Mechanical Vibrations: Theory and Applications

FRANCIS S.
TSE
University of Cincinnati
IVAN E. MORSE
University of Cincinnati
T. HINKLE
Michigan State University
Mechanical Vibrations
Theory and Applications
SECOND EDITION
Allyn and Bacon,

Boston Sydney Toronto
Copyright 1978, 1%3 by and Bacon, Inc.
470 Atlantic Avenue, Boston, Massachusetts 02210.
rights reserved. Printed in the United States of America.

No part of the material protected by this copyright notice
may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying, recording,
or by any information storage and retrieval system, without
written permission from the copyright owner.
Library of Congress Cataloging in Publication Data
Tse, Francis Sing

Mechanical vibrations.
(Allyn and Bacon series in Mechanical engineering
and applied mechanics)
Includes index.
1. Vibrations. I. Morse, Ivan E., joint
author. Hinkle, Theodore, joint
author. Title.
1978 620.3 77-20933
ISBN
ISBN (International)
Contents
Preface xi
CHAPTER 1 INTRODUCTION
1- 1Primary Objective 1
1-2 Elements of a Vibratory System 2
1-3 Examples of Vibratory Motions 5
1-4 Simple Harmonic Motion
1-5 Vectorial Representation
of Harmonic Motions 11
1-6 Units 16
1-7 Summary 19
Problems 20
CHAPTER 2 SYSTEMS WITH ONE DEGREE

OF FREEDOM-THEORY
2-1 Introduction 23
2-2 Degrees of Freedom 25
2-3 Equation of Motion-Energy Method 27
2-4 Equation of Motion-Newton's Law
of Motion 33
2-5 General Solution 34
Complementary Function 34
Particular Integral 38
General Solution 42
vi Contents
2-6 Frequency Response Method 45

Impedance Method 45
Transfer Function 49
Resonance, Damping, and Bandwidth 51
2-7 Transient Vibration 52
Impulse Response 53
Convolution Integral 55
Indicia1 Response 57
2-8 Comparison of Rectilinear
and Rotational Systems 58
2-9 Summary 58
Problems 62
CHAPTER 3 SYSTEMS WITH ONE DEGREE

OF FREEDOM-APPLICATIONS
Introduction 69
Undamped Free Vibration 70
Damped-Free Vibration 77
Undamped Forced Vibration-
Harmonic Excitation 80
Damped Forced Vibration-
Harmonic Excitation 86
Rotating and Reciprocating
Unbalance 87
Critical Speed of Rotating Shafts 89
Vibration Isolation and
Transmissibility 94
Systems Attached to Moving Support 98
Seismic Instruments 101
Elastically Supported Damped
Systems 106
Damped Forced Vibration-
Periodic Excitation 109
Transient Vibration-Shock Spectrum 116
Equivalent Viscous Damping 122
Summary 129
Problems 131
CHAPTER 4 SYSTEMS WITH MORE THAN ONE DEGREE

OF FREEDOM 142
4-1 Introduction 142
4-2 Equations of Motion:
Newton's Second Law 143
4-3 Undamped Free Vibration: Principal
Modes
4-4 Generalized and
Coupling
4-5 Principal Coordinates 158
4-6 Modal Analysis: ient Vibration
of Undamped S 160
4-7 Systems 165
4-8 Forced Vibration-Harmonic Excitation 169
4-9 Influence Coefficients 175
4-10 180
Problems 181
CHAPTER 5 METHODS FOR

NATURAL
5-2 Equation 190
5-3 Rayleigh Method 193
5-4 Method 197
5-5 Transfer Matrix 202
Myklestad-Prohl Method
5-7 213
Problems
CHAPTER 6 DISCRETESYSTEMS
6-2 Equations of Motion-Undamped
Systems
6-3 Undamped Vibration-Principal
Modes 223
6-4 Orthogonality and Principal Coordinates 226
6-5 Coordinates 229
6-6 Expansion Theorem 230
6-7 Quotient 231
6-8 Semidefinite Systems 232
6-9 Matrix Iteration 234
6-10 Undamped Forced
Vibration-Modal Analysis 238
6-1 1 Systems with Proportional
6-12 Orthogonality of Modes of Damped
Systems 241
6-13 Damped Forced
Contents
6-14 Summary 245

Problems 246
CHAPTER 7 CONTINUOUS SYSTEMS

7-2 Continuous
A Simple Exposition 253
7-3 Separation of the Time
and Space Variables 256
7-4 Problems Governed by
the Wave Equation 258
Longitudinal Vibration of Rods 258
Torsional Vibration of Shafts 261
7-5 Lateral Vibration of Beams 262
7-6 Rotary Inertia and Other Effects 265
Shear Deformation and Rotary
Inertia Effects 266
Effect of Axial Loading 268
7-7 The Eigenvalue Problem 270
7-8 Orthogonality 272
Boundary Conditions
Independent of A 273
Boundary Conditions Dependent
on A 275
7-9 Lagrange's Equations 280
7-10 Undamped Forced
7-11 Rayleigh's Quotient 288
7-12 Rayleigh-Ritz Method 290
7-13 Summary 294
Problems 295
CHAPTER 8 NONLINEAR SYSTEMS

8-2 Stability and Examples of Nonlinear
Systems 301
8-3 The Phase Plane 303
8-4 Stability of Equilibrium 306
8-5 Graphical Methods 316
Isocline Method 317
Pell's Method 319
8-6 Self-excited Oscillations 321
8-7 Analytical Methods 323
Contents
8-8 Free Vibration 323

Perturbation Method 323
Variation of Parameter Method 325
Balance 327
8-9 Forced Vibration 328
Jump Phenomenon 328
Subharmonic Oscillation 332
8-10 Summary 334
Problems 335
CHAPTER 9 SOLUTIONS BY DIGITAL COMPUTERS

9-2 One-degree-of-freedom
Systems-Transient Response 341
9-3 Program-TRESP1 342
9-4 Program-TKESPSUB 346
9-5 349
9-6 One-degree-of-freedom
Systems-Harmonic Response 352
9-7 N--degree-of-freedom
Systems-Harmonic Response 356
9-8 Transient Response of Undamped
Discrete Systems 360
9-9 Rayleigh's Method-Undamped
Multirotor Systems 365
9-10 Myklestad-Prohl Method-Transfer
Matrix Technique 369
9-11 Matrix Iteration-Undamped
Discrete Systems 371
9-12 Transient Response of Damped Systems 376
9-13 Summary 380
Problems 385
A Elements of Matrix Algebra

APPENDIX B Lagrange's Equations
APPENDIX C Subroutines
D Linear Ordinary Differential Equations
with Constant Coefficients
Index
Preface
Vibration is study of oscillatory motions. The ultimate goals of

this study are to determine the effect of vibration on the performance and
safety of systems, and to control its effects. With the advent of high per-
formance machines and environmental control, this study has become a
part of most engineering curricula.
text presents the fundamentals and applications of vibration
theory. It is intended for students taking either a first course or a one-year
sequence in the subject at the junior or senior level. The student is assumed
to have an elementary knowledge of dynamics, strength of materials, and
differential equations, although summaries of several topics are included in
the appendices for review purposes. The format of its predecessor is re-
tained, but the text material has been substantially rewritten. In view of the
widespread adoption of the International System of Units (SI) by the indus-
trial world, SI units are used in the problems.
The objectives of the text are first, to establish a sense of engineering
reality, second, to provide adequate basic theory, and finally, to generalize
these concepts for wider applications: The primary focus of the text is on
the engineering significance of the physical quantities, with the mathemat-
ical structure providing a supporting role. Throughout the text, examples of
applications are given before the generalization to give the student a frame
of reference, and to avoid the pitfall of overgeneralization. T o further
enhance engineering reality, detailed digital computations for discrete sys-
tems are presented so that the student can solve meaningful numerical prob-
lems.
The first three chapters examine systems with one degree of freedom.
General concepts of vibration are described in Chapter 1. The theory of
xii Preface
time and frequency domain analysis is introduced in Chapter 2 through the

study of a generalized model, consisting of the mass, spring, damper, and
excitation elements. This provides the basis for modal analyses in subse-
quent chapters. The applications in Chapter 3 demonstrate that the ele-
ments of the model are, in effect, equivalent quantities. Although the same
theory is used, the appearance of a system in an engineering problem may
differ greatly from that of the model. The emphasis of Chapter 3 is on prob-
lem formulation. Through the generalization and classification of problems
in the chapter, a new encounter will not appear as a stranger.
Discrete systems are introduced in Chapter 4 using systems with two
degrees of freedom. Coordinate coupling is treated in detail. Common
methods of finding natural frequencies are described in Chapter 5. The
material in these chapters is further developed in Chapter 6 using matrix
techniques and relating the matrices to energy quantities. Thus, the student
would not feel the artificiality in the numerous coordinate transformations
in the study.
The one-dimensional wave equation and beam equation of continu-
ous systems are discussed in Chapter 7. The material is organized to show
the similarities between continuous and discrete systems. Chapter 8, on
nonlinear systems, explains certain common phenomena that cannot be pre-
dicted by linear theory. The chapter consists of two main parts, conforming
to the geometric and analytical approaches to studies.
The digital computation in Chapter 9 is organized to follow the
sequence of topics presented in the prior chapters and can be assigned con-
currently with the text material. The programs listed in Table 9-1 are suffi-
cient for the computation and plotting of results for either damped or
undamped discrete systems. Detailed explanations are given to aid the stu-
dent in executing the programs. The programs are almost conversational
and only a minimal knowledge of FORTRAN is necessary for their execu-
tion.
The first five chapters constitute the core of an elementary,
quarter terminal course at the junior level. Depending on the purpose of the
particular course, parts of Section 3-5 can be used as assigned reading. Sec-
tions 3-6 through 3-8, Section 4-9, and Sections 5-4 through 5-6 may be
omitted without loss of continuity.
For a one-semester senior or dual-level course, the instructor may
wish to use Chapters 1 through 4, Chapter 6, and portions of Chapter 7 or
8. Some topics, such as equivalent viscous damping, may be omitted.
Alternatively, the text has sufficient material for a one-year sequence
at the junior or senior level. Generally, the first course in mechanical vibra-
tions is required and the second is an elective. The material covered will give
the student a good background for more advanced studies.
We would like to acknowledge our indebtedness to many friends,
students, and colleagues for their suggestions, to the numerous writers who
SEC. 1-5 Vectorial Representation of Harmonic Motions 13
wr
(a) representation Harmonic motions
FIG. 1-8. Displacement, velocity, and acceleration vectors.
Thus, each differentiation is equivalent to the multiplication of the vector

by Since X is the magnitude of the vector X, is real, = 1, each
differentiation changes the magnitude by a factor of Since the multipli-
cation of a vector by is equivalent to advancing it by a phase angle of
each differentiation also advances a vector by 90".
If a given harmonic displacement is = cos the relations be-
tween the displacement and its velocity and acceleration are
Displacement = =X cos
Velocity = = - sin
Acceleration = =-
= +
These relations are identical to those shown in (1-4) to (1-6). The
representation of displacement, velocity, and acceleration by rotating
vectors is illustrated in Fig. 1-8. Since the given displacement is a
cosine function, or along the real axis, the velocity and acceleration must
be along the real axis. Hence the real parts of the respective vectors give
the physical quantities at the given time
Harmonic functions can be added graphically be means of vector
addition. The vectors and representing the motions cos wt and
+ a), respectively, are added graphically as shown in Fig.
The resultant vector X has a magnitude
= + cos sin
and a phase angle
sin
=
X, cos
Introduction
1-1 PRIMARY
The subject of vibration deals with the oscillatory motion of dynamic
systems. A dynamic system is a combination of matter which possesses
mass and whose parts are capable of relative motion. All bodies posses-
sing mass and elasticity are capable of vibration. The mass is inherent of
the body, and the elasticity is due to the relative motion of the parts of
the body. The system considered may be very simple or complex. It may
be in the form of a structure, a machine or its components, or a group of
machines. The oscillatory motion of the system may be objectionable,
trivial, or necessary for performing a task.
The objective of the designer is to control the vibration when it is
objectionable and to enhance the vibration when it useful, although
vibrations in general are undesirable. Objectionable vibrations in a
machine cause the loosening of parts, its malfunctioning, or its
eventual failure. On the other hand, shakers in foundries and vibrators in
testing machines require vibration. The performance of many instruments
depends on the proper control of the vibrational characteristics of the
devices.
The primary objective of our study is to analyze the oscillatory motion
of dynamic systems and the forces associated with the motion. The
ultimate goal in the study of vibration is to determine its effect on the
performance and safety of the system under consideration. The analysis
of the oscillatory motion is an important step towards this goal.
Our study begins with the description of the elements in a vibratory
system, the introduction of some terminology and concepts, and the
discussion of simple harmonic motion. These will be used throughout the
text. Other concepts and terminology will be introduced in the appro-
priate places as needed.
Introduction CHAP. 1
1-2 ELEMENTS OF A VIBRATORY SYSTEM
The elements that constitute a vibratory system are illustrated in Fig.

1-1. They are idealized and called (1) the mass, (2) the spring, (3) the
damper, and (4) the excitation. The first three elements describe the
physical system. For example, it can be said that a given system consists
of a mass, a spring, and a damper arranged as shown in the figure. Energy
may be stored in the mass and the spring and dissipated in the damper in
the form of heat. Energy enters the system through the application of an
excitation. As shown in Fig. 1-1, an excitation force is applied to the mass
m of the system.
The mass m is assumed to be a rigid body. It executes the vibrations
and can gain or lose kinetic energy in accordance with the velocity change
of the body. From Newton's law of motion, the product of the mass and
its acceleration is equal to the force applied to the mass, and the
acceleration takes place in the direction in which the force acts. Work is
force times displacement in the direction of the force. The work is
transformed into the kinetic energy of the mass. The kinetic energy
increases if work is positive and decreases if work is negative.
Th spring k possesses elasticity and is assumed to be of negligible
mass. spring force exists if the spring is deformed, such as the extension
or the compression of a coil spring. Therefore the spring force exists only
if there is a relative displacement between the two ends of the spring. The
work done in deforming a spring is transformed into potential energy, that
is, the strain energy stored in the spring. A linear spring is one that obeys
Hooke's law, that is, the spring force is proportional to the spring
deformation. The constant of proportionality, measured in force per unit
deformation, is called the stiffness, or the spring constant k.
The damper c has neither mass nor elasticity. Damping force exists only
if there is relative motion between the two ends of the damper. The work
or the energy input to a damper is converted into heat. Hence the
damping element is nonconservatiue. Viscous damping, in which the
damping force is proportional to the velocity, is called linear damping.
Viscous damping, or its equivalent, is generally assumed in engineering.
Static Excitation
equilibrium force
position
t
Displacement
X
1-1. Elements of a vibratory system.

SEC. 1-2 Elements of a Vibratory System
FIG. 1-2. A periodic motion.
viscous damping c is measured in force per unit velocity.

Many types of nonlinear damping are commonly encountered. For exam-
ple, the frictional drag of a body moving in a fluid is approximately
proportional to the velocity squared, but the exact value of the exponent
is dependent on many variables.
Energy enters a system through the application of an excitation. An
excitation force may be applied to the mass and/or an excitation motion
applied to the spring and the damper. An excitation force applied to
the mass m is illustrated in Fig. The excitation varies in accordance
with a prescribed function of time. Hence the excitation is always known
at a given Alternatively, if the system is suspended a support,
excitation may be applied to the system through imparting a prescribed
motion to support. In machinery, excitation often arises from the
unbalance of the moving components. The vibrations of dynamic systems
under the influence of an excitation is called forced vibrations. Forced
vibrations, however, are often defined as the vibrations that are caused
and maintained by a periodic excitation.
If the vibratory motion is periodic, the system repeats its motion at
equal time intervals as shown in Fig. 1-2. The minimum time required for
the system to repeat its motion is called a period which is the time to
complete one cycle of motion. Frequency is the number of times that the
motion repeats itself per unit time. A motion that does not repeat itself at
equal time intervals is called an aperiodic motion.
A dynamic system can be set into motion by some initial conditions, or
disturbances at time equal to zero. If no disturbance or excitation is
applied after the zero time, the oscillatory motions of the system are
called free vibrations. Hence free vibrations describe the natural behavior
or the natural modes of vibration of a system. The initial condition is an
energy input. If a spring is deformed, the input is potential energy. If a
mass is given an initial velocity, the input is kinetic energy. Hence initial
conditions are due to the energy initially stored in the system.
If the system does not possess damping, there is no energy dissipation.
Initial conditions would cause the system to vibrate and the free vibration
of an undamped system will not diminish with time. If a system possesses
Introduction CHAP.
damping, energy will be dissipated in the damper. Hence the free vibra-
tions will eventually die out and the system then remain at its static
equilibrium position. Since the energy stored is due to the initial conditions,
free vibrations also describe the natural behavior of the system as it
relaxes from the initial state to its static equilibrium.
For simplicity, lumped masses, linear springs, and viscous dampers will
be assumed unless otherwise stated. Systems possessing these characteris-
tics are called linear systems. An important property of linear systems is
that they follow the principle of superposition. For example, the resultant
motion of the system due to the simultaneous application of two excita-
tions is a linear combination of the motions due to each of the excitations
acting separately. The values of m, c, and k of the elements in Fig. 1-1
are often referred to as the system parameters. For a given problem, these
values are assumed time invariant. Hence the coefficients or the parame-
ters in the equations are constants. The equation of motion of the system
becomes a linear ordinary differential equation with constant coefficients,
which can be solved readily.
Note that the idealized elements in Fig. 1-1 form a model of a vibratory
system which in reality can be quite complex. For example, a coil spring
possesses both mass and elasticity. In order to consider it as an idealized
spring, either its mass is assumed negligible or an appropriate portion of
its mass is lumped together with the other masses of the system. The
resultant model is a lumped-parameter, or discrete, system. For example, a
beam has its mass and elasticity inseparably distributed along its length.
The vibrational characteristics of a beam, or more generally of an elastic
body or a continuous system, can be studied by this approach if the
continuous system is approximated by a finite number of lumped parame-
ters. This method is a practical approach to the study of some very
complicated structures, such as an aircraft.
In spite of the limitations, the lumped-parameter approach to the study
of vibration problems is well justified for the following reasons. (1) Many
physical systems are essentially discrete systems. (2) The concepts can be
extended to analyze the vibration of continuous systems. (3) Many
physical systems are too complex to be investigated analytically as elastic
bodies. These are often studied through the use of their equivalent
discrete systems. (4) The assumption of lumped parameters is not to
replace the basic understanding of a problem, but it simplifies the analyti-
cal effort and renders a technique for the computer solution.
So far, we have discussed only systems with rectilinear motion. For
systems with rotational motions, the elements are (1) the mass moment of
inertia of the body J, (2) the torsional spring with spring constant and
(3) the torsional damper with torsional damping coefficient An angular
displacement is analogous to a rectilinear displacement x, and an
excitation torque is analogous to an excitation force The two
types of systems are compared as shown in Table 1-1. The comparison is
SEC. 1-3 Examples of Vibratory Motions
T A BLE Comparison of Rectilinear and

Rotational Systems
RECTILINEAR ROTATIONAL
Spring force= kx Spring torque =

dx
Damping force= c - . Damping torque = c,-
dt
Inertia force= Inertia =

dt dt
shown in greater detail in Tables 2-2 and 2-3. It is apparent from the
comparison that thp concept of systems can be extended easily
systems.
1-3 OF MOTIONS
To illustrate different types of vibratory motion, let us choose various

combinations of the four elements shown in Fig. 1-1 to form-simple
dynamic systems.
The spring-mass system of Fig. serves to illustrate the case of
undamped free vibration. The mass is initially at rest at its static
equilibrium position. It is acted upon by two equal and opposite forces,
namely, the force, which is equal to the product of the spring
constant k and the static deflection of the spring, and the gravitational
force mg due to the weight of the mass m. Now assume that the mass is
displaced from equilibrium by an amount and then released with zero
initial velocity. As shown in the free-body sketch, at the time the mass is
released, the spring force is equal to This is greater than the
gravitational force on the mass by the amount Upon being released,
the mass will move toward the equilibrium position.
Since the spring is initially deformed by from equilibrium, the
corresponding potential energy is stored in the spring. The system is
conservative because there is no damper to dissipate the energy. When
the mass moves upward and passes through equilibrium, the potential
energy of the system Thus, the potential energy is transformed to
become the kinetic energy of the mass. As the mass moves above the
equilibrium position, the spring is compressed and thereby gaining poten-
tial energy'from the kinetic energy of the mass. When the mass is at its
uppermost position, its velocity is zero. All the kinetic energy of the mass
has been transformed to become potential energy. Through the exchange
of potential and kinetic energies between the spring and the mass, the
system oscillates periodically at its natural frequency about its static
Free length
of spring Free-body
sketch
Static
Static
position
0
t
(a) Undamped free vibration
( b ) Damped free vibration Forced vibration
FIG. 1-3. Simple vibratory systems.
equilibrium position. Hence natural frequency describes the rate of

energy exchange between two types of energy storage elements, namely,
the mass and the spring.
It will be shown in Chap. 2 that this periodic motion is sinusoidal or
simple harmonic. Since the system is conservative, the maximum displace-
ment of the mass from equilibrium, or the of vibration, will not
diminish 'from cycle to cycle. It is implicit in this discussion that the
natural frequency is a property of the system, depending on the values of
m and k. It is independent of the initial conditions or the amplitude of the
oscillation.
A mass-spring system with damping is shown in Fig. The mass
at rest is under the influence of the spring force and the gravitational
force, since the damping force is proportional to velocity. Now, if the
mass is displaced by an amount from its static equilibrium position and
then released with zero initial velocity, the spring force will tend to
restore the mass to equilibrium as before. In addition to the spring force,
however, the mass is also acted upon by the damping force which opposes
its motion. resultant motion depends on the amount of damping in
the system. If the damping is light, the system is said to be underdamped
the motion is oscillatory. The presence of damping will cause (1) the
eventual dying out of the oscillation and (2) the system to oscillate more
slowly than without damping. In other words, the amplitude
SEC. 1-3 Examples of Vibratory Motions
with each subsequent cycle of oscillation, and the frequency of vibration

with viscous damping is lower than the undamped natural frequency. If
the damping is heavy, the motion is nonoscillatory, and the system is said
to be The mass, upon being released, will simply tend to
return to its static equilibrium position. The system is said to be critically
damped if the amount of damping is such that the resultant motion is on
the border line between the two cases enumerated. The free vibrations of
the systems shown in Figs. and (b) are illustrated in Fig. 1-4.
Ail physical systems possess damping to a greater or a lesser degree.
When there is very 'little damping in a system, such as a steel structure or a
simple pendulum, the damping may be negligibly small. Most mechanical
systems possess little damping and can be approximated as undamped
systems. Damping is often built into a system to obtain the desired
performance. For example, vibration-measuring instruments are often
built with damping corresponding to 70 percent of the critically damped
value.
If an excitation force is applied to the mass of the system as shown in
Fig. the resultant motion depends on the initial conditions as well
as the excitation. In other words, the motion depends on the manner by
which the energy is applied to the system. Let us assume that the
excitation is sinusoidal for this discussion. Once the system is set into
motion, it will tend to vibrate at its natural frequency as well as to follow
the of the excitation. If the system possesses damping, the part
of the motion not sustained by the sinusoidal excitation will eventually die
out. This is the transient motion, which is at the natural frequency of the
system, that is, the oscillation under free vibrations.
The motion sustained by the sinusoidal excitation is called the steady-
state or the steady-state response. Hence the steady-state re-
sponse must be at the excitation frequency regardless of the initial
F I G . 1-4. Free vibration of systems shown in Figs. and ( b ) .

displacement = initial velocity = 0 .
8
conditions or the natural frequency of the system. It will be shown in

Chap. 2 that the steady-state response is described by the particular
integral and the transient motion by the complementary function of the
differential equation of the system.
Resonance occurs when the excitation frequency is equal to the natural
frequency of the system. No energy input is needed to maintain the
vibrations of an undamped system at its natural frequency. Thus, any
energy input will be used to build up the amplitude of the vibration, and
the amplitude at resonance of an undamped system increase without
limit. In a system with damping, the energy input is dissipated in the
damper. Under steady-state condition, the net energy-input per cycle is
equal to the energy dissipation per cycle. Hence the of
vibration at resonance for systems with damping is Mite, and it is
determined by the amount of damping in the system.
1-4 SIMPLE HARMONIC MOTION
Simple harmonic motion is the simplest form of periodic motion. It will

be shown in later chapters that (1) harmonic motion is also the basis for
more complex analysis using Fourier technique, and (2) steady-state
analysis can be greatly simplified using vectors to represent harmonic
motions. We shall discuss simple harmonic motions and the manipulation
of vectors in some detail in this section.
A simple harmonic motion is a reciprocating motion. It can be rep-
resented by the circular functions, sine or cosine. Consider the motion of
the point P on the horizontal axis of Fig. If the distance OP is
OP = = X cos
where t = time, = constant, and X = constant, the motion of P about the

origin is sinusoidal or simple harmonic.* Since the circular function
repeats itself in radians, a cycle of motion is completed when =
* A sine, a cosine, or their combination can be used to represent a simple harmonic

motion. For example, let
= wt cos a ws sin a ) X + a)
where X = and a = It is apparent that the motion is sinusoidal
and, therefore, simple harmonic. For simplicity, we shall confine our discussion to a cosine
function.
In Eq. indicates that is a function of time Since this is implicit in the
equation, we shall omit (t) in all subsequent equations.
1-4 Harmonic Motion
FIG. 1-5. Simple harmonic motion: X cos
that is,
Period =-
1
Frequency - - or Hz*
7
is called the circular frequency measured in

If represents the displacement of a mass in a vibratory system, the
velocity and the acceleration are the first and the second time derivatives
the displacement,? that is,
Displacement x = X cos
Velocity x = - sin = + 90")
Acceleration = - cos = + (1-6)
These equations indicate that the velocity and acceleration of a harmonic
displacement are also harmonic of the same frequency. Each differentia-
tion changes the amplitude of the motion by a factor of and the phase
angle of the circular function by The phase angle of the velocity is
90" leading the displacement and the acceleration is leading the
displacement.
Simple harmonic motion can be defined by combining Eqs. (1-4) and
(1-6).
-
where a constant. When the acceleration of a particle with recti-

linear motion is always proportional to its displacement from a fixed
point on the path and is directed towards the fixed point, the particle is
said to have simple harmonic motion. It can be shown that the solution of
Eq. (1-7) has the form of a sine and a cosine function with circular
frequency equal to w.
* In 1965, the Institute of Electrical and Electronics Engineers, Inc. (IEEE) adopted new
standards for symbols and abbreviation (IEEE Standard No. 260). The unit :hertz (Hz)
replaces (cps) for frequency. Hz is now commonly used in vibration studies.
symbols x and represent the first and second time derivatives of the function
respectively. This notation is used throughout the text unless ambiguity may arise.
CHAP. 1
The sum of two harmonic functions of the same frequency but with
different phase angles is also a harmonic function of the same frequency.
For example, the sum of the harmonic motions = cos and
+ a ) is
= cos + wt cos a -sin sin a )

= + cos - sin a sin
= - sin sin
=X +
where X = + cos sin is the amplitude of the resultant
harmonic motion and = sin + cos a ) is its phase
angle.
The sum of two harmonic motions of different frequencies is not
harmonic. A special case of interest is when the frequencies are slightly
different. Let the sum of the motions and be
=2X cos- t cos

2
+-
where The resultant motion may be considered as a cosine
wave with the circular frequency + which is approximately equal
to and with a varying amplitude [2X The resultant motion is
illustrated in Fig. 1-6. Every time the amplitude reaches a maximum,
there is said to be a beat. The beat frequency as determined by two
FIG. 1-6. Graphical representation of beats.

