid line and the linear response is shown as the dashed
line. The difference between the two responses is not great since the initial
condition of 0 = 30 is not that large.

30
L=0.5; m=1; g=9.8;
theta0=30;
20 % Linear simulation
sys=tf([1 0 0],[1 0 g/L]);
[y,t]=step(theta0*sys,[0:0.01:10]);
10
% Nonlinear simulation
[t,ynl]=ode45(@pend,t,[theta0*pi/180 0]);
plot(t,ynl(:,1)*180/pi,t,y,'--');
(deg)

0
xlabel('Time (s)')
ylabel('\theta (deg)')
-10
function [yd]=pend(t,y)
L=0.5; g=9.8;
-20 yd(1)=y(2);
yd(2)=-(g/L)*sin(y(1));
yd=yd';
-30
0 2 4 6 8 10
Time (s)

FIGURE CP2.7
Plot of versus xt when 0 = 30 .
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

82 CHAPTER 2 Mathematical Models of Systems

CP2.8 The system step responses for z = 5, 10, and 15 are shown in Fig-
ure CP2.8.

z=5 (solid), z=10 (dashed), z=15 dotted)


1.5

1
x(t)

0.5

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (sec)

FIGURE CP2.8
The system response.

CP2.9 (a,b) Computing the closed-loop transfer function yields

G(s) s2 + 2s + 1
T (s) = = 2 .
1 + G(s)H(s) s + 4s + 3

The poles are s = 3, 1 and the zeros are s = 1, 1.


(c) Yes, there is one pole-zero cancellation. The transfer function (after
pole-zero cancellation) is

s+1
T (s) = .
s+3
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

Computer Problems 83

Pole?Zero Map
1

0.8

0.6

0.4

0.2
Imaginary Axi s

?-0.2

?-0.4

?-0.6

?-0.8

?-1
?-3 ?-2.5 ?-2 ?-1.5 ?-1 ?-0.5 0
Real Axi s
>>
Transfer function:
ng=[1 1]; dg=[1 2]; sysg = tf(ng,dg); s^2 + 2 s + 1
nh=[1]; dh=[1 1]; sysh = tf(nh,dh); -------------
sys=feedback(sysg,sysh) s^2 + 4 s + 3
%
pzmap(sys)
% poles p=
pole(sys)
zero(sys) -3
-1

zeros
z=

-1
-1

FIGURE CP2.9
Pole-zero map.

CP2.10 Figure CP2.10 shows the steady-state response to a unit step input and a
unit step disturbance. We see that K = 1 leads to the same steady-state
response.
2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.

84 CHAPTER 2 Mathematical Models of Systems

0.35

K=[0.1:0.1:10]; 0.3
sysg=tf([1],[1 20 20]);
for i=1:length(K)
0.25
nc=K(i); dc=[1];sysc=tf(nc,dc); Input Response Steady-State

Steadystate response
syscl=feedback(sysc*sysg,1);
systd=feedback(sysg,sysc); 0.2
y1=step(syscl);
Tf1(i)=y1(end); 0.15
y2=step(systd);
Tf2(i)=y2(end);
0.1
end
Disturbance Response Steady-State
plot(K,Tf1,K,Tf2,'--')
xlabel('K') 0.05
ylabel('Steady-state response')
K=1
0
0 1 2 3 4 5 6 7 8 9 10
K

FIGURE CP2.10
Gain K versus steady-state value.