1-5 of Motions
consecutive maximum amplitudes, is
where and are the frequencies of the constituting motions. The more
general case, for which the amplitudes of and are unequal, is left as
an exercise.
The phenomenon of beats is common in engineering. Evidently beating
can be a useful technique in frequency measurement in which an un-
known frequency is compared with a standard frequency.
REPRESENTATION OF
HARMONIC MOTIONS
It is convenient to represent a harmonic motion by means of a
rotating vector X of constant magnitude* X at a constant angular velocity
o. In Fig. 1-7, the displacement of from the center along the x axis
(a) Vectorial representation
(b) Harmonic motions
F I G . 1-7. Harmonic motions represented by rotating vector.
* In complex variables, the length of a vector is called the absolute value or modulus, and
the phase angle is called the argument or amplitude. The length of the vector in this
discussion is the amplitude of the motion. To avoid confusion, we shall use
magnitude to denote the length of the vector.
is = cos ot. This is the projection of the rotating vector X on
the diameter along the axis. Similarly, the projection of X on the y axis
is OQ = = X sin Naming the x axis as the "real" axis and the y
axis as the "imaginary" one, the rotating vector X is represented by the
equation*
X = X cos o t + sin = Xeiu'
where X is the length of the vector or its magnitude and = is called

the imaginary unit.
If a harmonic function is given as = X cos it can be expressed as
where the symbol Re denotes the real part of the
function Similarly, the function = X sin wt can be expressed as
y(t)=Im[Xe u'], where the symbol Im denotes the imaginary part of
J
It should be remembered that a harmonic motion is a reciprocating

motion. Its representation by means of a rotating vector is only a
convenience. This enables the exponential function to be used in
equations involving harmonic functions. The use of complex functions
and complex numbers greatly simplifies the mathematical manipulations
of this type of equations. In reality, all physical quantities, whether they
are displacement, velocity, acceleration, or force, must be real quantities.
The differentiation of a harmonic function can be carried out in its
vectorial form. The differentiation of a vector X is
* A complex number is of the form where is the real part and y the
imaginary part of Both and y may be time dependent. For a specific time, and y are
numbers and can be treated as a complex number. Let X in Fig. be a complex
number. The vector X is
jy = sin
where is the magnitude of the vector X. Defining = and expanding the
sine and cosine functions by Maclaurin's series, we obtain
The equation = cos 0 j sin 0 is called Euler's formula.

Preface
contributed to this field of study, and to the authors listed in the references.
We are especially grateful to Dr. James L. for his suggestions in
Chapter 9, and to K. G. Mani for his contribution of the subroutine
in Appendix C.
Francis S.
Ivan E. Morse
T.
CHAP. 1
(a) Vector addition ( b ) Vector addition with phasor notation (wt 0)
F I G . 1-9. Addition of harmonic functions: vectorial method.
with respect to Since the original motions are given along the real
axis, the sum of the harmonic motions is =X + The
addition operation can readily be extended to include the subtraction
operation.
Since both and are rotating with the same angular
only the relative phase angle of the vectors is of interest. It is convenient
to assign arbitrarily wt = as a datum of measurement of phase angles.
The vector and their sum X are plotted in this manner in Fig.
Note that the vector can be expressed as
- = cos + sin
The quantity = is a complex number and is called the complex

amplitude or phasor of the vector X,. Similarly, = in Fig. is
the phasor of the vector X.
Harmonic functions can be added algebraically by means of vector
addition. Using the same functions = cos w t and = a), +
their vector sum is
+ = +
, = (X, + cos a + sin
where
= + cos a + sin a
and
sin a
=
+ cos a
SEC. 1-5 Vectorial Representation of Harmonic Motions
Since the given harmonic motions are along the real their sum is
= = =X
In representing harmonic motions by rotating vectors, it is often

necessary to determine the product of complex numbers. The product can
be found by expressing the complex numbers in the exponential form. For
example, the product of the complex numbers and is
= (a, + jb,)
(1-13)
where A and B = are the magnitudes of the numbers
and and are their phase angles. Equation
(1-13) indicates that
Magnitude of = (magnitude of of (1-14)
Phase of = (phase of + (phase of
Obviously, the multiplication operation can be generalized to include the
division operation.
Example 1. Manipulation of Complex Numbers
(a) = 1 + = j sin 60") = =

The symbol is a convenient way of writing It represents a
vector of magnitude of two units and a phase angle of 60" or rad
counter-clockwise to the reference axis.
(c) ( + j3) =
=
(e)
A= +j = = 2/90" + 60"
-
(g)
The last two examples indicate that the multiplication of a vector by j

advances the vector counter-clockwise by a phase angle of 90" and a
division by retards it by
CHAP. 1
Since there will be a change from the English engineering (customary)

to the International System of Units (SI), the two systems of units will
co-exist for years. The student and the practicing engineer will need
to know both systems. We shall briefly discuss the English units and then
the SI units in some detai; in this section.
Newton's law of motion may be expressed as
Force = (1-16)
Dimensional homogeneity of the equation is obtained when the force is in
pounds the mass in slugs, and the acceleration in ft/sec2. This is the
English system in which the mass has the unit of A
body falling under the influence of gravitation has an acceleration of
g ft/sec2, where 32.2 is the gravitational acceleration. Hence one
pound-mass exerts one pound-force under the gravitational pull of
the earth. In other words, 1 lb, weighs one pound on a spring scale. If a
mass is given in pounds, or weight, it must be divided by g to obtain
dimensional homogeneity in Eq. (1-16).
The system is generally used in the study of vibrations. The
2
gravitational acceleration is = 386 in./sec . Hence the weight is
divided by 386 in order to obtain dimensional homogeneity in Eq. (1-16).
We assume that the gravitational acceleration is constant unless otherwise
stated. In the derivation of equations, the mass m is assumed to have the
proper units.
The International System of Units (SI) is the modernized version of the
metric system.*- SI consists of (1) seven well-defined base units, (2)
derived units, and (3) supplementary units.
The base units are regarded as dimensionally independent. Those of
interest in this study are the meter, the kilogram, and the second. The
meter m is the unit of length. It is defined in terms of the wave-length of a
krypton-85 lamp as "the length equal to 1 6 5 0 763.73 wave-lengths in
vacuum of the radiation corresponding to the transition between the
levels and of the krypton-86 atom." The kilogram kg is the unit
* A description of SI and a brief bibliography can be in the literature. A kit,
containing the above and several other publications, is obtainable from the American
Society of Engineering Education:
International System of Units Edited by C. H. Page and P. Vigoureux, US
National Bureau of Standards, Special Publication 330, Revised 1974.
Guide Orientation and Guide of (Metric) Units, 6th ed.,
United Engineering Center, 345 E. 47th St., New York, N.Y. 10017, 1975.
Some References on Information, US National Bureau of Standards, Special
Publication 389, Revised 1974.
ASEE Metric (SI) Resource Kit Project, One Dupont Circle, Suite 400, Washington, D C
20036.
, Units 17
Examples of Derived Units
SI UNIT
IN TERMS IN TERMS
OF OF OTHER
QUANTITY NAME SYMBOL BASE UNITS UNITS
Area square meter

Volume meter m3
Speed, velocity meter per second
Acceleration meter per second
squared m/s2
Density, mass kilogram per cubic
density meter
Specific volume cubic meter per
kilogram
Frequency hertz Hz
Force newton N kg .
Pressure, stress pascal Pa .
Energy, work joule
Power watt W m2 . kg
Moment of force meter newton N. m m2 . kg
TABLE Supplementary Units
SI UNIT
QUANTITY NAME SYMBOL
Plane angle radian rad

Solid steradian
of mass. The standard is a cylinder of platinum-iridium, called the

International Standard, kept in a vault at France. The second is
the unit of time. It is defined in terms of the frequency of atomic
resonators. "The second is the duration of 9 192 631 770 periods of the
radiation corresponding to the transition between the two hyperfine levels
of the ground state of the cesium-133 atom."*
The derived units are formed from the base units according to the
algebraic relations linking the corresponding quantities. Several derived
units are given special names and symbols. The supplementary units form
a third class of SI units. Examples of derived units and the supplementary
units are shown in Tables and respectively.
*Page and Vigoureux, p. 3.
CHAP. 1
TABLE 1-3. Prefixes for Multiples and Submultiples of SI Units

I
MULTIPLE PREFIX SYMBOL SUBMULTIPLE PREFIX SYMBOL
tera T deci d
G centi c
mega milli m
kilo k micro
1 hecto h n
10 deca dc P
femto f
a
The common prefixes for multiples and submultiples of SI units are

shown in Table 1-3. Examples of conversion from the English to the SI
units are given in Table 1-4. Note that a common error in conversion is to
become ensnared in too many decimal places. The result of a computa-
tion cannot have any more significant numbers than that in the original
data.
For uniformity in the use of SI units, the recommeiidations* are: (1) In
numbers, a period (dot) is used only to separate the integral part of
numbers from the decimal part. Numbers are divided into groups of three
to facilitate reading. For example, in defining the meter above, we
.. equal to 1 650 763.73 wave-lengths .. (2) The type used for
symbols is illustrated in Table 1-2. The lower case roman type is generally
used. If the symbol is derived from a proper name, capital roman type is
TABLE 1-4. Examples of English Units to SI Conversiona
TO CONVERT F R O M TO MULTIPLY BY
Inch meter (m) 2.540 000 E - 02

Pound-mass kilogram (kg) 4.535 924 E- 01
Pound-mass/inch3 (lb kilogram/meter3(kg/m3) 2.767 990 E + 04
Slug kilogram (kg) 1.459 390 E + 01
Pound-force newton (N) 4.448 222 E +
Pound-force-inch newton-meter (N . m)
1.751
pascal (Pa) 6.894 757 03
Horsepower (550 watt (W) 7.456 999 02
"The table gives the conversion from the units to the SI units. The second and
radian are used in both systems and no conversions are needed. For example, the
damping c from Table 2-2 has the units of The value of c is multiplied
by 175.1 to obtain the value of c in
*Page and Vigoureux, op. p. 10.

19
used for the first letter. symbols are not followed by a period. (3) The
product of units is denoted by a dot, such as . m shown in Table 1-2.
dot may be omitted if there is no risk of confusion with another unit
symbol, such as m but not (4) The of units may be
indicated by a a horizontal tine, or a negative power. For
m
example, velocity in Table 1-2 can be expressed as -, or m . The
must not be repeated on the same line unless ambiguity is avoided
by parentheses. For example, acceleration may be expressed as m/s2 or
m. but not (5) The prefix symbols illustrated in Table 1-3 are
used without spacing between the prefix symbol and the unit symbol, such
as in mm. Compound prefixes by the use of two or more SI
prefixes are not used.
1-7
Some basic concepts and commonly used vibration are

described in this chapter.
The idealized model of a simple vibratory system in Fig. 1-1 consists of
a rigid mass, (2) a linear spring, (3) a viscous damper, and (4) an
excitation. The inertia force is equal to the product of the mass and its
acceleration as defined by Newton's law of motion. The spring force is
proportional to the spring deformation, that is, the relative displacement
between the two ends of the spring. The damping force is proportional to
the relative velocity the two ends of the damper. An excitation
be applied to the mass other parts of the system.
If a system is the energy stored due to the initial conditions
will cause it to vibrate about its static equilibrium position. If damping is
zero. the system will oscillate at its natural frequency without diminishing
in amplitude. If the system is underdamped. the amplitude of oscillation
will diminish with each cycle and the frequency is lower that without
damping.
If a periodic excitation is applied to a system. the vibration consists of
a steady-state response and (2) a transient motion. 'The former is
being sustained by the excitation and is therefore at the excitation
frequency. The latter is due to the initial energy stored in the system and
is at its damped natural frequency. Resonance occurs when the system is
excited at its natural frequency. The amplitude at resonance is limited
only by the damping in the system.
A simple harmonic motion is a reciprocating motion as shown in Fig.
1-5. Alternatively, it can be represented by means of a sinusoidal wave or
a rotating vector as in Figs. 1-7 and 1-8. These representations are
artifices for the convenience of visualization and manipulation only. Using
these representations, it can be shown that the velocity leads the displace-
ment by and the acceleration leads the velocity by 90". A complex
amplitude, shown in Fig. 1-9, is called a phasor. It has a magnitude and a
phase angle relative to the reference vector.
A complex number has magnitude and direction. It can be added
(subtracted) by adding (subtracting) the real and imaginary parts sepa-
rately. T h e product (quotient) of complex numbers is determined by Eqs.
and
Magnitude of = (magnitude of of
Phase of = (phase of I?)+ (phase o f B)
The system is generally used in vibration. T h e gravitational

constant is 386 in./sec2. There will be a change to the International
Systems of Units (SI). T h e gravitational constant in S I units is
9.81 Examples of S I units and the conversion from the English t o
t h e SI units are given in Tables 1-2 and 1-4, respectively.
PROBLEMS
1-1 Describe, with the aid of a sketch when necessary, each of the following:
(a) Spring force, damping force, inertia force, excitation.
(b) Kinetic energy, potential energy.

(c) Free vibration, forced vibration, a conservative system.
(d) Steady-state response, transient motion.
continuous system.
(f) Natural frequency, resonance.
(g) Initial conditions, static equilibrium position.
(h) Rectilinear motion, rotational motion.
(i) Periodic motion, frequency, period, beat frequency.
Superposition.
(k) Underdamped system, critically damped system.
(I) Amplitude, phasor, phase angle
1-2 A harmonic displacement is = 10 - mm, where t is in sec-

onds and the phase angle in radians. Find (a) the frequency and the period
of the motion, (b) the maximum displacement, velocity, and acceleration,
and (c) the displacement, velocity, and acceleration at t = Repeat part
(c) for t = 1.2 s.
Problems
1-3 Repeat Prob. 1-2 if the harmonic velocity is = 150 +

1-4 A n accelerometer indicates that the acceleration of a body is sinusoidal at a
frequency of 40 Hz. If the maximum acceleration is 100 find the
amplitudes of the displacement and the velocity.
1-5 Repeat Prob. if the acceleration lags the excitation by What is the
excitation frequency?
1-6 A harmonic motion is described as =X + mm. The initial
conditions are = 4.0 mm and = 1.0
(a) Find the constants and

(b) Expressing in the form
and find the constants A and B.

1-7 Given X + = A cos + B sin find A, B, and
for each set of the following conditions:
(a) = -8.796 mm and = 10.762 mm

(b) = -8.796 mm and = -621.5
= -8.796 mm and = -10.76 mm/s 2

(d) = 4.0 mm and = -10.76 6
1-8 A table has a vertical motion with constant frequency. What is

the largest amplitude that the table can if an object on the table is to
remain in contact?
1-9 Find the algebraic sum of the harmonic motions and
Find X and a. Check the addition graphically.

The motion of a particle is described as = 4 If the motion has
two components, one of which is x, = 2 determine the other
harmonic component.
In a sketch of x versus for 0.4 s, plot the described by each
of the equations: x, = 5 sin =4 + =
3 -
1-12 A periodic motion is described by the equation
x = 5 sin + 3 sin
I n a plot of x versus sketch the motion for 1.5 s.
1-13 Repeat Prob. 1-12 if
(a) x = 5 + 3 +
(b) x = 5 + 90") 3 + 180")
1-14 Is the motion = cos +3 + periodic?

1-15 Find the period of the functions
(a) = 3 sin 3 t + 5 sin 4t

Z
(b) = 7 cos 3t
1-16 Determine the sum of the harmonic motions wt and

+ where w. If beating should occur, find the amplitude and
the beat frequency.
1-17 Sketch the motion described by each of the following equations:
(a) = +
(b) = + + 7 sin
for
1-18 Express the following complex numbers in the exponential form
(a)
1-19 The motion of a particle vibrating in a plane has two perpendicular

harmonic components: = 2 + and = 3 sin Determine the
motion of the particle graphically.
1-20 Repeat Prob. 1-19 using = 2 + and =3 sin wt.
Systems with One Degree of
Freedom-Theory
2-1 INTRODUCTION
The one-degree-of-freedom system is the keystone for more advanced
studies in vibrations. The system is represented by means of a generalized
model shown in Fig. 1-1. The common techniques for the analysis are
discussed in this chapter.
Examples of one-degree-of-freedom systems are shown in Fig. 2-1.
Though such systems differ in appearance, they all can be represented
by the same generalized model in Fig. 1-1. The model serves (1) to unify
a class of problems commonly encountered, and (2) to bring into focus the
concepts of vibration. The applications to different types of problem
will be discussed in the next chapter.
Four mathematical techniques are examined. These are (1) the energy
method, (2) Newton's l a w of motion, (3) the frequency response method,
and (4) the superposition theorem. Our emphasis is on concepts rather
than on mathematical manipulations.
Since vibration is an energy exchange phenomenon, the simple energy
method is first presented. In applying Newton's second law, the system is
described by a second-order differential equation of motion. I f , the
excitation is an analytical expression, the equation can be solved readily
by the "classical" method. If the excitation is an arbitrary function, the
motion can be found using the superposition theorem. The frequency
response method assumes that the excitation is sinusoidal and examines
the system behavior over a frequency range of interest.
Note that a system will vibrate in its own way regardless of the method
of analysis. The purpose of different techniques is to find the most
convenient method to characterize the system and to describe its be-
havior. We treat Newton's second law and the superposition theorem as
Systems with One Degree of Freedom-Theory CHAP. 2
Static
equilibrium
Spring-mass system Torsional pendulum
.......
Pulley
(c) Equivalent spring Mass-pulley-spring system

I
governor Simple pendulum
FIG. 2-1. Examples of systems with one degree of freedom.
time domain analysis, since the motion of the mass is a time function,
such as the solution of a differential equation with time as the independ-
ent variable. The frequency response method assumes that both the
excitation and the system response are sinusoidal and of the same
frequency. Hence it is a frequency domain analysis. Note that time
response is intuitive but it is more convenient to describe a system in the
frequency domain.
SEC. 2-2 Degrees of Freedom 25
It should be remarked that there must be a correlation between the

time and the frequency domain analyses, since they are different methods
to consider the same problem. In fact, superposition, which is treated as a
time domain technique, is the basis for the study of systems. The
convolution integral derived from superposition can be applied in the
time or 'the frequency domain. We are presenting only one aspect of this
very important theorem and shall not discuss methods of correlation. The
mathematical of the time and frequency analyses is not new.
Its implementation, however, was not practical until the advent of compu-
ters, instrumentation, and testing techniques in recent years.
2-2 DEGREES OF FREEDOM

The number of degrees of freedom of a vibratory system is the number
of independent spatial coordinates necessary to define its configuration. A
is defined as the geometric location of all the masses of the
system. If the inter-relationship of the masses is such that only one spatial
coordinate is required to define the configuration, the system is said to
possess one degree of freedom.
A rigid body in space requires six coordinates for its complete identifi-
cation, namely, three coordinates to define the rectilinear positions and
three to define the angular rotations. Ordinarily, however, the masses in a
system are constrained to move only in a certain manner. Thus, the
constraints limit the of freedom to a much smaller number.
Alternatively, the number of degrees of freedom of a system can be
defined as the number of spatial coordinates required to specify its
configuration minus the number of equations of constraint." We shall
illustrate these definitions with a number of examples.
The one-degree-of-freedom systems shown in Fig. 2-1 are briefly
discussed as follows:
1. The spring-mass system in Fig. has a rnass suspended from a

coil spring with a spring constant k. If m is constrained to move only
in the vertical direction about its static equilibrium position only
one spatial coordinate is required to define its configuration.
Hence it is said possess one degree of freedom.
2. The torsional in Fig. consists of a heavy disk and a
shaft of negligible mass with a torsional spring constant If the
system is constrained to oscillate about the longitudinal axis of the
shaft, the configuration of the system can be specified by a single
coordinate
*Such a system is called a holonomic system; it is the only type system considered in
this text. For a discussion on holonomic and nonholonomic see, for
Goldstein, Classical Mechanics, Addison-Wesley Publishing Inc., Mass,
1957, pp. 11-14.
26 Systems with One of Freedom-Theory CHAP.
3. The mass-spring-cantilever system in Fig. has one degree of

freedom if the cantilever is of negligible mass and the mass m is
constrained to move vertically. By neglecting the inertial effect of the
cantilever and considering only its elasticity, the cantilever becomes a
spring element. a simple spring-mass system is obtained from
the given mass m and an equivalent spring, constructed the
combination of the spring k and the cantilever.
4. The mass-pulley-spring system in Fig. has one degree of
freedom if it is ass that there is no slippage between the cord and
the pulley and the cord is inextensible. Although the system posses-
ses two mass elements m and the linear displacement of m and
the angular rotation of are not independent. Thus, either or
can be used to the configuration of the system.
5. A simple spring-loaded governor rotating with constant angular
velocity is shown in Fig. If a disturbance is applied to the
governor, its vibratory motion can be expressed in terms of
angular coordinate
The simple pendulum in Fig. is constrained to move in the xy
plane. Its configuration can be defined either by the rectangular
Cartesian coordinates and or by the angular rotation
The coordinates, however, are not independent. They are related
by the equation of constraint
L2
where the length L of the pendulum is assumed constant. Thus, if
is chosen arbitrarily, is determined from Eq. (2-1).
Several systems with two degrees of freedom are shown in Fig. 2-2.
(a) 2-mass-2-sprlng system Spring-mass system Spherical
Two-degree-of- freedom systems.

SEC. 2-3 Equation of Motion-Energy Method 27
1. The two-spring-two-mass system of Fig. possesses two degrees

of freedom if the masses are constrained to move only in the vertical
direction. The two spatial coordinates defining the configuration are
and
2. The spring-mass system shown in Fig. was described previously
as a one-degree-of-freedom system. If the mass m is allowed to
oscillate along the axis of the spring as well as to swing from side to
side, the system possesses two degrees of freedom.
3. The pendulum in space in Fig. can be described by the and
coordinates as well as by the and coordinates. The
latter are related by the equation of constraint x 2 + y 2 + z 2 = L 2 . Thus,
this pendulum has only two degrees of freedom.
2-3 EQUATION OF MOTION-ENERGY METHOD

The equation of motion of a conservative system can be
from energy considerations. If a conservative system in Fig. 2-3 is set into
motion, its total mechanical energy is the sum of the kinetic energy and
the potential energy. The kinetic energy is due to the velocity of the
mass, and the potential energy is due to the strain energy of the spring
by virtue of its deformation. Since the system is conservative, the total
mechanical energy is constant and its time derivative must be zero. This
can be expressed as
= (total mechanical energy) = constant
T o derive the equation of motion for the spring-mass system of Fig.

2-3, assume that the displacement of the mass m is measured from its
Spring
force
Free length
I
Static Static
deflection equilibrium
FIG. 2-3. Potential energy in spring.

28 Systems One of Freedom-Theory CHAP. 2
static equilibrium position. Let be positive in the downward direc-

tion. Since the spring element is of negligible mass, the kinetic energy T
of the system is
2
The corresponding potential energy of the entire system is the algebraic

sum of (1) the strain energy of the spring and ( 2 ) that due to the change in
elevation of the mass. The net potential energy of the system about the
static equilibrium is
=
I'(total spring force) dx -
= + -
Substituting Eqs. (2-4) and (2-5) into Eq. ( 2 -3 ) gives

d
dt
= + =
Since the velocity in the cannot be zero for all values of

time, clearly
where The equation of motion of the system can be expressed

as shown in Eq. (2-6) or ( 2 -7 ) .
It can be shown that the solution of Eq. ( 2 -7 ) is of the form
x = A, cos + sin
where A, and are arbitrary constants to be evaluated by the initial
conditions and It is apparent that in Eq. ( 2 -7 ) is the circular
frequency of the harmonic motion Since the components of the
solution are harmonic of the same frequency, their sum is also harmonic
and can be written as
where A = is the amplitude of the motion and = tan-'

is the phase angle.
Equation (2-9) indicates that once this system is set into motion it will
vibrate with simple harmonic motion, and the amplitude. A of the motion
will not diminish with time. The system oscillates because it possesses two
* T h e arbitrary constants A , and A, can be evaluated by conditions specified other than
at = It is customary and convenient. however, t o use initial conditions.
SEC. 2-3 Equation of Motion--Energy Method 29
types of energy storage elements, namely, the mass and the spring. The
rate of energy interchange between these elements is the natural fre-
quency of the system.*
Note that the natural frequency is a property of the system. It is a

function of the values of m and k and is independent of the amplitude of
oscillation or the manner by which the system is set into motion. Evi-
dently, only the amplitude A and the phase angle are dependent on the
initial conditions.
Example 1
Determine the equation of motion of the simple pendulum shown in Fig.
Solution:
Assume ( 1 ) the size of the bob is small as compared with the length L of t h e
pendulum and ( 2 ) the rod connecting the bob to the hinge point is of
negligible mass. The mass moment of inertia of the bob of mass about is
where is the mass moment of inertia of about its mass center. If the
bob is sufficiently small in size, then
The angular displacement is measured from the static equilibrium
position of the pendulum. The kinetic of the system is
The corresponding potential energy is = cos where
L(l 8 ) is the change in elevation of the pendulum bob. Substituting
these energy quantities in Eq. (2-3) gives
= (2-11)
The equation of motion of the simple pendulum is as shown in E q . (2-11)

or (2-12). If it is further assumed that the amplitude of oscillation is small,
then sin 8 8 and Eq. (2-12) becomes
This is of the same form as Eq. (2-7) and the solution follows. The
of oscillation of a simple pendulum is =
* I t is convenient to call the natural frequency instead of the natural circular

frequency. In the subsequent sections of the text, natural frequency will refer to or
unless ambiguity arises. Similarly, frequency may refer to f or o.
30 Systems with One of Freedom-Theory CHAP. 2
Note that, if small oscillations are not assumed, Eq. (2-11) is a

nonlinear differential equation and elliptical integrals are used for the
problem solution. The dependent variable and the independent
variable t are related by*
cos -
where and are the initial conditions at t = It is conceivable that, if

the pendulum is given a sufficient large velocity, the pendulum may
continue to about the hinge point. Thus, will increase with
time and the motion will not be periodic.
oscillations will be assumed throughout this text unless otherwise
stated. This assumption greatly simplifies the effort necessary to obtain
solution. Furthermore, the answers will be relevant for most prob-
lems, such as in predicting the onset of resonance in a vibratory system.
Example 2
Figure 2-4 shows a cylinder of mass m and radius R , rolling without
slippage on a curved surface of radius R. Derive the equation of motion of
the system by the energy method.
Solution:
The kinetic energy of the cylinder is due to its translational and rotational
motions. The translational velocity of the mass center of the cylinder is
(R - The angular velocity of the cylinder is - Since the cylinder
I
a
FIG. 2-4. Cylinder on curved surface.
*T. von and M. A. Biot, Mathematical Methods in Engineering, McGraw-Hill

Book Co., New York, 1940, pp.
SEC. 2-3 Equation of Motion-Energy Method 31
rolls without slippage, the arc and Hence the

angular velocity can be written as - The total kinetic energy T of
the cylinder is
where = is moment of inertia of the cylinder about its longitudi-

nal axis. The potential energy is due to the change in elevation of the
mass center of the cylinder with respect to its static equilibrium position,
that is,
= - - cos 8)
Substituting the and expressions into Eq. (2-3) gives
where for small oscillations. Comparing this with Eq. (2-7), the
natural frequency of the system is equal to
The natural frequency of a conservative system can be deduced by

Rayleigh's method. Natural frequency is the rate of energy interchange
between the kinetic and the potential energies of a system during its cyclic
motion. As the mass passes through the static equilibrium position, the
potential energy is zero. Hence the kinetic energy is maximum and is
equal to the mechanical energy of the system. When the mass is at a
position of maximum displacement, it is on the verge of changing direc-
tion and its velocity is zero. Correspondingly, its kinetic energy is zero.
Thus, the potential is maximum and is equal to the total mechani-
cal energy of the system. As indicated in Eq. (2-9), the motion is
harmonic when the is vibrating at its natural frequency. The
maximum displacement, or amplitude, is A and the maximum velocity is
Equating the maximum kinetic and potential energies, we have
= = total energy of the system (2-15)

or
=
Example 3. Equivalent mass of spring: Rayleigh method

The mass of the spring shown in Fig. 2-5 is not negligible. Determine
natural frequency of the system by Rayleigh's method.
Systems with One of Freedom-Theory CHAP. 2
Static
equilibrium
FIG. 2-5. Equivalent mass of spring: Rayleigh method.
Solution:
Let L b e the length of the spring k when the system is at its static
equilibrium position. Assume that, when the end of the spring has a
displacement an intermediate point of the spring has a displacement
5
equal to - Thus, defines the configuration and the system has only
L
one degree of freedom.
The kinetic energy of the system is due to the rigid mass m and the mass
of the spring k. The kinetic energy of an element of the spring of length
is , where p = of the spring. Let =A sin
Hence the maximum kinetic energy of the system is
From Eq. the maximum potential energy of the system is
The natural frequency is obtained by equating the maximum kinetic and

potential energies, that is,
This equation shows that the inertial effect of the spring can be accounted
for by adding one-third of the mass of the spring to the rigid mass m. The
natural frequency can then be calculated as if the system were to consist of
a massless spring and an equivalent rigid mass of ( m
SEC. 2-4 Equation of Motion-Newton's Law Motion
The approximate method above indicates that the natural frequency is

independent of the mass that is, the mass of the spring to that
of the rigid mass. For a heavy spring with a light mass, a larger fraction of
the spring-mass would have to be used for the frequency calculation. The
error, however, is less than one percent, as compared with exact
value, when the spring-mass is equal to the rigid mass?
2-4 EQUATION OF
LAW OF' MOTION
Newton's law of motion is used to establish the differential equation of

mqtion of one-degree-of-freedom systems in this section. The emphasis is
concepts off vibration rather than the technique for solving the equa-
tion.
The generalized model representing this class of problems is shown in
Fig. 2-6. The displacement of the mass m is measured from the static
equilibrium Displacement is positive in the downward direction,
and so are the velocity and the acceleration A positive force on
the mass m will produce a positive acceleration of the mass and
versa. Referring to the free-body sketch, the forces acting on the
are (1) the gravitational force mg which is constant, (2) the spring force
k(x+ which always opposes the displacement, (3) the damping force
which always opposses the velocity, and (4) the excitation force which
is assumed to equal to Fsin wt.
Newton's law of motion (second law) states that the rate of change of
momentum is proportional to the impressed force and takes place in the
direction of the straight line in which the force acts. If the mass is
constant, the change of momentum is equal to the mass times its
acceleration. From the free-body sketch in Fig. 2-6, the equation of
motion of the system is
k k(x +
equilibrium
Fsin
FIG. 2-6. Model of systems with one degree of freedom.
* S. Timoshenko and D. Young, Vibration Problems in Engineering, 3d ed., D . Van

Co., Inc., New York, 1954, pp.
Note that the force mg is equal to the static spring force

This is obvious in a simple problem. The implications are (1) that the
static forces must cancel in a vibratory system, and (2) that only the
dynamic forces need be considered. This concept may be helpful for more
complex problems.
The equation above can be derived whether the general position of the
mass is considered above or below the static equilibrium position or
whether the mass is moving upward or downward. Thus, the equation is
true for all time and for all positions of the mass. The verification of this
left as an exercise.
Using principle, Eq. (2-16) can be expressed as
The quantity is called the inertia force. In other words, introducing

the appropriate inertia force, we can say that the impressed force on the
mass is in equilibrium with the inertia force. Thus, the dynamic problem
is reduced to an "equivalent" problem of statics.
2-5 GENERAL SOLUTION

The equation of motion for the model in Fig. 2-6 is a second-order
linear ordinary differential equation with constant coefficients, Eq.
The general solution is the sum of the complementary function
and the particular integral as shown in Eq. App. D.
Let us consider the two parts of the solution separately before discussing
the general solution.
Complementary Function
The complementary function satisfies the corresponding homogeneous

equation
(2-21)
The solution is of the form
r
= Be S
where B and are .constants. (2-22)into (2-21)gives
Since the quantity Be" cannot be zero for all values of we deduce that
This is called the auxiliary or the characteristic of the system.

The roots the characteristic equation are
1
= (-c
Since there are two roots, the complementary function
where and are arbitrary constants to be evaluated by the initial
Let us rewrite the equations above in a more convenient form by

defining
k c c
and or (2-27)
m m
where is the natural circular of the system and is called the
damping factor. Since c, and k are positive, is a positive number.
Using the definitions of and and (2-25)become
(2-30) shows that the roots are real, distinct, and

negative, 1 Thus, no oscillation can be expected from the
complementary function in Eq. (2-26) regardless of the initial conditions.
Since both the roots are negative, the motion diminishes with increasing
time and is aperiodic.
When 1, Eq. (2-30) shows that both the roots are equal to
the complementary function is of the form
where and are constants. The motion is again aperiodic. Since the
the motion will eventually diminish to
zero.
When 1, the roots are complex conjugates.
36 Systems with One Freedom--Theory CHAP. 2
where = Defining
and using Euler's formula cos sin 8, the complementary func-

tion in Eq. (2-26) becomes
or
+ - (2-33)
Since the displacement is a physical quantity, the coefficients,

and in Eq. (2-33) must also be real. This requires
that and becomplex conjugates. Hence Eq. (2-33) can be rewritten
as .
= cos sin (2-34)
or
= (2-35)
where A , and A, are real constants to be evaluated by initial
conditions. The harmonic functions in Eq. (2-%)..are combined to give
Eq. where A and The motion
described by Eq. (2-35) consists of a harmonic motion of frequency
and an amplitude which decreases exponentially with time.
For the three cases enumerated, the type of motion described by
depends on the value of The system is said to be overdamped when
1, critically damped when = 1, and underdamped when 1. This
was explained intuitively in Chap. 1. Note that (1) is vibratory only if
the system is underdamped; (2) the frequency of oscillation is lower
the natural frequency of the system; and (3) in all cases, will
eventually die out, regardless of the initial conditions or the excitation.
Hence the complementary function gives the transient motion of the
system. As a limiting case, if the system has no damping, the amplitude of
will not diminish with time. Furthermore, Eq. (2-35) shows that the
frequency and the rate of the exponential decay in amplitude are
independent of the arbitrary constants of the equation. In other words,
they are properties of the system, independent of the initial conditions or
the by which the system is set into motion.
The critical damping is the amount of damping necessary
for a system to be critically damped, that is, 1. From Eq. when
1, we have
the damping factor can be defined as
It is a measure of the existing as compared with that necessary

for a system to be critically damped.
Example 4.
A machine of 20 kg mass (44 is mounted on springs and dampers as

shown in Fig. 2-7. total stiffness of the springs is 8 (45.7
Static
equilibrium
FIG. 2-7. vibration.
and the total damping is If the system is

initially at rest and a velocity of 100 (3.9 is imparted to the
mass, determine (a) the displacement and velocity of the mass as a time
function, and (b) the displacement at = 1.0 s.
Solution:
The is obtained by the direct application of Eq. (2-34).
The parameters of the equation are
(a) these values in Eq. we obtain
x= cos sin
x + sin
+ sin
with of CHAP.
Applying the initial conditions gives
= 100 A, = 100119.7 = 5.07

x= mm
= sin 100 cos
= + 9.5")
The displacement at t = s is
The particular integral for the excitation = F sin in Eq. (2-18) is

of the form
The values of X and can be obtained by substituting Eq. into

This is left as an exercise. It can be shown that the of
the steady-state or response is
(2-40)
+
and
k
= tan-' or (2-41)
I - w2m/k
X is the amplitude of the steady-state response and is the phase angle
of relative to the excitation F sin that is, the displacement lags
the excitation by rad. For convenience, the last two equations are
often expressed in nondimensional form. Substituting the relations =
and oclk = and defining = these equations become
and
where R is called the magnification factor and the frequency ratio of the
excitation frequency to the natural frequency of the system. Equations
(2-42) and (2-43) are plotted in Figs. 2-8 and 2-9 with as a parameter.
I 2
Frequency ratio
F I G . 2-8. factor-versus-frequency ratio; sysrem shown in
Fig. 2-6.
40 Systems of
The characteristics of the motion X due to the excitation

F sin can be observed from Eqs. (2-38) to (2-43).
1. The motion described by Eq. (2-38) is harmonic and is of the same

frequency as the excitation. For a given harmonic excitation of con-
stant amplitude F and frequency o , the amplitude X and the phase
angle of the motion are constants. Hence the particular integral
gives the steady-state response due to a harmonic excitation.
2. Since the particular integral does not contain arbitrary constants, the
steady-state response of a system is independent of initial conditions.
3. The quantity R is called the magnification factor. It is a
displacement ratio, where X is the amplitude of the steady-state
response and Flk is the corresponding displacement when 0. As
shown in Fig. 2-8, R can be considerably greater than or less than
unity, depending on the damping factor and the frequency ratio
4. At resonance, when r = = 1, the magnification factor R is limited
only by the damping in the system. This is observed in Eq. (2-42) and
Fig. 2-8 and it was explained in Chap.
5. The phase angle as shown in Fig. 2-9 ranges from to The
phase angle varies with the excitation frequency and the damping in
the system. Without damping, the phase angle can only be either or
180". At resonance, when r = the phase angle is always 90"
The interpretation of phase angle can be observed from Eq. (2-38).
- 4 )= X sin o ( t -
= sin -
where = is the time shift of relative to the excitation. In other

words, the sinusoidal displacement relative to the sinusoidal excitation is
shifted or delayed by an amount Note that phase angle is often
represented as an angle between two rotating vectors as illustrated in Fig.
1-9. Since the excitation and the response are harmonic, they can be
represented by vectors as discussed in 1-5. However, this is an
artifice, concocted for the convenience of presentation or manipulation.
This will be further examined in the next section.
Several methods are commonly used to plot (2-42) and (2-43).
The rectangular plots in Figs. 2-8 and 2-9 are intuitive. Using the
logarithmic plots in Figs. 2-10 and 2-11, it is possible to cover a wide
range of frequency, such as from 10 Hz to 3,000 Hz in vibration testing.
Correspondingly, the range of the magnification factor R, called the
dynamic range, can be presented conveniently. The logarithmic plots also
greatly facilitate the data interpretation in vibration testing.
SEC. 2-5 General 41
Frequency ratio r
FIG. 2-10. Magnification factor-versus-frequency ratio; system shown in

Fig. 2-6.
Another convenient method to present the steady-state response data

is shown in Fig. 2-12. This format is often used to present the perfor-
mance characteristics of instruments for vibration measurement. The
magnification factor R as defined in (2-42) is a displacement ratio.
Amplitude of steady-state displacement X
--
Amplitude of static displacement . (2-44)
Similarly, the velocity and the acceleration ratio can be
defined. Since the steady-state response is =X - the
steady-state velocity amplitude is and the acceleration amplitude is
Dividing these quantities by and respectively, the velocity
ratio R, and the acceleration ratio are
Frequency ratio
FIG. 2-11. angle-versus-frequency ratio; system shown in Fig.
2-6.
A combined plot of R, and versus the frequency ratio = is

shown in Fig. 2-12. The phase information is the same as before and
needs not be presented again. A t steady state, the velocity leads the
displacement by and the acceleration leads the velocity by The
steady-state response data can be presented in other convenient forms
but we shall not pursue the subject further.
General
The general solution of the equation of motion in Eq. represents
the system response to a harmonic excitation and the given initial
conditions. Assume that the system is underdamped, which is often
encountered in vibration. Substituting Eqs. (2-35) and (2-38) into
the general solution due to a harmonic excitation is
where and are calculated from Eqs. and respectively.

Note that only the constants A and are arbitrary. They are evaluated
by applying the initial conditions to the general solution in Eq.
The physical interpretation of this equation was explained in Chap. 1.
As the harmonic excitation and the initial conditions are applied to the
SEC. 2-5 Solution
Frequency ratio
F I G . 2-12. factor, velocity ratio, accebration

ratio r for various damping factor shown
in 2-6.
system, it tends to the excitation and to vibrate at its own natural

frequency. Since is sustained by the excitation, it be at the
excitation frequency. On the other hand, is not sustained by the
excitation and it is the transient motion. The frequency of the
transient motion is that of the free vibration of the system.
Example 5
Find the steady-state response and the transient motion of the system in
Example 4 , if an excitation force of 24 sin (5.4 sin is applied
to the mass in addition to the given initial conditions.
Solution:
The displacement of the mass m is obtained by the direct application of Eq.
(2-47). The system parameters are identical to those calculated in Example
Systems of Freedom-Theory CHAP. 2
4. The steady-state response from Eq. (2-42) is
= 6.0 - mm
where
The general solution is
Applying the initial conditions, we obtain
= =.A sin + 6.0 sin(-29.1")

= A(-3.25 sin 19.7 cos
Solving for A and we obtain tan-' 1.87 = 61.8" and A = Thus,

x= mm
The equation is plotted in Fig. 2-13.
FIG. 2-13. Displacement time; Example 5.

SEC. 2-6
2-6 FREQUENCY METHOD
response method is a harmonic analysis. A sinusoidal excita-

tion is applied to a system and its steady-state response is examined over
a frequency of interest. For a linear system, both the excitation and
the system response are sinusoidal of the same frequency. This can be
verified from the theory of differential equations.
The method is generally used for vibration measurement. The implica-
tion is that it is more convenient to describe a system by its Fourier
spectrum (see Chap With the advent of instrumentation and compu-
ters, the pulse technique has become a popular test procedure. The
results from pulse testing, however, are generally expressed as frequency
response data. A vast vibration measuring has been done in
the past decade or two. This field of study will gain further prominence in
the future.*
We shall discuss mechanical impedance method and sinusoidal
transfer function in this section.
Impedance
Mechanical impedance method is a harmonic analysis. the

sinusoidal in the equation of motion by means of rotating
vectors as discussed in 1-5. We shall first represent the forces in a
system by means of rotating vectors and then derive the mechanical
impedance of the system and its components.
The equation of motion of the one-degree-of-freedom system in Fig.
2-6 and its steady-state response from (2-38) are
Using the vectorial representation of hartnonic motions, the equations

above can be expressed as
* The importance of vibration measurement can be judged from a quotation: "Mechanical

Maintenance is one of the largest industries on earth. Vibration measuring has to
lower costs by as much as M. P. Blake, Monograph Inc.,
Grove, 1972.
For discussion of vibration measurement, see, for example, M. P. Blake and Wm. S.
Mitchell, Vibration and Acoustic Measurement Handbook, Spartan Books, New York, 1972
Imaginary
Real
FIG. 2-14. Displacement, velocity and vectors.
The force vector is the displacement vector is X =

and x are the phasors of and X, respectively.*
The force vector and the displacement vector X are-shown in Fig.
2-14. The corresponding velocity and acceleration vectors obtained
from X by differentiation as shown in Eq. The velocity vector is
and the acceleration vector is -w2X.The relative positions of the
vectors are illustrated in the figure.
The harmonic forces are obtained by multiplying each of these vectors
by the appropriate constants. The spring force always resists the displace-
ment. Hence the spring force vector is Similarly, the damping and
the inertia-force vectors are and respectively. These
vectors shown in Fig. Although these are rotating vectors,
their relative positions or phase angles are constant. For dynamic equilib-
rium, the vectorial sum of the forces due to the spring, the damper, and
the mass is equal and opposite to the applied force as indicated in
(2-49). Hence the force vectors form a closed polygon as shown in Fig.
Figure 2-16 shows the relation of these vectors for an excitation force
of constant magnitude but for frequency ratio r where r = The
*The phasor notation is often a source of confusion for some students. A phasor is a time
independent complex coefficient, which together with the factor gives a complex time
function.
From 1-5, a phasor is a complex amplitude or a complex number. denotes the
magnitude and phase angle a vector relative to the In this case, the force
is the reference vector and its phase angle is zero. Thus, F or where a If
given = F sin and x = X - the displacement vector is X = = XeJw'.
Hence the phasor of X is
More generally, if given = the steady-state is
X + - 4). In phasor notation. force vector is F = = and the
displacement vector is X = = Hence the phasor of is and that of X
is The relative amplitude and the phase angle between the force and the
displacement remain unchanged.
2-6 Frequency Response Method
(a) Force vectors ( b ) Polygon of vectors
FIG. 2-15. Vectors representing harmonic spring, damping, inertial, and

excitation forces.
magnitudes and the phase angles indicated may be compared with those
shown in Figs. 2-8 and 2-9. Since the interest is in the relative amplitudes
and phase angles of the vectors, the vectors are rotated clockwise by an
amount (wt - 4). This is equivalent to choosing (wt - as the datum of
measurement.
The is obtained by substituting Eq. in (2-49).
Factoring out the term, we get
+ + = (2-51)
or
= - (2-52)
k - w2m +
where
F
(2-53)
- k
where R is the magnification factor defined in Eq. (2-42).
(a) 1.0 r = 2.0

F I G . 2-16. Polygon of force vectors for frequency ratio 1: constant
and = 0.25.
T A BL E 2-1. Impedance of System

Elements
ELEMENT SYMBOL
Mass m --w2m
Damper c
Spring k k
The quantity (k - w 2 m + in Eq. (2-51) is called 'the mechanical

impedance of the system in Fig. 2-6. It has the dimension of force per unit
displacement. The definition follows conveniently from Newton's law of
motion in Eq. (2-51). Similarly, for the elements m, c, and k, the
corresponding mechanical impedances are defined as - 0 2 m , and k,
respectively. These are tabulated in Table 2-1. In the literature, mechani-
cal impedance is also defined as force per unit velocity, although this
definition is by no means standardized.*
*Mechanical impedance was defined by analogy from law. Let us briefly examine
the analogy.
Electrical impedance was defined from the generalization of Ohm's law RI, where V
is the voltage drop across the resistor R and I the current flow through R. The generaliza-
tion is where Z is the impedance of a component or a network.
Let us rewrite Eq. (2-49) and compare it with the RLC circuit in series and in parallel
shown in Fig. 2-17.
where and are the sources of force, voltage, and current, respectively. Since
all the equations above are of the same form, either the force-voltage analogy or the
force-current analogy can be used to define mechanical impedance.
Using the force-voltage analogy, mechanical impedance is defined as Using
the force-current analogy, mechanical impedance is defined as
The electrical circuits in Fig. 2-17 are self-explanatory. The mechanical "circuit" shows
that (1) the excitation force is the sum of the inertia force, the damping force, and the spring
force, and (2) the mass, spring, and damper have the same velocity at a common junction.
Hence the diagram represents the mechanical system.
examination of the diagrams reveals that, if the force-current analogy is used, the
mechanical circuit can be obtained directly from the electrical. In other words, if (1) force is
analogous to current, (2) velocity analogous to voltage, and mechanical impedance defined
as then both the equations and the circuits are analogous.
The force-voltage analogy is intuitive. 'The force-current analogy has the advantages
mentioned above. Furthermore, a force acts through a component; the forces at both ends
of a spring are equal. Note that a current flows through a component. Hence force and
current are both through variables. The velocity is measured across a component. Note also
that a voltage is measured across a component. Thus, velocity and voltage are both across
variables. Using this concept and the force-current analogy, the electrical and mechanical
circuits should be alike.
C Ref:
RLC in b, PLC in parallel circuit
FIG. 2-17. Comparison of and circuits.
Use the impedance method to find the steady-state response of the system
described in Example 5.
The of the system is

(k -- 0 2 m ) +
=
From (2-52) we have
which the or complex amplitude of the displacement vector X.
Since the given excitation force is F sin which is equal to the

displacement is which is X -
Transfer
The transfer function is a mathematical model defining the input-output
relation of a physical system. If the system has a single input and a single
it can be represented by means of a block diagram shown in Fig.
-18. The response is caused by an excitation Naming
Input I transfer Output

excitation) (response)
FIG. 2-18. Block diagram of linear systems.

as the output and the input, this causal is specified by
the transfer function.
(Output) = (transfer
or
= (transfer function)
Consider the system in Fig. 2-6 as an example. The equation of motion

is
Using the impedance method and substituting for the derivative

in the equation, we obtain
The symbol indicates that G is a function of Similarly, is a

function of o, that is, the symbol does not indicate a product of and
is the sinusoidal transfer function of the system.
Comparing Eqs. (2-57) and it is evident that the transfer
function is another technique to present the frequency response data of a
system. Moreover, the data in Figs. 2-8 to 2-12 are the nondimensionai
plots of the sinusoidal transfer function. The transfer function of a complex
system can be obtained from test data. Thus, a system can be identified
by data from its frequency response test.
Note that the transfer function defined in Eq. (2-55) is an operator. It
operates on the input to yield an output. It is often called a ratio of
output per unit input. This is not a ratio in the normal sense of the word
as a ratio of two numbers. As illustrated in Eq. the transfer
function is a complex number. Furthermore, it is dimensional, such as
displacement per unit force. It is more appropriate to think of a transfer
function simply as an operator.
Determine the frequency response of the system described in Example 5 by

means of its transfer function.
Solution:
From Eq. (2-57) the transfer function is
SEC. 2-6 Method
Hence = 0.25 Since the excitation is 24 sin the

F
magnitude of the displacement is
Damping, and Bandwidth
It is observed in Fig. 2-8 that the height of the resonance peak is a

function of the damping in the system. One of these frequency response
curves is reproduced in Fig. 2-19. It can be shown that the peak of the
resonance curves occur at = If 1, the peaks occur at r
Thus, from Eq. (2-42) the value of the maximum amplification factor is
The damping in a system is indicated by the sharpness of its response

curve near resonance and can be measured by the bandwidth. The
bandwidth is as shown in Fig. 2-19, where = is a frequency
ratio and and are the frequency ratios at the half-power The
amplification factor R at and is R = Substituting this in Eq.
(2-42) and letting 1/25 shown in Eq. we obtain
FIG. 2-19. response showing bandwidth and

power points.
* This terminology is commonly used in electrical engineering. The power dissipation P in

a resistor R is P , = At half power, = = Thus, I, =
52 Systems One of CHAP. 2
Assuming 0. solving for we get
Bandwidth = - (2-60)
A Q is also used to define the bandwidth and damping.

1 1
--
(I bandwidth 25
Q is used to measure the quality of a resonance circuit in electrical

engineering. It is also useful for determining the equivalent viscous
damping in a mechanical system.
2-7 TRANSIENT VIBRATION

We shall show that the transient vibration due to an arbitrary excitation
can be obtained by means of superposition. Although the method is
not convenient for hand calculations, it can be implemented readily using
computers. ,
The equation of motion of the model in Fig. 2-6 for systems with one
degree of freedom and an excitation is
One method to solve the equation approximate by a sequence

of pulses as shown in Fig. If the to a typical
pulse input is known,-the response to can be obtained by superposi-
tion. In other words, the system response to is the sum of the
responses due to each of the pulses in the sequence.
0 Time 0 Time
Sequence of pulses (b) Sequence of steps
FIG. 2-20. approximated by pulses and steps.
The response due to a unit impulse input with zero initial
conditions its A rectangular of duration
or width and height is shown in Fig. area of this
is unity. T o obtain a unit let the pulse approach
zero while the pulse area at unity. the we have a unit
impulse as by the relations
for
=1
This at = as shown in Fig. a unit impulse

occurs at = T, it is defined by the relations
for
Note that - a unit impulse translated along the positive time axis
by an amount T.
function, not necessarily a rectangular pulse. the
above can be used as a unit impulse and is called the delta function.
Mathematically. a unit impulse must have zero pulse unit area, and
infinite height. It seems that an impulse cannot be in applications.
In pulse testing of real systems, however. an excitation can be considered
,
as an impulse if its duration is very short compared with the natural
period of the system.
From Eq. the equation of motion with an excitation =
Assume that the system is at rest before the unit impulse: is applied,
Time 0
Rectangular unit area
FIG. 2-21. Rectangular puke and unit

Systems Degree of CHAP. 2
that is,
Since is applied at = it is over with at Thus, ( 1 ) the system

becomes unforced for and ( 2 ) the energy input due to becomes
.the initial conditions at =
T o find the initial conditions at = we integrate Eq. (2-65) twice for
t Thus,
dt+ dtdt
From Eq. the first integration of gives a constant and the

'second integration for is zero. Hence the right side of the
equation is zero. If does not become infinite, its integration over this
infinitesimal interval is also zero. Thus,
If = as indicated in Eq. we have =

Now, integrating Eq. (2-65) once for t
From Eq. the right side of this equation is unity. The third
on the left side is zero if does not become infinite. The second term is
zero as explained above. Thus,
For = = in Eq. the initial conditions at t = 0' due to a

unit impulse at = are
and (2-67)
The homogeneous equation equivalent to Eq. (2-65) is
with the initial conditions = and = This deduction is

almost intuitive, since an impulse would cause a momentum change. If
is constant and = = an impulse would cause a change in the
initial velocity.
It can be shown from the solution of Eq. (2-68) that the impulse
response is
1
for
If a unit impulse T) occurs at = as shown in Fig. the

SEC. 2-7
response is delayed by an amount T, that is,

1 (2-70)
where h(t - T) = for t T.
Convolution
Let an excitation be approximated by a sequence of pulses as
shown in Fig. The strength of a pulse is defined by the pulse area.
strength of a typical pulse in the sequence at time is the area
system response to a typical pulse is the product of its unit
impulse response and the pulse strength, that is, By
superposition, we sum responses due to each of the pulses in the
sequence and obtain
As A T approaches zero, the summation becomes the convolution integral
6' sin -
This is the system response for the input with zero initial conditions.
An alternative form of the integral is
In other words, the response of a linear system to an arbitrary excitation

is the convolution of its impulse response and the excitation.
statement is known as theorem.
If the initial conditions are not zero, the complete solution is obtained
by the superposition of the particular solution due to the excitation and
the complementary solution due to the initial conditions. Substituting the
initial conditions and = into Eq. (2-34) gives the com-
plementary solution
+
sin
"
d
particular solution is shown in Eq. (2-71). By superposition, the

complete solution is
= cos + + sin
"
d
56 with One Degree of Freedom-Theory 2
The convolution integral is a powerful tool in the study of linear

systems. Although Eq. (2-71) cannot be conveniently applied by hand
calculations, it can be implemented readily using computers. The example
to follow is not indicative of the amount of algebraic computation
involved in applying the convolution integral by hand calculations.
Example
A box shown in Fig. 2-22 is dropped through a height H. Find the
maximum force transmitted to the body m when the box strikes the floor.
Assume there is sufficient clearance between m and the box to avoid
contact.
Box
FIG. 2-22. Drop test.
Let be the relative distance between m and the box and = the
time for the box to strike the floor. Assume that on striking the floor the
box remains in contact with the floor. Let us consider the time interval for
the free fall and that after striking the floor separately.
During the free fall, the absolute displacement of m is ( x Hence the
equation of motion of m is
or
where
1 2
= or
Hence the equation of motion becomes
If the box is initially at rest before the fall, we have zero initial conditions.
Applying Eq. (2-74) where = we get
*R. D. Mindlin, "Dynamics of Package Cushioning," Bell Tech. Jour., 24, (July
1945) pp. 353-461.
SEC. 2-7 Transient Vibration
The expression for h(t - is obtained from Eq. (2-70). Since the system is
undamped, the equation above becomes
From the time that the box strikes the floor at t = the system becomes
unforced. Redefining the time from the instant of impact, the initial
conditions are
= = (1 - cos
where and are obtained from the above and is the

velocity of the box assembly at = Applying Eq. (2-7.4) with =
gives
The maximum force transmitted to m is X is the

amplitude of Thus, the maximum force is
Force = cos + sin
The system response due to a unit step input with zero initial conditions
is called the response. A unit step function shown in Fig.
has the property
1 for t > O
=
for
1 -
0 Time t Time
(a) Unit step function (b) Unit step function -
F I G . 2-23. Unit step functions.

58 Systems One Degree of Freedom-Theory CHAP. 2
A unit step translated along the positive time axis by an amount to

become - is shown in Fig.
1
for
An arbitrary function can alternatively be approximated by a
sequence of steps as illustrated in Fig. Following the steps
enumerated for the impulse response, it can be shown that the
response is
where 4 = sin-' The system response to an arbitrary input is

the superposition of the responses due to each of the individual steps.
Thus,
where is the time derivative of estimated at = The term

is to account for the step at t = since the slope does not take
into account Equation (2-78) is referred to as Duhamel's integral or
the superposition integral.
2-8 COMPARISON OF AND

ROTATIONAL SYSTEMS
The discussions in the previous sections centered on systems with

rectilinear motion. The theory and the interpretations given are equally
applicable to systems with rotational motion. The analogy between the
two types of motion and the units normally employed are tabulated in
Table 2-2. The responses of the two types of systems are compared in
Table 2-3.
Extending this analogy concept, it may be said that systems are
analogous if they are described by equations of the same form. The
theory developed for one system is applicable to its analogous systems. It
will be shown in the next chapter that our study of the generalized model
for systems with one degree of freedom is applicable to a large number of
physical problems, the appearance of which may bear little resemblance
to one another.
2-9 SUMMARY
The number of degrees of freedom of a system is the number of spatial

coordinates required to specify its configuration minus the number of
TABLE 2-2. Analogy between Rectilinear and Rotational Systems
RECTILINEAR SYSTEM ROTATIONAL SYSTEM

UNIT UNIT
QUANTITY SYMBOL ENGLISH UNIT SYMBOL ENGLISH SI UNIT
Time t t
Displacement x in. 8 rad rad
Velocity
Acceleration in./sec2 m/s2 rad/sec2 rad/s2
Mass, moment of
inertia m2 . kg
Damping factor m . Nlrad
Spring constant m
Force, torque m N = m2
Momentum kg .
Impulse m2 . kg
Kinetic energy J
Potential energy J
Work
Natural frequency = =
TABLE 2-3. Response of Rectilinear and Rotational Systems
--
ITEM SYSTEM SYSTEM
System
Equation of motion + cx + k x =
System response x= +
Initial conditions = x,,,
Complementary function,
= Ae sin + sin +
= = =
Particular integral,
(1) = F sin wt = X -
2
r= tan-" -r ) r= = tan r 2)
1 1
= e sin = sin
(3) arbitrary = -r )
SEC. 2-9 61
equations constraint. Many practical problems can be represented by

systems with one degree of freedom, the model of which is shown in Fig.
The methods of study in this chapter can be broadly classified as the

time and the frequency response methods.
The energy method treats the free vibration of a conservative system.
The total mechanical energy is the sum of the kinetic T and the
potential energy Since the total energy ( T + is constant, the
equation of motion is derived from
method assumes that (1)the motion is sinusoidal, and ( 2 ) the

maximum kinetic energy is equal to the maximum potential energy. Thus,
if the mass--of a spring is not negligible, an equivalent mass of the
spring can be calculated from its kinetic energy. the natural
frequency of the system is = +
The equation o f motion from Newton's law is
All the static forces can be neglected if is measured from the static
equilibrium position of the system. The general solution by the "classical"
method is
where is the complementary function the transient

motion and the particular integral due to the excitation. The roots of
the characteristic equation from Eq. (2-24) or (2-29) dictate the form of
This gives the "natural" behavior of the system. If the system is
underdamped, is sinusoidal with exponentially decreasing amplitude
as shown in Eq. (2-35).
If the is harmonic, is harmonic and at the excitation
frequency. The harmonic response, described by Eq. is shown
graphically in Figs. 2-8 to 2-12. Many schemes can Ibe used for this
graphical presentation.
The general solution in Eq. (2-47) shows that the arbitrary constants A
and only in They are evaluated by applying the initial
conditions to the general solution, since it is the entire solution that must
satisfy the initial conditions.
In the response method, vectors are used to represent the
sinusoidal functions in the equation of motion. Denoting the excitation by
the vector and substituting for in the equation of motion,
the amplitude X and the phase angle of the steady-state response
62 of CHAP. 2
=X - are as shown in Eqs. (2-53) and (2-54). The mechani-

cal impedance and the sinusoidal transfer function methods are variations
of this technique, although they are very important in vibration measure-
ment.
An arbitrary excitation can be approximated by sequence of
pulses as shown in Fig. The system response due to is the
sum of the'responses due to the individual pulses. In other words, if the
impluse response in Eq. (2-69) of a system is known, its response due to
can be obtained by superposition. This gives the convolution integral
in Eq. (2-71).
Zero initial conditions are generally assumed when applying the con-
volution integral. If the initial conditions are not zero, the complementary
solution (2-73) due to the initial conditions is added directly to
yield the complete solution in Eq. (2-74). In contrast, the classical method
first obtains the general solution of the nonhomogeneous equation and
then evaluates the constants of integration by applying the initial condi-
tions to solution.
The theory discussed in this chapter is equally applicable to systems
with rectilinear and rotational motions. The two types of systems are
compared in Tables 2-2 and 2-3.
PROBLEMS
Assume all the systems in the figures to follow are shown in their static
equilibrium positions.
2-1 Use the energy method to determine the equations of motion and the
natural of the systems shown in the following figures:
(a) Figure Assume the mass of the torsional bar k, is negligible.

(b) Figure Assume there is no slippage between the cord and the
pulley.
(c) Figure Consider the mass of the uniform rod L.
(d) Figure Assume there is no slippage between the roller and the
surface.
(e) Figure Assume there is no slippage between the roller and the
surface. Neglect the springs and let the springs be under initial
tension.
(f) Repeat part e. including springs k, and all the springs are
under initial compression.
(g) Figure Assume there is no the pulley and the
cord.
Problems 63
Roller
FIG. P2-1. Vibratory systems.
Figure Assume m,.
(i) Figure The U tube is of uniform cross section.

(j) Figure The cross sectional areas are as
2-2 A connecting rod of 2.0 kg mass is suspended on a knife edge as shown in

Fig. If the period of oscillation is 1.2 s, find the moment of
inertia of the rod about its mass cg.
Knife
edge
Rod
+
(a) Connecting rod and segment

F I G . P2-2. Pendulums.
2-3 A counter weight in the form of a circular segment as shown in Fig.

is attached to a uniform wheel. The mass of the wheel is 45 kg and that of
the segment 4 kg. The wheel-and-segment assembly is swung as a pen-
dulum. If = 250 mm, R, = 230 mm, and L = 500 mm, find the period of
the oscillation.
rnaa ............
..........
(a) Cantilever supported beam

F I G . P2-3. frequency of beams.
2-4 A uniform cantilever beam of is shown in Fig.

Assume that the beam deflection during vibration is the same as its
deflection for a concentrated load at the free end, that is,
(a) Determine the natural frequency of the beam. ( b )
Define an equivalent mass at the free end of the beam for this mode of
vibration.
2-5 Repeat Prob. 2-4 if the deflection curve is assumed as = What is
the percentage error in the natural frequency as compared with Prob. 2-4?
Note that the assumed deflection curve does not satisfy the boundary
condition at the fixed end, since the slope at the fixed end must be zero.
2-6 Repeat Prob. 2-4 if a mass m is attached to the free end of the cantilever.
2-7 A simply supported uniform beam with a mass m attached at is
shown in Fig. The mass of the beam is p Assume that
the deflection during vibration is the same as the static deflection for a
concentrated load a t that is, x = - for 0
(a) Find the fundamental frequency of the system. ( b ) What is the
equivalent mass of the beam at
2-8 Repeat Prob. 2-7 if the deflection curve is assumed for
2-9 A bar of p with an attached rigid mass m is shown in

Fig. Assume the elongation of the bar is linear, is, =
Find the frequency for the longitudinal vibration of the bar.
Bar
FIG. Fundamental frequency of bars.
2-10 A uniform bar of p is shown in Fig. Assume the

maximum deflection of the bar is due to its own weight. Find the fundamen-
tal frequency for the longitudinal vibration.
2-11 Repeat parts (a) to of Prob. 2-1, using Newton's law of motion.
2-12 Referring to Fig. 2-4, let the cylinder be of mass m = 45 kg, =
and 500 mm. (a) Derive the equation of motion by means of Newton's
second law. (b) Find the natural frequency.
2-13 Find the solutions of the homogeneous equation x + = for the following
initial conditions:
(a) = 1 and =0
(b) = and =2
2-14 Find the solutions of the homogeneous equation +4x = for the
following initial conditions:
(a) = and x =
and 2
(c) = and 2
2-15 Assuming the initial conditions = = 0, find the solution of each of

the following nonhomogeneous equations:
(a) x +4x + =
(b)
(c) x +4x + = sin 4t

66
2-16 The equations of motion are given as + + kx = F + a),and

m i t + kx F + Derive the steady-state response of each of
these equations by the method of undetermined coefficients.
2-17 A machine of 20 kg mass is mounted as shown schematicallyin Fig. 2-7. If
the total stiffness of the springs is 17 the total damping is
300 N find the motion for the following initial conditions:
= 25 and =
= 25 mm and = 300
=0 and = 300
2-18 Repeat Prob. 2-17 if an excitation force 80 N is to the mass

of the system.
2-19 An excitation of 20 - 30") N is applied to the mass or a mass-spring
system with m = 18 kg and k = 7 (8) Find the motion of
for the initial conditions = = (b) Repeat part if the
the system is c 200 N .
2-20 The equation of motion of the system in Fig. 2-6 is + + = F sin
Represent the forces by rotating vectors and indicate the positions of the
vectors for the following conditions:
(a) m is moving downward and it is below the equilibrium

(b) is moving upward and it is below the equilibrium
m is moving upward and it is above the equilibrium
(d) m moving downward and it is above the equilibrium
2-21 The equations of motion are given as (a) + cx = F + + and

(b) mx + cx kx F +
Find the steady-state response for each of
the equations by the method of mechanical impedance.
2-22 Find the steady-state response of the system described in Prob. 2-14 if an
excitation force F = 20 cos N is applied to the mass of the system.
2-23 A mass-spring system with damping has m = 2 kg, c = 35 k
4 and an excitation F = 30 cos N applied to the mass. Use the
mechanical impedance method to find the steady-state amplitude X and the
phase angle for each of the following excitation frequencies:
(a) =6
(c) = 120
2-24 Derive the equations of motion for each of the systems shown in Fig.
Derive expressions for the steady-state response of the systems by the
mechanical impedance method.
Problems 67
X, sin = sin
2
F I G . P2-5. Vibratory systems.

2-25 A constant force is applied to an underdamped mass-spring system at t =
Assuming zero initial conditions, (a) derive the equation for the response
(b) find the time at which the first peak of occurs, amd (c) derive an
equation relating this peak response and the damping factor of the system.
2-26 Consider the equation + = where 7 and are If =
find the solution (a) by the method of undetermined coefficients, and (b) by
the integral in Eq. (2-72). (c) Repeat the problem for = x,.
2-27 Repeat 2-26 for the equation + = Ct.
2-28 Given thie equation of motion of an undamped system
+ kx = or x+ =
derive the equation for the transient response shown in Eq. (2-74) by
( 1 ) multiplying the equation above by sin and (2) integrating by
parts for 0 7 t, that is,
(t) cos +- sin + sin ( I! -
2-29 Given equation of motion of an

+ + kx = or x+ + =
derive the equation for the transient response shown in Eq. (2-74) by
( 1 ) multiplying the equation above by sin - (2)
ing by parts for 0 t, that is,
+ sin -
2-30 Given the system equation and the initial conditions

and
find the transient response by means of Eq. (2-74). Clheck the answer
by means of the classical method.
2-31 Assuming zero initial conditions, find the transient response of a system
described by the equation
= At
by means of Eq. where A =constant. Use the method to
check thie answer.
68 Systems with One Degree of Freedom-Theory CHAP. 2
Computer problems:
2-32 Use the program in Fig. to find the transient
response of the system
+ cx + kx =
Let be as shown in Fig. Choose values for k, and
Assume appropriate values for the initial conditions and Select about
two cycles for the duration of the run and approximately twenty data points
per cycle.
Consider the problem in three parts as follows:
= = and =
(b) = and =
(c) = and =
Verify from the computer print-out that in part c is the sum of the
parts a and b. In other words, this is to demonstrate Eq. (2-74) in which the
response due to the initial conditions and the excitation can be considered
separately.
(a) Rectangular pulse Step input with time A half sine pulse
FIG. Excitation forces.
2-33 Repeat Prob. 2-32 for the excitation shown in Fig.
2-34 Repeat Prob. 2-32 for the excitation shown in Fig.
2-35 Select any transient excitation and repeat Prob. 2-32.
2-36 It was shown in the pendulum problem in Example 1 that the equation of
motion is nonlinear for large amplitudes of vibration. Consider a variation of
the pendulum problem in Eq. (2-11).
+ + sin = torque (t)
+ sin =
where is a viscous damping factor and a constant torque applied to

the system. Select values for and the initial conditions and
Using the fourth-order Runge-Kutta method as illustrated in Fig.
write a to implement the equation above.
2-37 Repeat Prob. 2-36 but modify the program for plotting, as illustrated in
Fig. (b) Plot the results using in Fig.
Systems with One Degree of
Freedom-Applications
INTRODUCTION
'This chapter is devoted to the application of the theory developed in
Chap. 2 to a large class of problems, the appearance of which may differ
appreciably from that of the generalized model in Fig. The emphasis
is on problem, formulation and the generalization of each type of system.
The approach is to reduce the equation of the system to the form of a
one-degree-of-freedom system shown in Eq. that is,
where and are the equivalent mass, damping, coeffi-

cient, spring constant, and excitation force, respectively.* Once the equa-
tion is developed, the interpretation follows the general theory discussed
in the last chapter.
The equivalent quantities in Eq. (3-1) may be self-evident for simple
problems. example, the equivalent spring force is that which
tends to restore the mass to its position. The restoring force
can be due to a spring, gravitation, the buoyancy of a liquid., a centrifugal
field, or their combinations. Alternatively, from considerations,
is a quantity in the total potential energy of the system due to a
displacement in the direction. Similarly, is a quantity in the total
kinetic energy of the system due to The accounts for all the
energy dissipation associated with x. could be to a force
*For convenience of writing, the subscript is omitted from equations

unless ambiguity arises.
70 Systems with One of Freedom-Applications CHAP. 3
a motion applied to the system or an unbalance in the machine.

The product of and the displacement x has the unit of work."
The generalized model shown in Fig. 2-6 consists of four elements,
namely, the mass, the damper, the spring, and the excitation. The systems
considered in this chapter are grouped according to the elements in-
volved. If a system does not possess one of these elements, such as a
damper, it is simply omitted from the equation of motion in Eq. (3-1). We
shall begin with the simple mass-spring system.
UNDAMPED FREE VIBRATION
The simplest vibratory system is one that consists of a mass and a

spring element. If a system is lightly damped, it can be approximated by a
simple spring-mass system. Neglecting the damper and the excitation, Eq.
(3-1) becomes
From Eq. the solution of Eq. (3-2) is

= A, cos + sin
where A, and A, are constants. Substituting the initial conditions
and = gives
= cos +- sin
where
and = (3-5)
Example 1. Equivalent Mass

A machine component at its static equilibrium position is represented by a
uniform bar of mass m and length L and a spring k in Fig. Derive
the equivalent system shown in Fig.
Solution:
The equivalent mass is obtained by considering the kinetic energy T of
the system as illustrated in Example 3, Chap. 2. Assuming the spring is of
Although the concept of equivalent quantities may not be fully utilized in this, chapter,
they are introduced early in the text because (1) the one-degree-of-freedom system is basic
in vibration, and (2) the concept of equivalent or generalized quantities is essential for more
advanced studies in later chapters.
SEC Undamped Free Vibration
3-1. mass
negligilble mass, we get
-
where
= = + =
Substituting in the kinetic energy equation yields
Example 2. Equivalent Mass Moment of Inertia

A pinion-and-gear assembly is shown in Fig. 3-2. It is convenient to
refer the mass moment of inertia of the assembly to a common shaft. Find
the of the assembly referring to the motor shaft.
Solution:
Let be the number of teeth on the pinion and that of the gear. The
gear is n = Let and be the angular rotations of the pinion
and the gear respectively. The kinetic energy T of the assembly
referred to the motor shaft is
Substituting = in we get
Hence the equivalent mass moment of inertia of referring to the

shaft 1 is n 2J 2.
FIG. 3-2. Equivalent mass moment of inertia

Systems One of CHAP. 3
Static
(a) Vibratory system Equivalent
FIG. 3-3. Equivalent spring.
Similarly, equivalent springs can be calculated. Let be the torsional

of shaft 2. It can be shown by equating potential energies that the
equivalent stiffness of shaft 2 referring to shaft 1 is n 2k r2. The equivalent
spring in a system can assume various forms. We shall illustrate the
equivalent spring with the examples to follow.
Example 3.
Figure 3-3 shows that the static deflection of a cantilever beam is due to
the mass m attached to its free end. Find the natural frequency of the
system.
Solution:
The equivalent system is as shown in Fig. if (1) the cantilever is of
negligible mass and (2) m is small in size compared with L. The static
due to the force mg at the free end of a beam of
length L is
here EI is the flexural of the beam. The equivalent spring

is defined as force per unit deflection.
From the equivalent system, the natural frequency is
Example 4. Springs in
Springs are said to be in series when the deformation of the equivalent
spring is the sum of their deformations. Assume the cantilever in Fig.
is of negligible mass. Show that the cantilever and the spring are
in series.
SEC. 3-2 Free
(a) Vibratory system (b) and ( c ) Equivalent systems
FIG. 3-4. Springs in series.
T h e cantilever can be replaced by an equivalent spring of spring constant

as in Example 3. The equivalent system is as shown in Fig. A
unit static force applied at m in the x direction will the and
to elongate by and respectively. The corresponding elongation
of the equivalent spring is Thus,*
The system in Fig. consists of a torsional shaft an extended

arm and a spring in series. If the mass of the shaft and its arm are
negligible, this system reduces to that of Fig.
Example 5. Springs in Parallel

Springs are said to be in parallel when ( 1 ) the equivalent spring force is the
sum of the forces of the individual springs, and (2) the springs have the
same deformation. .A disk is connected to two shafts shown in Fig.
(a) Show that the shafts are equivalent to springs in parallel
(b) Determine the frequency of the system for torsional vibrations.
* Students often wonder why the equation for springs in series is like electrical impedance
in parallel, where Z = V = voltage, and I = current. Referring to the discussion on
analogy in 2-6 and using the current-force analogy, we define the impedance of a spring
= where =deformation, force, and k =spring constant. It can be shown
in a mechanical "circuit" that the impedance of the springs in Fig. 3-4 are in series. Thus,
1 1 1
or
Equation (3-7) for springs in parallel can be explained in the same manner.
Systems with .. CHAP. 3
.,
......................
Or-
(a) Vibratory system Equivalent system

F I G . 3-5. Springs in parallel.
Solution:
(a) Since both shafts tend to restore to its equilibrium position, the
equivalent system is as shown in Fig. The restoring torque of a
circular shaft is
where G is the shear modulus and d and are the diameter and length
of the shaft, respectively. If the disk J is rotated by an angle 8, the
restoring torque T is the sum of the restoring torques of the individual
shafts.
= (k,, +
or
(b) From the equivalent system, the natural frequency is
Example 6. Effect of Orientation

The equivalent spring force could be due to a combination of the spring and
gravitational forces. Determine the equation of motion of the systems
shown in Fig.
Solution:
Owing to the difference in the orientations of the systems, the restoring
torque due to gravitation on the mass is different for the three systems.
Assuming small oscillations and taking moments about 0, the equations of
motion are
(a) = (torque),
=- sin - (ka sin cos 0)
+
SEC. 3-2 Undamped Free Vibration
F I G . 3-6. of orientation.
(b) = sin cos

+ =
(c) = sin - (ka sin cos 8 )

+ (ka 2
- =
Note that the quantity mg does not appear in all the equations in the
examples above. Furthermore, the pendulum could be at the slant; that is,
its static equilibrium positions need not be vertical or These
considerations are left as home problems.
The next two examples are variations of the pendulum problem in
which the equivalent spring force is due to gravitation a centrifugal
field.
Example 7
The equation of motion of the system in Fig.. 2-4 was derived by the energy
method in Example 2, Chap. 2. Derive the equation of by Newton's
law of motion.
Solution:
Since 8, is the rotation of the cylinder relative to the curved surface, the
rotation of the cylinder is Using the relation =
and moments about the instantaneous center of rotation b, the
equation of is'
= sin
sin 8 =
where
76 Systems with One of Freedom-Applications CHAP. 3
Assuming sin 0 0 gives
Example 8. Effect of Centrifugal Field

A helicopter blade and rotor assembly is shown in Fig. Make the
necessary assumptions to simplify the problem and deduce the equation of
motion for the flapping motion of the blade.
Solution:
Assume (1) the blade of mass m is a uniform bar hinged at 0, (2) the rotor
angular velocity is constant, and (3) the gravitational field is negligible
compared with the centrifugal tield. Each element of the blade is
subjected to a centrifugal force where p is the of the
blade. The corresponding moment about is sin Since
= R + cos the total moment is
+ 0) =
where sin 0 1, and m = The mass moment of inertia of the

blade about is Taking moments about gives
Rotor Blade
(a) Rotor and blade Force analysis

FIG. 3-7. Helicopter rotor and blade.
3-3 DAMPED-FREE VIBRATION
T o a greater or lesser degree, all physical systems possess damping. For
free vibration with damping, Eq. (3-1) becomes
+
If the system is underdamped, from Eq. (2-34) and for the initial
conditions and x,, the solution of Eq. (3-8) is
where
(3-11)
+
and are defined in Eq. (2-27) and =
Example 9
A component of a machine is represented sch'ematically in Fig. 3-8. Derive
its equation of motion.
Solution:
Assuming small oscillations and taking moments about 0, the equation of
motion is
=
= - -
which is of the form as Eq. (3-8).
....................... . . ...........
FIG. 3-8. One-degree-of-freedom system with damping.
78 Systems One of Freedom-Applications CHAP. 3
10. Logarithmic Decrement

A mass-spring system with viscous damping is shown in Fig. 2-7. The mass
m is displaced by an amount from its static position and then
released with zero initial velocity. Determine the of any two consecu-
tive amplitudes.
Solution:
From Eq. the maximum amplitude occurs when the product
and is a maximum. Rewriting the equation with as the
independent variable and equating = for maximum, we have
Hence the maximum amplitude occurs at
Let the motion be as illustrated in Fig. 3-9 and and

correspond to the maxima x, and x,. The last equation indicates that
+ = + $). Hence t,) = is a period and
= The consecutive amplitude ratio is
FIG. Free vibration with damping: initial conditions = and

= 0.
SEC. 3-3 Damped-Free Vibration
The natural logarithm of this ratio is called the logarithmic decrement 6,

that is, = Hence
(3-12)
or
for
The logarithmic decrement is a measure of the damping factor and it
gives a convenient method to measure the damping in a system.
It was pointed out in Chap. 2 that the rate of decay of vibrations is
a property of the system. Hence the logarithmic decrement must be
independent of initial conditions. Furthermore, any two points on the
curve in Fig. 3-9 one period apart serve to evaluate the logarithmic
decrement. The use of consecutive amplitudes, however, is a convenience.
Example
The following data are given for a system with viscous damping: mass
4 (9 lb,), spring constant k = 5 (28 and the amplitude
to 0.25 of the initial value after five consecutive cycles. Find the
damping coefficient the damper.
Solution:
The ratio of any consecutive amplitudes is
Hence
The damping factor and the damping coefficient c are

= 0.044
= = 12.5 s .
Following the method in the example above, the nunnber of cycles n
required to reduce amplitude by a factor of N is given by the
expression
Eq. is plotted in Fig. 3-10. Note that it takes

Systems with One Degree of Freedom-Applications CHAP. 3
FIG. 3-10. Number of cycles n to the amplitude by a factor of N

for small of damping.
less than four cycles to attenuate the amplitude by a factor of 10 when

= 0.1. This implies that it not take many cycles for the transient
motion to die out even for lightly damped systems.
3-4 UNDAMPED FORCED

VIBRATION-HARMONIC EXCITATION
The usual interest in the study of forced vibration with harmonic

excitation is the steady-state response of the system. As discussed in the
last section, the transient motion will soon die out, even for lightly
damped systems. Except near resonance, the steady-state response can be
approximated by that of an undamped system.
Neglecting the damping term in Eq. the equation of motion for a
system with harmonic excitation is
If the transient motion is assumed to have died out, the steady-state

response from Eq. (2-42) is
sin
1 - r2
where r = and
The equation above is plotted in Figs. 2-8 to 2-12. The corresponding
curves are when = Resonance occurs when the frequency ratio r = 1.
SEC. 3-4 Undamped Forced Vibration-Harmonic Excitation
The amplification factor R is infinite at resonance and the amplitude of

the also becomes infinite.
Alternatively, the behavior at resonance can be deduced from the
particular integral of Eq. (3-15). Dividing the equation by m and sub-
stituting for in the excitation term, we obtain
F
+ = - sin
m
(3-17)
Example 2, App. the particular integral is of the form

x= sin cos
where A, and A, are undetermined coefficients. this into Eq.
(3-17) and solving for the coefficients, we get
Thus, the amplitude increases proportionately with and would

theoretically become infinite.
Equation (3-18) indicates that it takes time for the amplitude to build
up at resonance. Hence, if the resonance is passed through rapidly, it is
possible to bring up the speed of a machine, such as a turbine, to beyond
resonance or its critical speed. At frequencies considerably above reso-
nance, Fig. 2-8 shows that the magnification factor is less than unity. It
may be advantageous to operate the machine in this speed range. It
should be cautioned that during the shut-down the machine would again
pass through the critical speed and excessive vibration might be encoun-
tered.
Example 12. Determination of Natural Frequency

The control tab of an airplane elevator is shown schematically in Fig. 3-11.
The mass moment of inertia of the control tab about the hinge point is
Elevator Control tab
e sin
F IG . 3-11. Determination of natural frequency.
Systems with One of Freedom-Applications CHAP.
known, but the torsional spring constant due to control linkage is difficult
to evaluate. T o determine the natural frequency experimentally, the
elevator mounted and the tab is excited as illustrated. The
excitation frequency is varied until resonance occurs. If the resonance
frequency is o , , find the natural frequency of the control tab.
Solution:
Taking moments about the hinge point 0, the equation of motion of the test
system is
=- - - - e sin
where (LO - e sin w t ) is the deformation of the spring Rearranging, this
equation becomes
+ k,)L 2 ]8 = sin o t
A t resonance,
k + f k 2 )L 2 = +
Hence
The next three examples illustrate an application of the simple pen-

dulum in a dynamic absorber for vibration control.
Example 13
The simple pendulum in Fig. 3-12 is hinged at the point The hinge point
is given a horizontal motion = e sin (a) the angular displace-
ment 8 of the pendulum for frequency ratios 1, and (b) the force
required to move the hinge point.
x e sin
equilibrium
(a) Vibratory system r
FIG. 3-12. Pendulum excited at support.

3-4 Forced Excitation
Solution:
(a) Assuming small oscillations, the horizontal acceleration of m in Fig.
is and the vertical acceleration is of second order.
Taking moments about 0 , the equation of motion is
+ =-
= - = sin (3-19)
+ = sin
where = is the amplitude of the equivalent torque. This
equation is of the same as Eq. From the
steady-state response is
.
=
1- r2
sin -
=
-
-
= and = Note that = when and =

when In other words, and are in phase with one
another when and 180" out of phase when r These phase
relations are illustrated in Figs. and (c).
( b ) For dynamic equilibrium of the pendulum, the horizontal force at the hinge
point is equal the horizontal component of the inertia force of the
pendulum, that is,
where is if the motion is to the of the static

equilibrium position. The equation shows that near resonance when
a large force could associate with a small amplitude e at the
hinge point
Example 14. Pendulum Dynamic Absorber

A simple pendulum in a centrifugal field can be used to nullify the torsional
disturbing moment on a rotating machine member, such as a crank shaft. A
rotating disk with a pendulum hinged at B is shown in 3-13. The disk
has an average speed and a superimposed oscillation sin where
n is the number of disturbing cycles per revolution of the: disk. Derive
the equation of motion of the system. ( b ) Find the amplitude ratio
Briefly discuss the application of the pendulum.
Solution:
The pendulum is under the influence of a centrifugal field when the system
is in rotation. Assume the gravitation is negligible compared with the
Systems sf
FIG. 3-13. Centrifugal pendulum.
centrifugal field at moderate and high speeds of rotation. As shown in the

figure, the pendulum bob m is subjected to a centrifugal force The
component of this force normal to L is sin From the triangle
we have
R 5
sin 8) sin sin a
(a) For small oscillations, the tangential acceleration of m is

(R Taking moments about B, the equation of motion of the
pendulum is
+ + =
Since =- sin not, the equation becomes

+ = m ( R+ sin
which is of the same form as Eq. (3-15).
(b) Comparing with Eq. we have 2

= mL , = mw 2 RL, and
= m ( R+ and is analogous to The steady-state
solution of the equation is
The amplitude ratio can be obtained by substituting the corre-

sponding eqivalent quantities into Eq. (3-16). Performing the substi-
tution and simplifying, we obtain
3-4 Forced 85
(r) The equation above shows that if n = the amplitude ratio =

Physically, this means that a finite value of is possible for an
arbitrarily small value of If the superposed oscillation = sin nwt is
due to a disturbing torque, the resultant oscillation of the machine
member can be very small for a finite of In other words. if the
centrifugal pendulum is tuned such that the disturbing torque
can be balanced by the inertia torque of the pendulum.
The number of disturbing cycles n per revolution in a rotary
as an internal combustion engine, is constant. When a centrifugal
pendulum is tuned for n = it is effective as a dynamic absorber
for all speeds of operation of the machine. This is not a friction type
damper, since it creates an equal and opposite torque to nullify the
disturbing torque.
From Fig' 3-13, the reacting moment on the shaft is due to the
tension + L ) of the pendulum. The moment arm R sin 0
Hence the magnitude of the reacting torque
= +
Example 15
A n eight cylinder, four-stroke cycle engine operates at 1,800 rpm. The
fluctuating torque is absorbed by the flywheel a dynamic absorber.
Assume the disturbing torque to be balanced' by the absorber is
500 N . m (4,425 The most convenient length for R as shown in Fig.
is 96 mm (3.8in.). If the maximum amplitude of oscillation of the
pendulum is find the length L and the mass m of the pendulum.
Solution:
T h e eight cylinder engine has four power strokes per revolution. Hence
= 4. The length of a properly tuned pendulum is
From the expression for in Example 14. the mass m of the
F I G . 3-14. Bifilar-type centrifugal-pendulum dynamic absorber.

86 One of 3
pendulum is
= kg (18.5 lb,)
The length of the pendulum is small; = 6 mm. This poses a design
in providing sufficient mass with a small L. problem is solved
by using a bifilar-type centrifugal pendulum as shown in Fig. 3-14. half
of the pendulum mass be mounted two loosely fitted pins on each side of
the crank. The diameter of the pins is and that of the holes through the
mass and the crank is Thus, each point on the mass moves in an arc of a
circle of radius - The Iength of the pendulum is L = -
3-5 DAMPED FORCED

VIBRATION-HARMONIC
Periodic excitation generally occurs in under steady-state

operation. Considering one of the harmonic components of the periodic
excitation, the equation of motion, becomes
+ + kx = sin o t
This is identical to Eq. (2-18) except for the substitution of for
The objective is to the sources of F and the manner it affects
"4
the system response In other words, the aim is in problem formula-
tion and in reducing the equation of motion to the form shown above.
The solution of the equation and the interpretation of its response
fully discussed in Chap. 2. Note that the phase angle of relative to
would be identical to the previous discussions as shown in Eq. (2-54).
Unless it is justifiable, the phase relation will not be discussed.
The applications are divided into six cases, which appear as distinct
types of problems. The equations of motion of Case 1 to 5 are reduced to
Eq. (3-22). Case 6 shows a generalization of vibration isolation. The
subject is further generalized for periodic excitations in the next section.
It is advantageous to use the impedance method in 2-6 to treat the
problem. Denoting sin by the vector = and the response
= - by the vector X = = Eq. (3-22) be-
comes
mx + cx =
Substituting = gives
-0 2m + = (3-23)
where = is the phasor of the excitation vector The mag-
nitude of is = and its phase angle is a, relative to a reference
3-5 Forced Vibration-Hamode Excitation
vector. If the excitation is the reference, then zero. Similarly,

is the phasor of the vector X. The magnitude of X is
and its phase angle is relative to the reference vector. It is understood
that the force and the response must be along the same axis in a physical
problem. it is unnecessary to denote the real or the imaginary
parts of the vectors all subsequent discussions.
the term in Eq. (3-23) and rearranging, we get
which is identical lo Eq. (2-52).
where = = and =
Case Rotating Reciprocating
A turbine, an electric or any device with a rotor as a working

part is a rotating machine. Unbalance exists the mass center of the rotor
does not coincide with the axis of rotation. The unbalance me is meas-
ured in terms of an equivalent m with an eccentricity e.
A rotating machine of total mass m, with an unbalance me is shown in
Fig. 3-15. The eccentric mass m rotates with the angular velocity o and
its vertical displacement is ( x sin ot). The machine is constrained to
move in the vertical and it has one degree of freedom. The
sin
(a) Vibratory system Equivalent
F I G . 3-15. Rotating unbalance.

88 Systems with One Degree of Freedom-Applications CHAP. 3
displacement of the mass - m) is Hence the equation of motion

of the system is
Rearranging the equation yields

+ + kx = mew 2
sin = sin cut
2
where = mew is the amplitude of the excitation force. Hence the
equivalent system is as shown in Fig. The steady-state solution is
given in Eqs. (3-25) and (3-26).
From Eq. the amplitude of the harmonic response is
This can be expressed in a nondimensional form. Multiplying and dividing

the equation by recalling = and simplifying, we
obtain
This is plotted in Fig. 3-16.

At low speeds, when the force mew2 is small and the amplitude
Frequency ratio r
FIG. 3-16. Harmonic response of systems with inertial excitation;
shown in Fig. 3-13.
Forced
F I G . 3-17. Reciprocating unbalance.

of vibration X is nearly zero. At resonance, when = 1 and the amplifica-
tion factor = 1/25, the: mass m) has an amplitude equal to
Hence the amplitude of vibration is limited1 only by the
presence of in the system. Furthermore, the mass - m) is 90"
out of phase with the unbalance mass m. For example, (m, - m) is
moving and passing its static equilibrium position, the mass m is
directly above its center rotation. At high speeds, when 1, the mass
m) has an amplitude In other words, the amplitude
remains independent of the frequency of excitation or the
damping in the system. The phase angle is 180"; that is, (m, - m) is
at its topmost position, m is directly below its center of rotation.
The discussion of rotating unbalance can be used to a recip-
rocating unbalance. A reciprocating engine is illustrated in 3-17. The
reciprocating mass m consists of the mass of the piston, the wrist pin, and
part of the connecting rod. The exciting force is equal to the: inertia force
of the reciprocating mass, which is approximately equal to +
where e is the crank radius and the length of the
connecting rod. If the ratio is small, the second-harmonic term,
can be neglected.? Thus, the problem reduces to that of the
rotating unbalance.
Case 2. Speed of
Shafts
A rotating shaft carrying an unbalance disk at its is shown in
Fig. Critical speed occurs when the speed of rotation of the shaft
*See, for example, R. T. Hinkle, Kinematics of Machines, 2d ed., Prentice-Hall, Inc.,
Englewood Cliffs, NJ, 1960, p. 107.
It will be shown in Case 3 for vibration isolation that if an isolator is adequate for the
fundamental frequency it would be adequate for the higher harmonics.
Systems One Degree of Freedom-Applications CHAP. 3
Disk
( a ) Vibratory system (b) General position of disk.
FIG. 3-18. Critical speed of rotating shaft.
is equal to the natural frequency of lateral (beam) vibration of the shaft.

Since the shaft has distributed mass and elasticity along its length, the
system has more than one degree of freedom. We assume the mass of the
shaft is negligible and its lateral stiffness is k.
The top view of a general position of the rotating disk of mass m is
shown in Fig. Let G be the mass center of the disk, P the
geometric center, and the center of rotation. Assume the damping
force, such as air friction opposing the shaft whirl, is proportional to the
linear speed of the geometric center P and that the flexibility of the
bearings is negligible compared with that of the shaft.
Resolving the forces in the and y directions gives
+ + kx = mew 2
cos o f cos wt
+ cy + ky = mew 2
sin = sin wt
Applying the impedance method illustrated in Eq. the equations
above become
The phase angle in the second equation above indicates that the
3-5 Damped
3-19. Phase relation of rotating shaft for r=
in the generic and y directions are at with one

another. It is evident that the amplitudes X and Y are equal.
Since the two harmonic motions and are
the same frequency, and at to each other, their sum is a circle.
Thus, the radius of the circle is equal X or Y. motion of the
geometric center of disk in Fig. 3-18 describes a circle of radius u
about the center of rotation From Eq. we get.
Substituting and = and simplifying we
This is identical to Eq. if m. This is true because the total

mass is also the eccentric mass for a rotating disk. Hence the response
curves in Fig. 3-16 for rotating unbalance also represents for
the whirling of rotating shafts.
The phase relation for various operating frequencies shown in Fig.
3-19. It is interesting to note that, when the frequency ratio r the
mass G tends to coincide with the center of rotation This can
be demonstrated readily. Assume that an unbalance rotor is rotating in a
balancing and that a piece of chalk is moved towards the rotor
until it barely touches. When the rotational speed is below critical, the
chalk mark is found on the side closer to the mass of the rotor.
When the speed is above critical, the-chalk mark is on the side away from
the mass center.
Elasticity Bearings and Supports
Rigid bearings were assumed in the above discussion of critical speed.
Figure 3-20 shows a pulley assembly, in which the brackets can
be more easily in the vertical direction than in the lateral
FIG. 3-20. Pulley assembly with flexible bearing supports.
direction. The effect of the elastic bearings is to render the system more
flexible and therefore lowering the critical speed. The critical speed can be
lowered by 25 percent in some installations. (a) Derive the equation of
motion of the system and briefly discuss the effect of the unequal elastic
supports on the system performance.
Solution:
A schematic representation of the system is shown in Fig. The
elasticity of the bearings and supports is represented by springs mounted in
rigid frames. The equivalent spring constants and are due to the
stiffness of the shaft, the bearings, and the supports in the and y
directions. A general position of the disk is shown in Fig. which
may be compared with Fig. P is the geometric center and G the
mass center of the disk. the center of rotation of the system correspond-
ing to the static equilibrium position of shaft.
(a) Schematic
.
(b) General position of rotating disk.
F I G . 3-21. System with elastic bearing supports.

3-5 Forced Vibration--Harmonic Exatation
FIG. 3-22. Rotation of disk about for various frequencies.
(a) For simplicity, the system is assumed undamped. The equations of

motion are
+ = cos
+ sin
Since the equations indicate that the system has two natural
frequencies and therefore two critical speeds. define =
2
= = = and = mew . Eq.
the amplitude ratios are
The two harmonic motions and are of the same and

at to each other. Since their amplitudes are unequal, the sum of
their motions is an ellipse. Thus, the geometric center moves in an
ellipse about as shown in Fig. 3-22. By neglecting the damping in the
system, the phase angle can be either zero for below the critical speed
or for above the critical speed. Since there are two natural
frequencies, we may consider the operations at speeds above and below
the critical.
When and both the disk and P rotate in the same
direction with the same speed as shown in Fig. The heavy side of
the disk and the position of the shaft key are marked for purpose of
identification. Assume When the disk and rotate
in opposite directions with the same speed as shown in Fig. When
is than both natural frequencies, again the disk and P rotate in
the same direction with the same speed as shown in Fig.
It is interesting to note that when the excitation is above or below the
critical speeds, there is no reversal in stresses in the shaft; is, while the
shaft is revolving, the compression side of the shaft in compression
of Freedom-Applications 3
and the tension side remains in tension. When the excitation is between the
critical speeds, the shaft undergoes two in stress per revolu-
tion.
of machines and field balancing' are further examples of

speed calculations. Since the subject is usually covered in the
dynamics of machines, it not be pursued here.
Machines often mounted on springs and dampers as in Fig.

3-23 to minimize the transmission of forces between the machine m and
its foundation.
We shall first consider the system in Fig. a
harmonic force is applied to and the deflection of the foundation is
negligible, the equation of motion is identical to force
transmitted to foundation is the sum of the spring force kx and the
damping force
Force transmitted =
If the excitation is harmonic, the magnitude and the phase angle of
excitation force and the other forces are as illustrated in Fig. 3-24.
The phase angle is generally is of secondary interest. Using
the force transmitted is
The ratio of the amplitude of the force transmitted and the am-
plitude of the driving force is called the transmissibility TR. From the
equation above, we have
Foundation Foundation (Base)

FIG. 3-23. Vibration isolation.
SEC. 3-5 Damped Vibration-Harmonic Excitation
F I G . 3-24. Relation of force transmitted and other vectors.
where = and = The equation is plotted in Fig. 3-25. Note

that all the curves in the figure cross at r = Hence transmitted
force is greater than the driving force below this frequency ratio and less
than force when the machine is operated above this frequency
ratio.
a speed machine, the amplitude of the exciting force is
constant. Hence the force transmitted is proportional to the value of the
Frequency ratio r
F I G . 3-25,. Transmissibility versus frequency ratio; in Fig.
96 Systems One of CHAP. 3
Frequency ratio
F I G . 3-26. Force ratio versus frequency ratio for excitation; sys-
tem shown in Fig. 3-15.
TR. It i s advantageous to operate a constant speed

machine at w
For a variable speed machine, the driving force due to an unbal-
ance me, is where is the operating frequency. Let us define a
constant force = Substituting = mew2 into Eq. dividing
both sides of the equation by and simplifying, we obtain
where is as defined in Eq. (3-32). Hence the magnitude of the force

transmitted can be high in spite of the low transmissibility. The equation
is plotted in Fig. 3-26.
The reduction of the force transmitted in buildings is of interest. For
example, the mechanical equipment of a tall office building is often
located on the roof directly above the penthouse or the boardroom of the
company.
The fractional reduction of the force transmitted is
Force reduction -
SEC. Damped Vibration-Harmonic
Static deflection
FIG. 3-27. Percentage reduction in force transmitted to foundation in
isolation, 0.
where and are the amplitudes of the excitation and the transmitted
force, respectively. It is observed in Fig. 3-25 that low natural frequency
and low damping are desirable for vibration isolation. Assume and
1 in Eq. (3-32). Thus, TR = l/(r 2 - and the force reduction becomes
Force reduction =-
r2 -
Since r2 and the static deflection of a spring =
the equation above reduces to
-
Force reduction =
-
The equation is in Fig. 3-27
Example 17
A n air compressor of 450 kg mass (992 lb,) operates at a constant speed of
1,750 rpm. The rotating parts are well balanced. The reciprocating parts are
of 10 kg (22 lb,). The crank radius is (4 in.). the damper for the
mounting introduces a damping factor 0.15, (a) specify the springs for
the mounting such that only 2 0 percent of the unbalance force is transmit-
ted to the foundation, and (b) determine the amplitude of the transmitted
force.
98 Systems with Degree of Freedom-Applications CHAP. 3
Solution:
= = 183.3
Since TR = 0.20 = +
(b) Amplitude of force transmitted 0.20

2
= 0.20meo = 6.72 (1,510
Case 4. System Attached to

Moving Support
When an excitation motion is applied to the support or the base of the
system instead of applying to the mass, both the absolute motion of the
mass and the relative motion between the mass and the support are of
interest. W e shall consider the absolute motion of the mass in this case.
Let the base of the system in Fig. be given a harmonic
displacement The corresponding displacement of m is The
spring force is and the damping force is Applying
Newton's second law to the mass yields
which can be rearranged to
Applying the impedance method gives x,=X l e l "'. The quantity

(kx, + = (k + = can be considered as an equivalent
force as shown in Eq. (3-22). Hence the equation above reduces to
where
and klm, = and =

Since Eq. (3-37) for is identical to Eq. (3-32) for both
can be called the transmissibility equation, although the former denotes
the transmission of motion from the base to the mass and the latter the
SEC. 3-5 Damped Forced Vibration-Harmonic Excitation
transmission of force from the mass to its foundation. Hence Fig. 2-25 is
a plot of both equations.
Example 18. Mounting of Instruments
An instrument of mass m is mounted on a vibrating table as shown in Fig.
Find (a) the maximum acceleration of the instrument, and the
maximum force transmitted to the instrument. Assume the motion of
the table is harmonic at the frequency w.
Solution:
(a) The equation of mass m is as shown in Eq. (3-35). The maximum
acceleration of m is Applying (3-37) yields
where r = and Alternatively, we can compare the accel-

eration of m with that of the table, that is,
Thus, the characteristics of the acceleration ratio is the same as the

displacement ratio and plotted in Fig. 3-25.
(b) is transmitted to m through the spring and the damper. From Eq.
2
the sum of these forces is = mw X 2 . Applying Eq.
the maximum force transmitted is
Comparing with the maximum acceleration of the support, we

have and
+
Hence, with the exception of the constant m , this equation is rep-
resented in Fig. 3-25. Comparing with the maximum displacement
of the support, we have
where m w Z = = Hence, with the exception of the constant

k, this equation is represented in Fig. 3-26.
Example Vehicle Suspension

A vehicle is a complex system with many degrees of freedom. As a first
Fig. 3-28 may be considered as a vehicle driven on a rough
FIG. 3-28. Schematic sketch of vehicle moving over rough road.

road. It is assumed that (1) the vehicle is constrained to one degree of
freedom in the vertical direction, (2) the spring constant of the tires is
infinite, that is, the road roughness is transmitted directly to the suspension
system of the vehicle, and (3) the tires do not leave the road surface.
Assume a trailer has 1,000 kg mass (2,200 lb,) fully loaded and 250 kg
empty. The spring of the suspension is of 350 (2,000 The
damping factor 0.50 when the trailer is fully loaded. The speed is
100 (62 mph). The road varies sinusoidally with 5.0 mlcycle
(16.4 Determine the amplitude ratio of the trailer when fully
loaded and
Solution:
The excitation frequency is
From Eq. the damping coefficient is Since c and are

constant, varies inversely with the square root of m. Thus, 1.0 when
the trailer is empty. Applying Eq. the calculations are tabulated as
follows:
ITEM TRAILER, FULLY LOADED TRAILER,
Natural frequency
= = =
= 18.7 = 37.4
= 2.98 Hz = 5.96 Hz
=
Ratio of
The amplitude ratio when fully loaded and empty is = 111.67.

SEC. 3-5 Damped Forced Vibration-Harmonic Excitation
Case 5. Seismic Instruments

A schematic sketch of a seismic instrument is shown in Fig. 3-29. It
consists of a mass m attached to the base by means of springs and
dampers, that is, the base of the instrument is securely attached to a
vibrating body, the motion of which is be measured. Let be the
motion of the base and the motion of The relative motion
is used to indicate As illustrated, is recorded by
means of a pen and a rotating drum.
Since this system is essentially the same as that shown in Fig.
its equation of motion is identical to Eq. Let x, = X , sin
Defining = - and substituting = into Eq.
(3-35) yields
mf + + kx = - = sin ot (3-39)
which is of the same form as Eq. (3-22) with
Using the impedance method 2-6, Eq. (3-39) gives
0 2m -
1 -
k - 0 2m +
where = tan-' r 2 ), = and = The: amplitude ratio
the equation above is
The right side of this equation is identical to Eqs. (3-28) and

Hence the characteristic of the equation is also portrayed in Fig. 3-16.
The relative motion between the mass m and its in a seismic
instrument, or vibration pickup, can be measured mechanically as illus-
trated in Fig. 3-29. For high speed operations and convenience, this
motion is often converted into an electrical signal. The schemes illustrated
in Figs. and ( b ) for this conversion are self-evident. A typical
piezoelectric accelerometer is illustrated in Fig. The piezoelectric
Rotating
Record
I
Base
F IG. Schematic sketch of seismic instrument.
Piezoelectric discs
Frame
Mounting stud
[a) Pickup with strain ( b ) Variable-reluctance (c) Piezoelectric accelerometer

sensitive element pickup
FIG. 3-30. Vibration pickups with electrical output.
elements are sandwiched between the mass and the frame. The voltage
output of the device is due to the cyclic deformation of the piezoelectric
crystals. The effective spring, damping, and mass in this accelerometer,
however, are not self-evident. Since vibration measurement is a separate
study, we shall not pursue the subject further.
Vibrometer: Referring to Fig. 3-16, if or r

the ratio in
Eq.(3-40) approaches unity regardless of the value of In other words,
the relative displacement is equal to the displacement which is
the motion to be measured. The phase angle is approximately 180". An
instrument for displacement measurement is called a vibrometer. Since
there is no advantage in introducing damping in the system, a vibrometer
is designed with damping only to minimize the transient vibration.
The performance characteristics of a typical inductive type velocity
pickup is illustrated in Fig. 3-31. The useful range is marked off in the figure
with bold lines. Displacement and acceleration can be obtained from
the electrical output of the velocity pickup by integration and differen-
tiation. The natural frequency of this pickup is 8 Hz. The displacement
amplitude ranges from 0.0025 to 1 0 mm.
A seismic instrument to measure acceleration is

called an accelerometer. Due to their small size and high sensitivity, most
vibration measurements today are made with accelerometers. The veloc-
ity and displacement can be obtained from the electrical output of the
accelerometer by integration. For example, the motion of the piston of an
internal combustion engine can be indicated in this manner.
If the motion to be measured is sin the amplitude of the
acceleration is 02X,. From Eq. we obtain
SEC. Damped Forced Vibration-Harmonic 103
The quantity is a constant, since is a property of the system. The

relative motion is proportional to the acceleration if the
magnification factor is constant for all ranges sf operation.
A periodic vibration generally has a number of harmonic components,
each of which gives a corresponding value of r ( = in Eq. (3-41).
Amplitude distortion occurs if the magnification factor - +
changes with the harmonic components. In other words, the magnification
of each harmonic component must be identical in order to reproduce
the input waveform. Since R 1 when r approaches zero, an accelerome-
ter is constructed such that or The percent amplitude
distortion is defined as
Amplitude distortion = ( R - 1 ) 100% (3-42)
FIG. 3-31. characteristics of a typical inductive velocity

pickup.
Frequency ratio
FIG. 3-32. Amplitude distortion in accelerometer.
This is plotted in Fig. 3-32. Note that (1)an accelerometer should be built
with in order to minimize the amplitude distortion and (2)
the usable range is 0.6.
Phase distortion occurs if there is a shift in the relative phase between
the harmonic components in a periodic signal. Assume a periodic signal
in Fig. has two harmonic components. The recorded signal in
Fig. consists of the same components without amplitude distor-
tion. The relative phase between the components, however, has changed.
Evidently the distortion in the wave-form of the recorded signal is due to
the phase distortion. Phase distortion is secondary for applications.
It is important for the applications in which the wave-form must be
preserved.
For zero phase distortion, the phase shift of each of the harmonic
components in the signal must increase linearly with frequency. Consider
the equations
= +
- 4,) + -
= sin sin -
+ sin -
Harmonic components
(a) Original signal Recorded signal
F I G . 3-33. Phase distortion.

SEC. 3-5 Forced 105
Frequency ratio
FIG. 3-34. Phase distortion in accelerometer.
The quantities and denote the delay or the shift of the signal
along the positive time axis. If each harmonic component of a signal
is shifted by the same amount of time, the wave-form is preserved. This
requires that = = constant, that is, = = constant. In
other words, the phase angle varies linearly with the frequency
It is observed in Fig. 2-9 that, for the phase angle vary linearly
with frequency over an acceptable range the phase shift is
r). Hence the phase distortion of an accelerometer is defined as
Phase distortion = - 90r) deg (3-43)
This is plotted in Fig. 3-34. Again, it is seen that an appropriate damping
in an accelerometer is necessary in order to minimize the phase distor-
tion.
Example 20
A machine component is vibrating with the motion
Y = +
= sin + sin
= 0.10 sin + 0.05 sin
Determine the vibration record that would be obtained with an accelerome-
ter. Assume = 0.65 and = = 1,500 H z .
Solution:
From the equation of motion is
Systems with One of Freedom-Applications CHAP. 3
The harmonic response due to of the components in the input can be

obtained from Eq. (3-40). By superposition, we have
where
= tan-' - and = tan-'
1-
The frequency ratios are = = = 1/50 and =
= 2/50. The values of the magnification factors and are
almost unity. Thus,
= tan-'
1-
Hence the acceleration record is
The output of an accelerometer is usually converted to an electrical signal

and amplified. Hence the value of the recorded acceleration depends on the
amplification used in the data processing. If the equation above is integrated
twice to give a displacement, the measured value of the given motion is
y = -
(measured) + 0.05
Note that the measured value of has practically no amplitude distortion
and only a slight shift in the phase angles. There is no phase distortion,
however, because the phase shift is linear with the frequency of the
harmonic components. There is a slight time delay between the input
and its measured value. The time delay is For the given values, we
have time delay = 1.49' 0.14 ms. The results above are due
to the high natural frequency of the accelerometer.
Case 6. Elastically Supported Damped Systems
T h e damping of a real system can be considerably m o r e complex than a

simple damper shown in Fig. 3-23. Equivalent viscous damping will b e
discussed in 3-8. W e shall consider the elastically supported damper
as illustrated in Fig. 3-35.
From the free body sketch, the equation of motion for the mass m is
SEC. 3-5 Damped Excitation
F sin wt
F sin wt
Free-body. sketches
F I G . 3-35. Elastically supported damper.

Since the damper c and the spring are in series, the damping force is
equal to the spring force.
-
Since the motions and are harmonic, the equations above can
be solved by the mechanical impedance method cliscussed in
2-6. Substituting for Fsin for the time derivatives, and using
phasor notation, the equations become
The phasors and can be solved for by
Defining the stiffness ratio of the springs as =

r= and
we obtain
-
= tan-' 2 tan-' -
1- r
(3-48)
= tan-'
1- r2 2
Example 21. Vibration Isolation*

A machine of mass m is mounted on shown in
Fig. 3-35. Define = = r= and N = (a) De-
rive the equation for the transmissibility T R of the system. (b) If r 0.6,
= 2 , and 0.4, find the transmissibility TR. (c) Repeat part b if r 10
and the other parameters remained unchanged. (d) Compare the values of
TR from part b and part with that expressed in Eq. (3-32).
Solution:
(a) The force transmitted to the foundation is the sum of the forces
transmitted through the springs k and
The phasors and are given in Eq. Using vectorial addition,

the phasor of the transmitted force is
kZ+
+ jwc)+
- + -
where is the phase angle of the transmitted force relative to the

excitation and is defined in Eq. (3-46). The transmissibility T R is the
ratio of the magnitude of the transmitted force relative to that of the
excitation; that is,
(b) For = 0.6, N = 2 , and = 0.4, we obtain
For r = 10, N = 2, and 0.4, we have
(d) Substituting the corresponding values in Eq. the values of T R

for the system in Fig. are
for
for r- 10
* A detailed analysis is shown in J. C . Vibration and Shock in Damped Mechani-
cal Systems, John Sons, Inc., New York, 1968, pp. 33-38.
SEC. Forced Excitation 109
Comparing parts (b), (c), and (d), the transmissibility for the two
isolation systems are approximately the same for r = 0.6. When operat-
ing at r 10, the system in Fig. 3-35 seems to be superior to that in Fig.
3-6 DAMPED FORCED

VIBRATION-PERIODIC
Harmonic response of systems was presented in the last section. Forces

arising from machinery are commonly periodic but seldom harmonic. In
considering periodic excitations, we are in effect generalizing the previous
applications. We shall briefly review the Fourier series and then illustrate
the applications.
As illustrated 1-2, a function is periodic if
F(t T) (3-50)
where is the period, or the minimum time required for to repeat
itself. The Fourier series* expansion of is
-+ (a, cos not + sin not)
where n is a positive integer, a, and are the coefficients of the infinite

series. Note that gives the average value of The fundamental
frequency of the periodic function is o = that is, when n = 1. The
frequency of the nth harmonic is n o = that is, when n 1.
The following relations are used to evaluate a, and
if
cos nor dt =
if m=n
[sin mot sin not dt = if (3-52)

if m=n
cos mot sin not dt = whatever m and n
where m and n are integers and = the period of Rewriting

the series in an expanded form, we get
+ (b, sin + sin + .

A particular coefficient a, can be obtained by multiplying both sides of
*See, for example, I. S. Sokolnikoff and R. M. Redheffer, Mathematics of Physics and
Modern Engineering, McGraw-Hill Book Co., New York, 1958, p. 175.
Systems with One of Freedom-Applications 3
this equation by (cos and integrating each term using the relations in
Eq. (3-52). Except for the term containing a,, all the integrals on the
right side are identically zero. Thus,
pot = + ..
a
2
or
Similarly, a particular coefficient can be obtained by multiplying the

series by pot) and applying the relations in Eq. (3-52). Thus, the
coefficients of the Fourier series in Eq. (3-51) are
= nor dt
Let a periodic force be applied to a one-degree-of-freedom

system. may represent the equivalent force in any of the five cases
enumerated in the previous section. Expanding in a Fourier series
and applying Eq. the equation of motion becomes
+ + kx = cos not sin not) (3-54)
The steady-state response due to each of the components of the excita-

tion can be calculated. By superposition, the steady-state response of the
system is
a, - - (3-55)
2k +
where
= z and
1- n r 2
Example 22
Find the Fourier series of the square wave in Fig.
Solution:
For any one cycle, the given periodic function is
1 for
for
SEC. 3-6 Damped Forced Vibration-Periodic Excitation
(a) A periodic square wave
Harmonic components of a square wave
FIG. 3-36. Fourier series analysis of a square wave.
Applying Eq. (3-53) and for = the coefficients of the Fourier series
of are
dt- =
a. = not dt
4
for n odd
for n even
the Fourier series expansion of the square wave is

4
for
The first four harmonics of and their sum are plotted in Fig.
The impedance method in 2-6 can be applied readily to this type

of problem. We shall consider (1) the Fourier spectrum of the periodic
excitation (2) the transfer function of the system, and then (3) the
technique to combine the two spectra to obtain the spectrum of the
response. This general technique is applicable to any linear system.
Consider the two terms in Eq. (3-51) of the same frequency no. Their
sum can be expressed as
cos nwt + b, sin not = - a,) (3-56)
where
= and = (3-57)
Note that (1) is the amplitude and a, the phase angle of the excitation
at the frequency and (2) when n = we have = and = 0.
Thus, using the vectorial notation, a periodic excitation can be
expressed as
where = is the phasor of the harmonic component at the

frequency no. The plot of c, versus frequency is called the frequency
spectrum and versus frequency the phase spectrum of The two
plots as shown in Fig. are known as the Fourier spectrum.
From Eq. the sinusoidal transfer function of a
freedom system is
The plot of versus frequency is analogous to the frequency spectrum

and versus frequency the phase spectrum. In other words, the transfer
function plots in Fig. are the continuous plots of the Fourier
spectrum of the system for excitations of unit magnitude and zero phase
angle for all frequencies.
The equation of motion of the system is obtained by substituting Eq.
(3-58) into (3-22). Thus,
Using the impedance method and Eqs. (3-24) to the response due
3-6 Damped Forced Vibration-Periodic Excitation
w Frequency
(a) Fourier spectrum of periodic input
Frequency
transfer function
Frequency
Fourier spectrum of system response
F IG . 3-37. Construction of response spectrum input spectrum and

transfer function.
to a typical component, c, = = , of the excitation at

the frequency nw is
where
=
k - n 2w 2m + jnwc
and
jnwc
= tan-'
k-
Note that the phasor of a harmonic response in Eq. (3-62) is the
product of and - n 2w 2m both of which are complex
numbers at the given frequency In other words, the Fourier spectrum
of the system response is the product of the Fourier spectrum of the
excitation and the system transfer function. The rules for the product
of complex numbers are given in Eqs. (1-14) and that is, (1) the
magnitude of the product is the product of the magnitudes and (2) the
phase angle of the product is the algebraic sum of the individual phase
angles. At the frequency nw, the magnitude of the response is -
n 2w 2m + which is the product of the frequency spectrum of
and the magnitude of the system transfer function; the phase angle
+ of the response is the algebraic sum of the phase spectrum of
and the phase angle of the system transfer function. Thus, the
Fourier spectrum of the system response can be constructed as shown in
Fig. considering all the harmonic components of'
By the superposition of the individual responses in Eq. the
system response is
which is Eq. in vectorial form. The waveform of the time response

can be constructed from the Fourier spectrum of the response by super-
position.
Example 23
A cam actuating a spring-mass system is shown in Fig. 3-38. The total cam
lift of the sawtooth is 25 mm in.). The cam speed is 60 rpm. Assume
SEC. 3-6 ,Damped Forced Vibration-Periodic Excitation
FIG. 3-38. Periodic excitation.
m = 20 kg (44 and = k = 3.5 (20 The damping coeffi-

cient is c = 0.2 . (1.14 Find the response
Solution:
A cycle of the sawtooth motion can be expressed as
1
for
Since the fundamental frequency is 1 Hz or w = the period is

Applying Eq. it can be verified readily that
Hence the Fourier series expansion of is
The equation of motion of the system
We define = (k + = and = response due to

the constant term is
The response due to a typical harmonic excitation term at the frequency

no = is as shown in Eq. (3-62).
=
+
where
tan-' z
- n 2r
By superposition, the total response due to the excitation is
For the given data, we have
Thus, the response of the system is
where
3-7 TRANSIENT VIBRATION-SHOCK SPECTRUM
The design of equipment to withstand shock is of concern to the

engineer. Vibrations induced by the steady-state operation of a machine
are generally periodic. This was discussed in the last two sections.
Vibrations due to shock and transients usually originate from sources
outside the machine or from a sudden change in the machine operation.
The transient will die out, but the machine may be damaged or may
malfunction momentarily, both of which should be well considered.
A shock is a transient excitation, the duration of which is short
compared with the natural period (reciprocal of natural frequency) of
oscillation of the system. The transient response due to a transient
excitation was discussed in 2-7. The recording from a vibration test,
in the form of a "wavering line" versus time, cannot be used directly by
the designer. The shock spectrum is a common method to reduce the test
data to a more usable form.
SEC. 3-7 Transient Vibration-Shock Spectrum 117
A shock spectrum (response spectrum) is a plot of the peak response

versus frequency due to the applied shock. The peak response is that of a
number of one-degree-of-freedom systems, each tuned to a different
natural frequency. The frequency is that of natural frequency of the
individual systems. The response may be expressed in units of accelera-
tion, velocity, or displacement.* For example, a vibrating reed shown in
Fig. 3-3 is a simple mass-spring system. A reed gage consists of a number
of reeds of different natural frequencies. Using a reed gage in a shock
test, the maximum displacements of the tips of the reeds give the
maximum response for the various natural frequencies. The reed gage can
be replaced by a single accelerometer and computers employed to simu-
late the reeds.? The one-degree-of-freedom system is variously called a
an oscillator, or a simple structure.
The objective of shock spectrum is to describe the effect of shock
rather than the shock itself. Shocks are difficult to characterize and a
specific pulse shape is difficult to obtain in a test machine. It is necessary
to correlate test data from different laboratories. The shock spectrum a
"common denominator" on the assumption that shocks having the same
spectrum would produce similar effect. The spectrum may be regarded as
indicative of the potential for damage due to the shock. For example, the
peak relative displacement between the mass m and its base for the
system in Fig. is related to the stress in the spring. In other words,
the envelope of the spectrum establishes an upper bound of the stress
induced, or the damage potential, by a specific shock on the equipment
under test.
Types of are usually categorized using the undamped
resonator as the standard system. Types of shocks and methods of data
reduction can be found in the It is paradoxical that a complex
study like shock is treated in a seemingly simple manner. The reason is
that the time and expense for a detail study must be justified by the past
experience of the engineer. An undamped resonator is used, since the
largest excursion occurs within the first cycle of the transient and the
error introduced by neglecting damping tends provide a
safety. Note that a system under test commonly has more than one degree
of freedom. It will be shown in later chapters that a complex system has
discrete modes of vibration and a natural frequency is associated with
* C. E. Crede, Shock and Vibration Concepts in Engineering Design, Book

Co., New York, 1965, p. 138.
C. T. Morrow, Shock and Vibration Engineering, John Wiley and Sons, Inc., New York,
1963, p. 111.
S . Ayre, "Transient Response to Step and Pulse Functions," Chap. 8 of Shock and
Vibration Handbook, vol. 1 , C. M. Harris and C. E. Crede (eds.) McGraw-Hill Book, Co.,
New York, 1961.
S. "Concepts in Shock Data Analysis," Chap. 23 of Shock and Vibration
Handbook, vol. 2, C. M. Harris and C. E. Crede (eds.), McGraw-Hill Book Co., New
1961.
each of the modes. Hence a complex system can be described in terms of

equivalent one-degree-of-freedom systems. Thus, a shock will excite all
the modes of a system.
If a shock is due to a sudden change in the machine m in Fig.
the equation of motion of m is identical to Eq. (3-1). Assuming
zero initial conditions, the response from Eq. (2-71) is
where
as defined in Eq. (2-69). On the other hand, if the excitation is applied to

the base of the machine as shown in Fig. the equation for the
relative displacement between m and its base is identical to Eq.
(3-39).
mi! + + -
where and and are the absolute motions

indicated in the figure. Applying Eq. (2-71) yields
where is as defined above. The maximum response and the corres-

ponding shock spectrum can be obtained from Eq. (3-66) or (3-69)
depending on the application. Computers can be used for the calcula-
tions.*
To illustrate a shock spectrum by hand calculation, let a one-half sine
pulse shown in 3-39 be applied to the mass m of the system in
FIG. 3-39. A half-sine pulse.
* See, for example, J. B. Vernon: Linear Vibration and Control System Theory, John Wiley
Sons, Inc., New York, 1967, pp.
SEC. 3-7 Transient Vibration--Shock
Fig. 3-23. Assume the system is undamped. is described by the
sin for t
=
for
For t the system response from Eq. (3-66) is
which can be integrated to yield
= (sin
Equating = to find we get

COS - COS =
where is the time when is a maximum or minimum. The roots of

this equation is deduced using the identity
cos - cos = - 2 sin + sin -
Thus,
2
=-- for n = integer
Consider = + Defining = the terms in Eq. (3-70)

can be expressed as
2
sin sin = sin
sin
Hence sin = -sin Recalling klm, from Eq. we get
2
=
(1+-
):
-=-
1
sin
Systems with Degree of Freedom-Applications CHAP. 3
Similarly, if = - we obtain
Comparing the last two equations, it is evident a value of n can be

selected to have occur at = + Equation (3-71) is
plotted in Fig. 3-40. The initial shock spectrum, defined by Eq.
gives the response within the duration of the shock pulse for
Note that there is no solution for 1, since the maximum response
does not occur during the pulse if the natural frequency is smaller than
the pulse frequency.
For t the system response from Eq. (3-66) is
The upper limit of the integration is because = for t

Performing the integration, substituting = and simplifying, we get
convenient to define t- a new origin for the time axis.

Recalling = defining r and = the equation can be
Undamped natural frequency

Pulse frequency
F I G . 3-40. Shock spectra for half-sine pulse applied to m in Fig.

Eq. (3-71) shows initial spectrum; Eq. (3-72) residual
spectrum (Crede).
SEC. 3-7 Transient Spectrum
simplified to
The maximum value of can be expressed as
-
r 2- 1
This gives the residual ,shock spectrum, as it occurs after the shock has
terminated. The equation is plotted in Fig. 3-40 and shown as a dash
line.*
Now consider the shock spectrum due to a one-half sine pulse
applied to the base of the system in Fig. Assume the system is
undamped. The equation of is
for
for
For t the equation for the absolute motion is

+ = (t)
Since the excitation is sin wt, the response can be obtained from
Eq. (3-70) by substituting for that is,
Hence the initial shock spectrum is deduced directly from Eq. (3-71):
-- .
sin
l- r l+r
The relative motion between m and its base is given by the relation
where is as defined in Eq. (3-73). The expression for can be

maximized to give the corresponding spectrum.
The system is unforced for The residual shock spectrum can be
calculated as before. shock spectra for the system shown in Fig.
excited by a half sine pulse is illustrated in Fig.
* C. E. Crede, op. p. 85.

L. S. Jacobsen and R. S. Ayre, Engineering Vibrations, Book New
York, 1958, p. 163.
Systems with One of Freedom-Applications CHAP.
Pulse duration
Natural period of system T
F I G . 3-41 Shock spectra for half-sine pulse applied to the base in Fig.
X is relative displacement; the initial spectrum; the
residual spectrum (Jacobsen Ayre).
3-8 EQUIVALENT VISCOUS DAMPING
Damping is a complex phenomenon.* It exists whenever there is energy

dissipation. Viscous damping was assumed in the previous sections. This
occurs only when the velocity between lubricated surfaces is sufficiently low
to ensure laminar flow condition. More than one type of damping may
exist in a problem. Since damping is nonlinear, the superposi-
tion of different types of damping in a calculation does not always give
reliable results. Furthermore, damping may be dependent on the operat-
ing conditions and the past history of the damping mechanism or even the
shape of the damper, such as in a visco-elastic material. In order to have a
simple mathematical model, we shall examine the viscous equivalent of
different types of damping.
Most mechanical systems are inherently lightly damped. The effect of
damping may be insignificant for some problems, which may be treated as
undamped except near resonance. Damping must be considered, how-
ever, in order to control (1) the near resonance conditions of a dynamic
system, and (2) the performance of a machine, such as an accelerometer
or the riding quality of an automobile. The control can be achieved by
using (1) an energy transfer mechanism, such as a dynamic absorber
*The interest and knowledge in damping has increased exponentially in recent years.
Between 1945 and 1965, two thousand papers were published in this area.
B. Lazan, Damping of Materials and Members in Pergamon Press
Ltd., 4 5 Square, London 1968, p. 36.
SEC. 3-8 Equivalent Damping
Excitation
citation force
force
Foundation
Transmitted
force
(a) Lumped-parameter ( b ) Isolation to Isolation to

minimize minimize
F I G . 3-42. Model of a one-degree-of-freedom system.
discussed in Examples 14 and 15, or (2) an energy dissipating mechanism,

which is the present topic of discussion.
Consider the one-degree-of-freedom system in Fig: Damping
occurs in the damping mechanism DM, which can be viscous or other-
wise. The elements are shown separate in order to form a model for the
study. In reality, may not be separable, such as in an isolator of
visco-elastic material. An isolator is shown in Figs. and (c) to
minimize the transmission of forces between the madhine m and its
foundation.*
Let us find the equivalent damping from energy considerations. Under
cyclic strain, the energy dissipation in a damper is measured by the area
of the hysteresis loop shown in Fig. 3-43. Without energy dissipation, the
cyclic stress strain curve is a line with zero enclosed area, that is, the loop
degenerates into a single valued curve. From Fig. if
x and is sinusoidal, the equation of motion is
where is the damping force. The energy dissipation per cycle AE is
where the period in time per cycle.
*See, for example, J. E. Ruzicka and T. F. Derkley, Influence of Damping in Vibration

Isolation, SVM-7, The Shock and Vibration Information Center, U.S. Department of
Defense, 1971.
Force
FIG. 3-43. Hysteresis loop of a damping mechanism.
If the damping is viscous, the displacement and the damping force

are
= -
(3-77)
where c is the viscous damping coefficient. Combining Eqs. (3-75) to
(3-77) gives
(a) Viscous damping (b) Velocity squared damping
(c) Coulomb damping (d) Hysteretic damping
Harmonic displacement across

damping mechanism DM Damping force
FIG. 3-44. Wave form of damping force and relative displacement of

damping mechanism.
SEC. 3-8 Viscous Damping 125
Combining Eqs. and (3-77) yields
ocx
Hence the hysteresis loop for viscous damping is an ellipse having the
major and minor of and respectively.
If the damping is nonviscous, the equivalent viscous damping
is obtained from Eq. (3-78).
AE (3-80)
Note that the criteria for equivalence are (1)equal energy dissipation per
cycle of vibration, and (2) the same harmonic relative displacement. The
assumption of harmonic motion is reasonable only for small nonviscous
damping. The wave forms of damping forces for harmonic displacements
across typical dampers are illustrated in Fig. 3-44.
Example 24. Coulomb Damping

A mass-spring system with Coulomb (dry friction) damping is shown in Fig.
3-45. Assume (1) the-frictional force is proportional to the normal
force and (2) the initial conditions are = and = 0. Find the
motion and the change in amplitude per
FIG. 3-45. Coulomb damping.
Solution:
Let = where is a frictional coefficient. If the motion
is from left to right the frictional force is and vice versa. From
Eq. (3-74) the equation of motion is
where = accounts for the sign change. The corresponding solu-

tion is
cos + sin -
where and A, and are constants. Substituting the initial

conditions, we have
x - cos -
Systems with One Degree of Freedom-Applications CHAP.
For the first half cycle = displacement is
Hence the decrease in amplitude is per half cycle. The amplitude

change per cycle for free vibration with Coulomb friction is
Example 25. Equivalent Damping Coefficient

If a force F sin is applied to the mass m in Fig. 3-45, find the equivalent
viscous-damping coefficient and the magnification factor R.
Solution:
Since the damping force = is constant and the total displacement
per cycle is the energy dissipation AE per cycle is
AE
Comparing this with Eq. we have
From Eq. the amplitude X of the steady-state response is

F
(3-81)
-
This is an implicit equation, since is a function of and X. Substituting

the expression for and simplifying, the magnification factor is
where r = Since X is real, the equation is valid only if

Note that the amplitude at resonance is always theoretically infinite. T h e
resonance amplitude can also be viewed from energy considerations. The
energy input per is and the dissipation is If
the excess energy is used to build up the amplitude of
oscillation.
Example 26. Quadratic Damping

Quadratic or velocity squared damping is encountered in the turbulent flow
of a fluid. Determine the equivalent viscous-damping coefficient and the
amplitude of the steady-state response. Assume x = wt.
* Let the force = F sin and the harmonic response be x X - 4). resonance
= The energy input per cycle is
where = period. This is integrated to give

SEC. 3-8 Viscous 127
Solution:
Assume = where c, is constant and is the relative velocity
across the damper. Since damping always opposes the motion, the equation
of motion from Eq. (3-74) is
+ c 2x 2 + k x = F sin
where = accounts the sign change.
The energy dissipation per cycle from Eq. (3-75) is
Substituting the in Eq. (3-80) gives

8
=
Again, is not a constant as assumed for viscous damping.

The amplitude X for the steady-state response is obtained by substituting
into Eq. Simplifying the resultant equation yields
2 nkm
Example 27. Velocity-nth Power

the for the velocity-nth power or exponential damping.
Solution:
The energy dissipation AE per cycle is
Substituting the AE into Eq. we
where
Example 28. Hysteretic Damping

Assume the damping force in a one-degree-of-freedom system is in phase
with the relative velocity but is proportional to the relative
across the damper. Find the harmonic response of this system and its
equivalent viscous-damping coefficient.
Solution:
Writing the equation of motion in the vectorial form as shown in 3-5,
we have
+ jhx + kx =
mx + k ( l + =
where (jhx) is the damping force, h a constant, = is called the loss

and the complex Solving Eq. (3-82) by the
impedance method, the amplitude X of the harmonic response is
where = Comparing Eqs. (3-81) and the equivalent

damping coefficient is
(3-84)
Solid damping, hysteretic damping, and structural damping are terms

commonly used to describe the internal damping of material. It is
assumed that the energy dissipation per cycle is independent of frequency
and is proportional to the square of the strain amplitude. Substituting Eq.
(3-84) into we obtain AE = which confirms the assumption.
The nature of structural damping is rather complex. For mild steel, the
energy dissipation is proportional to For other cases, the value of
the amplitude exponent may range from 2 to 3. The damping of a
material may decrease slightly with increasing frequency instead of being
constant. In contrast, the common viscous-damping theory assumes that
the loss coefficient increases linearly with frequency.
Owing to their high damping characteristics, visco-elastic materials
have gained importance in vibration control in recent years. The physical
properties of such materials are more complex than those of metals. The
properties, are influenced by the operating conditions, the past history,
and the geometry of the damping mechanism. The variation of properties
temperature and frequency of a typical visco-elastic material is
shown in Fig. The complex modulus is defined by the
relation
(Stress) =
where is the loss coefficient. Note that high damping can be achieved in
the transitional region. Furthermore, if the spring constant of a damping
mechanism increases with preloading, it is feasible to tune a dynamic
absorber by adjusting its preload.
* D. J. Jones, Material Damping, ASA Damping Conference, Cleveland, Ohio, Nov. 21,
1968.
SEC. 3-9 129
Low frequency or
High frequency or
temperature high temperature
I
< .- I
Glassy Transition I Rubbery I
I
I
I
Loss coefficient
I
I
I
Complex modulus
I
I I
I I I
Increasing temperature (at constant frequency)

Decreasing frequency (at constant temperature)
FIG. 3-46. Temperature and frequency of a "typical" polymer.
The frequency dependence of the complex modulus is often considered

as a disadvantage using visco-elastic material for isolation mounting.
However, for a tuned dynamic absorber with damping (see the
frequency dependency also moves the "resonant frequency" of the ab-
sorber progressively higher or with the exciting frequency. Thus,
the absorber is effective for a greater frequency range of operation.*
3-9 SUMMARY
theory of systems is applied to a variety of

problems, the model of which is as shown in Fig. 2-6. The generalized
equation of motion is
The examples in the chapter are grouped to illustrate the equivalent

quantities and The emphasis is on problem formula-
tion and interpretation, since the general theory was developed in the last
chapter.
The equivalent mass and equivalent spring are illustrated in
Examples 1 to 8.
* J. C. Vibration and Shock in Damped Mechanical Systems, John Wiley

Sons, New York, 1968, p. 96.
130 Systems One Degree of Freedom-Applications CHAP. 3
Damped free vibration is examined in 3-3. The logarithmic decre-

ment in Eq. (3-12) is a convenient way to measure the damping in a
system. As shown in Fig. 3-10, it takes relatively few cycles for the
transient vibration to die out, even for lightly damped systems.
Examples of in vibration testing and control for undamped systems
are shown in 3-4. A lightly damped system can be considered as
undamped, except at near resonance. The harmonic response of an
undamped system is theoretically infinite at resonance, Eq. (3-16). It
takes time, however, for the amplitude to build up at resonance, Eq.
(3-18). In practice, if a resonance frequency is passed through quickly, a
machine may operate at a desired speed between resonant frequencies.
Systems with damping under harmonic excitation are treated in
3-5 in six cases. The rotating unbalance in machines in Case 1 and critical
speed of shafts in Case 2 are two views of the same problem, because the
excitation in both are due to an unbalance in the machine. Hence the
response curves in Fig. 3-16 can be used to present both cases.
Vibration isolation and in Case 3 and the response to
moving support in Case 4 are again two aspects of the same problem as
evident by comparing Eqs. (3-32) and (3-37). The response of both cases
are shown in Fig. 3-25.
The seismic instrument in Case 5 uses the relative displacement
between a mass and its base to measure the motion of the base. The
equation of motion in Eq. (3-39) is of the same form as those for Cases 1
and 2. Hence the response can also be represented in Fig. 3-16, although
they are distinct types of problems. Two types of instruments can be
constructed from this general theory, depending on the excitation fre-
quency and the'natural frequency of the system. If we have
a vibrometer for displacement measurement and = regardless
of the damping as shown in Fig. 3-16. If we have an accelerome-
ter and is proportional to Appropriate damping is
an accelerometer in order to minimize the amplitude and phase distor-
tions.
A periodic excitation can be expressed as a Fourier series. The system
response due to each harmonic component of the periodic excitation
be calculated as illustrated in Fig. 3-37. The total response is obtained by
superposition as shown in Eq. (3-65). Hence this is a generalization of the
harmonic excitation in 3-5.
Shock is a transient phenomenon. The shock spectrum in 3-7
describes the effect of shock on the assumption that shock excitations
having the same spectrum would produce similar effect on the system,
Damping is seldom purely viscous in a real system. For nonviscous
damping, an viscous damping coefficient is defined in
Eq. (3-80). The criteria for equivalence are (1) equal energy dissipation
per cycle of vibration, and (2) harmonic oscillations of the same
amplitude.
Problems 131
PROBLEMS
Assume all the systems in the figures to follow are shown in their static
equilibrium positions.
3-1 Find the natural frequency of the system in Fig. Assume that (1)
the cantilevers are of negligible mass and (2) their equivalent spring
constants are k, and
, Pulley
PIG. P3-1.
Neglecting the mass of the pulleys, find the natural frequency of the system
in Fig.
A mass m is attached to a rigid bar of negligible mass as shown in Fig.
Find the natural frequency of the system, if (a) the bar is con-
strained to remain horizontal while m oscillates, vertically; (b) the bar is free
to pivot at the hinges A and B. (c) Show that the natural frequency
determined in part a is higher than that of part b.
A mass m is suspended as shown in Fig. If the beam is of negligible

mass and its deflection is given by the equation = find the
natural frequency of the system.
Referring to Fig. let k 7 and m 18 kg. (a) Find the natural
frequency of the system. If the ends the spring are fixed and the mass
m is attached to the midpoint of the spring, find the natural frequency. If

the ends of the spring are fixed and m is attached to some intermediate
point of the spring, show that the natural frequency of this configuration is
higher than that of part b.
3-6 A rocker arm assembly is shown in Fig. Let =mass of rocker
arm and =the mass moment of inertia about the pivot A. Find the
equivalent mass and the equivalent spring of the system referring to
the coordinate.
Push rod Spring
3-7 An engine valve arrangement is shown in Fig. where is the

mass moment of inertia of the rocker arm about the pivot A. Assume the
effective mass m, and the effective stiffness of the are known.
Reduce the valve arrangement to an system.
3-8 A mechanism is shown schematically in Fig. Assuming that the
tension of the spring is constant, derive the equation of motion of the
system.
rnrn
Uniform bar
mass
Disk
9
Problems 133
3-9 A machine component is depicted as a pendulum shown in Fig.

Determine its natural frequency by: (a) Newton's second law; the energy
method.
3-10 The mass moment of inertia of a connecting rod of mass m is deter-
mined by placing the rod on a horizontal platform of mass m, and timing
the periods of oscillation. The platform, shown in Fig. is sus-
pended by equally spaced wires. With the platform empty and an amplitude
of the period is With the mass center of the rod coinciding with that
of the platform and an amplitude of the period is Find of the
connecting rod.
Rotor
3-11 The mass moment of inertia of the rotor of an electrical generator of

mass m is found by attaching a small mass at R, from its
longitudinal axis and timing the periods oscillation. The test setup is
shown in Fig. (a) Find of the rotor. (b) Show that small variation
of R will have the least effect when R, = + l)R.
3-12 A machine component is shown in Fig. The mass m is constrained
by rails to move only in the x direction. Neglecting the mass of the arm, find
the equation of motion.
3-13 A machine part is schematically shown in Fig. Find the equation of

motion.
3-14 Find the equation of motion of the mass m for the system shown in Fig.
Assume that the horizontal bar is rigid and is of negligible mass.
3-15 A walking-beam configuration, consisting of a uniform beam of mass

and a cylindrical float of cross-sectional area A, is shown in Fig. If
the mass of the float and the rod is determine the equation of motion.
..
3-16 For a one-degree-of-freedom system, if, m = 7 kg, k = 6 and c =
35 N . find: (a) the damping factor (b) the logarithmic decrement and
(c) the ratio of any two consecutive amplitudes.
3-17 From the data in Prob. 2-14, find the logarithmic decrement for each of the
given sets of initial conditions.
3-18 A device bought from a surplus store is depicted as a
freedom system. It is desired to find: (a) its natural frequency, (b) the mass
moment of inertia of the rotor, and (c) the damping required for it to be
critically damped. It is not possible, however, to disassemble the device. It is
found that (1) when the rotor is turned a of 0.35 N . m is needed
to maintain this position, (2) when the rotor is held in this position and
released, it swings to -5.5' and then to and (3) the time of the swing is
1.0 s. Calculate the required information.
3-19 Derive the equations of motion for the systems in Fig. P3-8. Assume the
bars are rigid and of negligible mass.
Problems 135
3-20 Derive the equations of motion for the system shown in Fig. if an
excitation force F sin is applied to: (a) the mass m the free end A of
the bar.
3-21 A force F sin ot is applied to the mass m of the system shown in Fig.
If = (1 determine the motion of Assume zero initial
conditions and 1.
3-22 A harmonic motion is applied to each of the systems shown in Fig. P3-9.
Derive the equations of motion.
e cos
... .
sin
3-23 Referring to Fig. the general position of the fulcrum can be above
or below the static equilibrium position of the system and the bar can be
rotated clockwise or counterclockwise. Show that the equation of motion
derived in 3-22 is for all positions of the system.
3-24 A wide-flange I beam is cantilevered from the foundation of a building. The
beam is 2 m in length with a total mass of 60 kg. The I of the beam section
is 30 mm4. A construction worker places a small electric motor of 4 kg
at the end of the beam. The mass of the armature of the motor is 1.5 kg
with an eccentricity of 0.05 mm. If the motor speed is 3,600 estimate
the amplitude of vibration at the end of the beam.
3-25 Show analytically that the maximum value of the curves in Fig. 2-8 occur at
. a locus through the maxima of the curves.
Show analytically that the maximum values of the curves in Fig. 3-16 occur
at r Sketch a locus through the maxima of the curves.
3-27 A table for sorting seeds requires a reciprocating motion with a stroke of
1.0 mm and frequency from 2 to 20 Hz. The excitation is provided by an
eccentric weight shaker. The total mass of the table and shaker is 200 kg. (a)
Propose a scheme for mounting the table. Specify the spring constant,
the damping coefficient, and the unbalance of the exciter.
3-28 A machine of 100 kg mass has a 20 kg rotor with 0.5 mm eccentricity. The
mounting springs have k = 85 and the damping is negligible. The
operating speed is rpm and the unit is constrained to move vertically.
(a) Determine the dynamic amplitude of the machine. (b) Redesign the
mounting so that the dynamic amplitude is reduced to one half of the
original value, but maintaining the same natural frequency.
3-29 A variable-speed counter-rotating eccentric-weight exciter is attached to a
machine to determine its natural frequency. With the exciter at 1,000 rpm, a
136 Systems with One Degree Freedom-Applications CHAP. 3
stroboscope shows that the eccentric weights of the exciter are at the top the
instant the machine is moving upward through its static equilibrium posi-
tion. The amplitude of the displacement is 12 mm. The mass of the machine
is 500 kg and that of the exciter is 20 kg with an unbalance of m.
Find (a) the natural frequency of the machine and its mounting and the
damping factor of the system.
3-30 A rotating machine for research has an annular clearance of
between the rotor and the stator. The mass of the rotor is 36 kg with an
unbalance of 3 kg . m. The rotor is mounted symmetrically on a
round shaft, 300 mm in length and supported by two bearings. The operat-
ing speed ranges from 600 to 6,000 rpm. If the dynamic deflection of the
shaft is to be less than 0.1 mm, specify the size of the shaft.
3-31 A circular disk of 18 kg is mounted symmetrically on a shaft, 0.75 m in
length and 20 mm in diameter. The mass center of disk is 3 mm from its
geometric center. The unit is rotated at 1,000 rpm and the damping factor
is estimated to be 0.05. (a) Compare the static stress of the shaft with the
dynamic stress at the operating speed. (b) Repeat part a for a shaft of
30 mm in diameter.
3-32 It is proposed to use a three-cylinder two-stroke-cycle diesel engine to drive
an electric generator at 600 rpm. The generator consists of a 2 kg rotor
mounted on a hollow shaft, 2 m in length with a 200-mm O D and a 100-mm
bore. A preliminary test shows that, when the rotor is suspended horizon-
tally with its axis 0.95 m from the point of suspension, the period of
oscillation is s. If you are the consulting engineer, would you approve
this proposal?
3-33 A turbine at is mounted as shown in Fig. The 14 kg
rotor has an unbalance of m. (a) Neglecting the mass of the
shaft, find the amplitude of vibration and the at each bearing for a
diameter shaft. (b) Repeat for a 25-mm shaft. (c) Repeat for a
30-mm shaft. (d) Estimate the error due to neglecting the mass of the shaft.
3-34 A 180-kg steel disk is on a 100-mm O D and 75-mm I D shaft as

shown in Fig. (a) Neglecting the flexibility of the bearing supports,
find the critical speed of the (b) If the bearings are flexible with a
spring constant k = 70 in any direction normal to the shaft axis,
find the change in critical speed. Assume that the mass of the shaft and the
gyroscopic effect of the disk are negligible.
Problems 137
3-35 A 140-kg disk is mounted on a 75-mm diameter shaft as shown in Fig.

The bearing supports are essentially rigid in the vertical direction
but are flexible with a spring constant k = 50 each in the horizon-
tal direction. Find the critical speeds of the assembly.
3-36 Show analytically that the crossover points of the transmissibility curves in
Figs. 3-25 and 3-26 occur at =
3-37 A refrigeration unit of 30-kg mass operates at The unit is
supported by three equal springs. (a) Specify the springs if 10 percent or less
of the unbalance is transmitted to the foundation. (b) Verify the calculation,
using Fig. 3-27.
3-38 A vertical single-cylinder diesel engine of 500-kg mass is mounted on
springs with k = 200 and dampers with = 0.2. The rotating parts are
well-balanced. The mass of the equivalent reciprocating parts is 10 kg and
the stroke is 200 Find the dynamic amplitude of the vertical motion,
the transmissibility, and the force transmitted to the foundation, if the
engine is operated at (a) 200 (b) at 600 rpm.
3-39 A 50-kg rotor is mounted as shown in Fig. It has an unbalance
0.06 kg - m and operates at 800 rpm. If the dynamic amplitude of the rotor
is to be less than 6 mm and it is desired to have low transmissibility, specify
the springs and the dampers for the mounting.
FIG.
3-40 A 15-kg electric motor is supported by four equal springs as shown in Fig.
The stiffness of each spring is 2.5 The radius gyration of
the motor assembly about its shaft axis is 100 mm. The operating speed is
Find the transmissibility for the vertical and the torsional
vibrations.
3-41 An instrument in an aircraft is to be isolated from the engine vibrations,
ranging from 1,800 to 3,600 cycles per minute. If the damping is negligible
and the instrument is of 20-kg mass, specify the springs for the mounting for
80 percent isolation.
3-42 A 250-kg table for repairing instruments is isolated from the floor by
springs with k = 20 and dampers with c = 4 . If the floor
vibrates vertically 2.5 mm at a frequency of 10 Hz, find the motion of the

table.
3-43 Referring to the vehicle suspension problem in Example 19, if = 0.1 m,
find the amplitude when the speed of the trailer is: (a) 70
120
3-44 A body m, mounted as shown in Fig. is dropped on a floor.
Assume that, when the base first contacts the floor, the spring is unstressed
and the body has dropped through a height of 1.5 m. Find the acceleration
of m. If m = 18 kg, c = 7 2 N . and k = 1.8 determine the
maximum acceleration of m.
Pendulum
Base
I I
Base
3-45 A vibrometer for measuring the rectilinear motion is shown in Fig.

The pivot constrains the pendulum to oscillate in the plane of the
paper and viscous damping exists at the pivot. Derive the equation of
motion of the system.
3-46 A torsiograph is a seismic instrument to measure the speed fluctuation ot a
shaft. A torsiograph consisting of a hollow cylinder of 0.5 kg
40-mm radius of gyration is mounted coaxially the shaft and connecting
to it by a spiral spring. Assuming that (1) viscous damping exists between
the and the shaft, (2) the average shaft speed is 600 and (3)
the of fluctuations varies from 4 to 8 times the shaft speed,
specify the spring constant and the damping coefficient if the torsiograph is
to measure relative displacement.
3-47 A vibrometer to measure the vibrations of a variable speed engine is
schematically shown in Fig. 3-29. The vibrations consist of a fundamental
and a second harmonic. The operating speed ranges from 500 to 1500
It is desired to have the amplitude distortion less than 4 percent. Determine
the natural frequency of the vibrometer if: (a) the damping is negligible;
the damping factor = 0.6.
3-48 An accelerometer with 0.6 is used to measure the vibrations described in
4-47. The amplitude distortion is to be less than percent. (a) From
Fig. 3-32, find the natural frequency of the accelerometer. If the engine
operates at 1,000 find the amplitude distortion of the second harmonic.
(c) Find the phase distortion from Fig. 3-34 and calculate the phase shift in
unit of time.
3-49 Find the Fourier series of each of the periodic functions shown
in Fig. P3-13.
3-50 From the answers in 3-49, find the Fourier spectrum of periodic
functions in Figs. and (b).
3-51 Find the Fourier series expansions of the periodic functions in Fig. P3-13 if
each of the functions is delayed by an amount
3-52 From the answers in Prob. 3-51, find the Fourier spectrum of the periodic
functions of parts a and b.
The excitation of a system has two harmonic components.
(a) Sketch the wave form of the excitation. Use the impedance method
in Eq. (3-62) and (3-63) to find the steady-state response due to each of the
components. (c) Sketch the wave form of the composite
state response.
3-54 If the system in Fig. 3-38 is actuated by a cam with the profile as shown in
Fig. find the steady-state response of the system. Assume that
170 kg, = k 7 c = 1.7 slm, total cam lift 50 mm, and
the cam speed = 60
3-55 Derive the transmissibility equation for each of the systems shown in Fig.
P2-5. (See Prob. 2-24, Chap. 2).
3-56 A periodic force, with the as shown in Fig. is applied to
a mass-springsystem. Will there be a resonance if the fundamental frequency
of the excitation is one-half of the natural frequency of the system?
3-57 Find the relative motion of the mass m in Fig. 3-23, if the base
is given an excitation: (a) g; where C and a are constants.
Assume that the damping is negligible and the system is 'initially at rest.
3-58 For the system with Coulomb damping shown in Fig. 3-45, deduce from
energy considerations that the amplitude decay for the free vibration is
per cycle, where is the frictional force.
3-59 For the system with Coulomb friction shown in Fig. 3-45, assume that
m = 9 kg, k = 7 and the friction coefficient 0.15. If the initial
conditions are = 25 mm and = 0, find: (a) the dectease in displace-
ment amplitude per cycle, (b) the maximum velocity, (c) the decrease in
velocity amplitude per cycle, and (d) the position at which the body m
would stop.
3-60 For the system with Coulomb damping in Fig. 3-45, let an excitation
ot be applied to the mass m. (a) Use Eq. (3-75) to show that the
energy dissipation per cycle is where F is the frictional force. (b) Show
that the transmissibility TR is infinite at resonance. (c) Find TR for the
frequency ratios r where r =
3-61 For a one-degree-of-freedom system with velocity-squared damping, as-
sume the force applied to the mass is wt. (a) Find the resonance
amplitude from energy considerations. (b) Check part a from the expression
for X in Example 26.
3-62 A machine of 350-kg mass and 1.8-kg. m eccentricity is mounted on
springs and a damper with velocity squared damping. The damper consists
of a 70-mm diameter cylinder-piston arrangement. The piston has a nozzle
for the passage of the damping fluid, the density of which is = 960
The natural frequency of the system is 5 Hz. Assuming that the equivalent
viscous-damping factor = 0 2 at resonance, determine: (a) the resonance
amplitude; (b) the diameter of the nozzle if the pressure drop across the
nozzle is p = where (velocity) is that at the throat of the
nozzle.
Computer problems:
3-63 Use the program listed in Fig. to find the free vibration of
the system + + = Choose appropriate values for
and Select about two cycles for the duration of the run.
3-64 (a) Repeat Prob. 3-63 by modifying the program for plotting, as illustrated
in Fig. (b) Plot the results using the program listed in
Fig.
3-65 Consider the equation of motion and the harmonic response
+ cx + kx = sin
The response can be expressed as shown in Eqs. (3-25) and (3-26). (a)
Modify the program listed in Fig. to write the values of the
amplitude ratio and the phase angle in separate files for plotting.
Let the damping factor 0.2 and (b) Plot the results using
the program listed in Fig.
3-66 Repeat Prob. 3-65 but plot for a set of values of as in Figs.
2-10 and 2-11. Let 0.2, 0.3, 0.4, 0.55, 0.7, and 1.0.
Problems 141
3-67 (a) Classify and tabulate the harmonic response equations for Cases 1 to 5
in 3-5. (b) Write a program to plot the amplitude-ratio versus
frequency-ratio r for each type of response a range of damping factor
In other words, the object is to plot the response curves illustrated in Figs.
3-16, 3-25, 3-26. Hint: Modify the program as listed in Fig.
and use PLOTFILE, listed in Fig.
3-68 Repeat Prob. 2-32, 2-33, 2-34, or 2-35, but (1) modify the program for
plotting, as illustrated in Fig. and (2) write the program such that
only every nth data point is plotted. Use n = 2 for this program.
Systems with More Than
One Degree. of Freedom
4-1 INTRODUCTION
The degree of freedom of discrete systems was defined in Chap. 2.

Since there is no basic difference in concept between systems with two or
more degrees of freedom, we shall introduce multi-degree-of-freedom
systems from the generalization of systems with two degrees of freedom.
Computers, however, are mandatory for the numerical solution of prob-
lems with more than two degrees of freedom.
An n-degree-of-freedomsystem is described by a set of n simultaneous
ordinary differential equations of the second-order. The system has as
many natural frequencies as the degrees of freedom. A mode of vibration
is associated with each natural frequency. Since the equations of motion
are coupled, the motion of the are the combination of the motions
of the individual modes. If the equations are uncoupled by the proper
choice of coordinates, each mode can be examined as an independent
one-degree-of-freedom system.
To implement the topics enumerated above, computer sub-routines are
listed in App. C, such as for (1) finding the characteristic equation from
the equations of motion, (2) solving the characteristic equation for the
natural frequencies, (3) obtaining the modal matrix in order to uncouple
the equations of motion, and (4) solving the corresponding one-degree-
of-freedom system. Computer solutions are given as home problems. For
purpose of organization, typical computer programs are grouped and
illustrated in Chap. 9.
We shall begin by formulating the equations of motion from Newton's
second law, and then discuss natural frequencies, coordinate coupling and
transformation, modal analysis, and applications. The method of influence
SEC. 4-2 of Motion: Newton's Second Law 143
coefficients is presented in the latter part of the chapter. Orthogonality of

the modes and additional concepts for a better understanding of the
material will be discussed in Chap. 6.
4-2 EQUATIONS OF MOTION: NEWTON'S

SECOND LAW
The equations of motion for the two-degree-of-freedom system in Fig.
can be derived by applying Newton's second law to each of the
masses. Assume the damping is viscous and the displacements and
measured from the static equilibrium positions of the masses.
Summing the dynamic forces in the vertical direction on each mass shown
in the free-body sketches, we get
which can be rearranged to
where and are the excitation forces applied to the respective

masses. Note that the equations are not independent, because the equa-
tion for contains terms in and Hence the coupling terms in the
first equation in Eq. (4-1) are Similarly, the coupling terms
in the second equation are + kx,). In other words, the motion
of is influenced by the motion of and vice versa. Coordinate
coupling will be in detail in 4-4.
(a) System (b) Free-body sketches
FIG. 4-1. A two-degree-of-freedom system.

144 Systems More One Degree of Freedom CHAP. 4
For conciseness, Eq. (4-1) can be expressed in matrix notations as
By simple matrix operations, it can be shown that Eqs. (4-1) and (4-2) are
equivalent. The quantities in Eq. (4-3) can be identified by comparing
with Eq. (4-2). The matrices M, C, and are called the mass
matrix, damping matrix, and matrix, respectively. The 2 1 matrix
{x) is called the displacement vector. The corresponding velocity vector is
{x) and the acceleration vector is {f}. The 2 1 matrix is the force
vector.
It will be shown in 4-4 that if another set of coordinates is
used to describe the motion of the same system, the values of the
elements in the matrices M, C, will differ from those shown in Eq.
(4-2). The inherent properties of the system, such as natural frequencies,
must be independent of the coordinates used to describe system.
Hence the general form of the equations of motion of a
freedom system is
The 2 2 matrices M, and K associated with the coordinates {q) can be

identified by comparing the last two equations. The 2 1 matrix is
the force vector associated with the displacement vector {q).
Generalizing the concept, Eq. (4-5) also describes the motion of an n-
degree-of-freedom system if the matrices M, and are of nth-order,
that is,
C
where i, = . . . , n. The coefficients and are the elements

of the matrices M, C, and respectively. The generalized coordinates {q)
and the generalized force vector are
Vibration: 145
Hence Eq. (4-5) is also the general form of the equations of motion of an
n-degree-of-freedom system.
4-3 UNDAMPED VIBRATION: PRINCIPAL

MODES
A dynamic system has as many natural frequencies and modes of
vibration as the degrees of freedom. The general motion is the superposi-
tion of the modes. We shall discuss (1) a method to find the natural
frequencies, and (2) the modes of vibration of an undamped system at its
natural frequencies.
In the absence of damping and excitation, the system in Fig. 4-1
reduces to that in Fig. Hence the equations of motion from
Eq. (4-2) are
The equations are linear and homogeneous and are in the form of Eq.
App. D. Hence the solutions can be expressed as
Motions at first mode
Time
Time
Motions at second mode
Vibratory system at modes
FIG. 4-2. of
146 Systems More Degree of CHAP. 4
where and are constants. Since the system is undamped, it can

be shown that the values of are imaginary, By
formula, cos ot sin ot, and recalling that the x's are real, the
solutions above must be harmonic and the general solution must consist
of a number of harmonic components.
Assume one of the harmonic components is
where and are constants and is a natural frequency of the

system. If the motions are harmonic, the choice of sine or cosine functions
is arbitrary.
Substituting Eq. (4-11) in dividing out the factor + #), and
rearranging, we have
which are homogeneous linear algebraic equations in and The

determinant of the coefficients of and is the charac-
teristic determinant. If is equated to zero, we obtain the characteris-
tic or the frequency equation of the system from which the values of are
found, that is,
From algebra, (4-12) possesses a solution only if the

ant is zero.
Expanding the determinant and rearranging, we get
which is quadratic in This leads to two real and positive values* for
Calling them and the values of from Eq. (4-14) are and
Since the solutions in Eq. (4-11) are harmonic, the negative signs for
merely change the signs of the arbitrary constants and would not lead
to new solutions. Hence the natural frequencies are and
The example shows that there are two natural frequencies in a
degree-of-freedom system. Each of the solutions of Eq. (4-9) has two
harmonic components at the frequencies and respectively. By
* Note that the values of s in (4-10) are and in order to have the periodic
solutions assumed in Eq. (4-11). If is not real and positive, it can be shown that
solutions by (4-10) would either diminish to zero or increase to with increasing
time.
SEC. Free Vibration: Principal Modes
superposition, the solutions from Eq. are
where the A's and are arbitrary constants. The lower frequency term
is called the fundamental and the others are the harmonics. Double
subscripts are assigned to the amplitudes; the first subscript refers to the
coordinate and the second to the frequency. For example, A,, is the
amplitude of at the frequency =
The relative amplitudes of the harmonic components in Eq. (4-15) are
defined in Eq. (4-12). Substituting and in Eq. (4-12) and rearrang-
ing, we obtain
k k+ -- =-
A,, k+ - k
(4-16)
k
k+ - k
where the are constants, defining the relative amplitudes of and

at each of the natural frequencies and Thus, Eq. (4-15) becomes
+ + A
where A,,, and are the constants of integration, to be deter-

mined by the initial conditions. There are four constants because the
system is described by two second-order differential equations. Note that
(1) from the homogeneous equation in Eq. only the ratios
and : can found, and (2) the relative amplitudes at a given natural
frequency are invariant, regardless of the initial conditions.
A principal or natural mode of vibration occurs when the entire system
executes synchronous harmonic motion at one of the natural frequencies
as illustrated in Fig. For example, the first mode occurs if A,, =
in Eq. that is,
where is called a modal vector or eigenvector. Note that

as above. It represents the relative amplitude, or
the mode shape, of the motions and at w = w , . Hence a
principal mode is specified the modal vector a t the given natural
frequency. The quantity = is It shows
148 Systems with More Than One Degree of Freedom CHAP. 4
that the entire system executes synchronous harmonic motion at a princi-

pal mode. Similarly, the second mode occurs if in Eq. (4-17) is zero,
that is,
The modal vector for second mode is

The harmonic functions of the motions and in Eq. (4-17) can
be expressed as
where the modal matrix [u] is
and = + and = + Note that in

Eqs. (4-18) through (4-20) only the relative values in a modal vector can
be defined, as shown in Eq. (4-16). A modal matrix [u] in Eq. (4-22) is
simply a combination of the modal vectors. The actual motions {x) in Eq.
(4-20) are specified constants A's and which are determined
by the initial conditions.
The vector {p) consists of a set of harmonic functions at the frequencies
and The vector {p) is called the principal coordinates. Each
principal coordinate and its associated modal vector describe a
mode of vibration as shown in Eqs. and (4-19). Principal coordi-
nates will be further discussed in 4-5 and Chap. 6. The coordinate
transformation between the {x) and {p) coordinates is shown in Eq.
(4-21).
The extension of the concept to n-degree-of-freedom systems is im-
mediate. For example, a system may be described by the
{x, ... x,) coordinates. Analogous to Eq. each of motions
has n harmonic components. A principal mode occurs if the entire
system executes synchronous harmonic motion at one of the natural
frequencies. The corresponding principal coordinates is
.. . The modal matrix [u] consists of n modal vectors
where i, = . . ., n. The transformation between the and the

coordinates is analogous to that in Eq. (4-21).
SEC. 4-3 Undamped Free Modes 149
Example 1
Referring to Fig. let = = m and = = k. If the initial
conditions are = 0 ) and = find the natural frequen-
cies of the system and the displacement vector {x).
Solution:
From Eq. the natural frequencies are
and
Substituting and in Eq. we obtain = and Hence
the displacement vector { x ) from Eq. (4-20) is
For the initial conditions = we get
Premultiplying the equation by the inverse of [ u ] gives
1 1
and A,,=-
2 sin +, 2 sin
For the initial conditions = we have
Premultiplying the equation by the inverse of [ u ] gives
Since the A's and to's are nonzero, we have cos = cos Let =
and = where m and are odd integers. It can be shown that
the choice of m and n other than 1 will not to new solutions. Thus,
From Eq. we obtain
The motions are plotted in Fig. 4-3 for 2m. The example can be
repeated f o r different initial conditions to show that the relative amplitudes
of the principal modes remain unchanged. is left as an exercise.
FIG. 4-3. Superposition of modes of vibration: Example 1.
Example 2. Natural modes

Find the initial conditions that would set a two-degree-of-freedom into its
natural modes of vibration, that is, or in Eq. (4-17) becomes zero.
Solution:
From Eq. we have
Applying the initial conditions

I
we get
1 sin
Premultiplying the equation by the inverse of [u] gives
and
- - $2
Similarly, using the initial conditions we get

-
and
$2
SEC. 4-3 Undamped Free Vibration:
The first mode occurs at if that

= and =
In other words, the system will vibrate at its first mode if the initial
conditions are ={ 1 with zero initial velocities. Alternatively,
the initial conditions can be = with zero initial displace-
ments. Any combination of the above conditions would also set the system
in its first mode. It is only necessary to set the initial values of {x) and/or
to conform to their relative values for the first mode, as indicated by the
corresponding modal vector ={1
Similarly, the second mode occurs when that is, = and
= Any combination of these conditions will give the second mode.
Example 3. Vehicle suspension

An automobile is shown schematically in Fig. 4-4. Find the natural frequen-
cies of the car body.
Solution:
An automobile has many degrees of freedom. Simplifying, we assume that
the car moves in the plane of the paper and the motion consists of (1) the
vertical motion of the car body, (2) the rotational pitching motion of the
body about its mass and (3) the vertical motion of the wheels. Even
then, the bas more than two degrees of freedom.
When the excitation frequency due to the road roughness is high, the
wheels move up and down with great rapidity but little of this motion is
transmitted to the car body. In other words, the natural frequency of the car
body is low and only the low frequency portion of the road roughness is
being transmitted. (See Case 4 in 3-5.) Because of this large separation
of natural frequencies between the wheels and the car body, the problem
can be further simplified by neglecting the wheels as shown Fig. 4-5.
Assuming small oscillations, the equations of motion in the and
coordinates are
mi= (forces),
=- - -
Car body
FIG. 4-4. Schematic of an automobile.

Systems with More One of Freedom CHAP. 4
FIG. 4-5. Simplified representation of an automobile body.
and
= (moments),
= - -
Rearranging, we obtain
which is of the same form as Eq. (4-9). The frequency equation from Eq.
(4-13) is
Expanding the determinant and solving the equation, we get
The natural frequencies are and Hz.
Example 4
A vehicle has a mass of 1,800 kg (4,000 Ib,) and a wheelbase of 3.6 m
(140 in.). The mass center cg is 1.6 m (63 in.) from the front axle. The radius
of of the vehicle about is 1.4 m (55 in.). The spring constants of
the front and the rear springs are 42 and 48
(275 respectively. Determine (a) the frequencies, (b) the
principal modes of vibration, and the motion and of the vehicle.
SEC. 4-3 Undamped Free Vibration: Principal Modes 153
( a ) First mode: f = 1.09 Hz (b) Second mode: f 1.50 Hz
F I G . 4-6. Principal modes of vibration a car body (not to scale).
Solution:
(a) From the given data and the equations in Example 3, we have
amplitude ratios for the two modes of vibration are

X -
+ -
The two principal modes of vibration are shown schematically in Fig.
4-6. The mode shape at 1.09 Hz is { X Thus, when
is positive, is negative from the assumed direction of rotation.
When =1 = -114.69 rad, that is, the node is 4.69 m from
the cg of the car body. Similarly, at the mode shape is
{X
From Eq. the and motions are
where and are the constants of integration.
Example 5. A three-degree-of-freedom system
A torsional system with three degrees of freedom is shown in Fig. 4-7. (a)
Determine the equations of motion and the frequency equation. (b) If
Systems More One of Freedom CHAP. .
F IG. A three-degree-of-freedom torsional system; Example 5.
= = = and = k,, = = k,, find the natural frequencies and the

equation for the displacement
Solution:
(a) From Newton's second law, the equations of motion are

=- - -
- -
Substituting = +), for = and 3, in
factoring out the term, and rearranging, we have
The frequency equation is obtained by equating the determinant

of the coefficients and to zero.
(b) If = = = J and = = = the frequency equation is
The roots of the equation are 0 2 = 0 . 1 9 8 ( k J J ) , and

The corresponding frequency vector is
From Eq. the amplitude ratios are
and
+
SEC. Coordinate 155
FIG.4-8. Principal modes of vibration; amplitudes normal to axis

of rotation; Example 5.
Thus, a modal vector can be calculated for each of the
natural frequencies The modal matrix for = and 3, is
formed from a combination of the modal vectors as in Eq. (4-21).
1
= = -1.247
2.25 -0.802 0.555
where and are the modal vectors for the frequencies
= and = respectively. The
principal modes of vibration are illustrated in Fig. 4-8.
By superposition of the principal modes, the motions of the rotating
disks are
The to be determined by the initial conditions.
4-4 GENERALIZED COORDINATES AND

COORDINATE COUPLING
The general form of the equations of motion of a
freedom system is shown in Eq. (4-4). For undamped free vibration, we
have
The system is described by the coordinates and which are

elements of the displacement vector {x). The coupling terms in the
equations are and We shall show that the values of the
Static equilibrium
(a) Static coupling Dynamic coupling Static and

dynamic coupling
FIG. 4-9. Generalized coordinates and coordinate coupling: a two-
degree-of-freedom system described b y the (x, 0), and
coordinates.
elements in the matrices M and K are dependent on the coordinates

selected for the system description.
A vibratory system can be described by more than one set of indepen-
dent spatial coordinates, each of which can be called a set of generalized
coordinates. We often use the displacements from the static equilibrium
positions of the masses and the rotations about the mass centers for the
coordinates. This choice is convenient, but, nonetheless, arbitrary. We
shall describe the system in Fig. 4-9 by the displacement vectors
Referring to Fig. and assuming small oscillations, the equations

motion in the coordinates are
Rearranging, we obtain
The coupling term, occurs only in the stiffness matrix,

and the system is said to be statically or elastically coupled (see Examples
3 and 4). If a static force is applied through the cg at point 1, the body
will rotate as well as translate in the direction. Conversely, if a torque
is applied at point 1, the body will translate as well as rotate in the 8
direction.
The same system is described by the coordinates in Fig.
The distance e is selected to give = If a static force is applied
at point 2 to cause a displacement the body will not rotate. Hence no
static coupling is anticipated in the equations of motion. During vibration,
however, the inertia force through cg will create a moment
Coordinate Coupling 157
about point 2, tending to rotate the body in the direction. Conversely, a

rotation 8 about point 2 will give a displacement at cg and therefore a
force in the direction. Hence. dynamic coupling is anticipated in
the equations.
The equations of motion in the coordinates are
The coupling terms axe associated with the inertia forces and the system is
said to be dynamically, or inertia, coupled.
Lastly, let the same system be described by the coordinates as
shown in Fig. It can be shown that the equations of motion are
the description, the equations are statically and dynamically

coupled.
Note that (1) the of coordinates for the system description is a
mere convenience, (2) the system will vibrate in its own natural way
regardless of the coordinate description, (3) the equations for one coordi-
nate description can be obtained from those for other descriptions, and
(4) coupling in the equations is not an inherent property of the system,
such natural frequencies.
The examples above show that the matrices M and K are symmetric,
that is, = and Symmetry is assured if the deflections are
from a position in space. This can be deduced from
reciprocity theorem in 4-9. Let us select a set of
generalized coordinates based on relative deflections to illustrate the
nonsymmetic in the equations of motion.
Example 6
Consider the system shown in Fig. 4-2 and assume the generalized coordi-
nates = and = that is, is proportional to the spring force
due to k. Find the equations of motion of the system.
Solution:
The coordinates ( q } and { x ) are related by
158 Systems More Degree of Freedom CHAP. 4
Replacing the displacement and vectors {x} and by {q) and

{q} in Eq. (4-9) for the same system, we get
4-5 PRINCIPAL COORDINATES
It was shown in the last section that the elements of the matrices M and
depend on the coordinates selected for the system description. It is
possible to select a particular set of coordinates, called the principal
coordinates, such that there is no coupling terms in the equations of
motion, that is, the matrices M and become diagonal matrices. Hence
each of the uncoupled equations can be solved independently. In other
words, when the system is described in terms of the principal coordinates,
the equations of motion are uncoupled, and the modes of vibration are
mathematically separated. Thus, each of the uncoupled equations can be
solved independently, as if for systems with one degree of freedom.
Assume an undamped two-degree-of-freedom system is uncoupled by
the principal coordinates {p). The corresponding equations .of motion
from Eq. (4-4) are
Expanding the equations gives
+ =
The solutions of the equations are
= A,, +
= +
where = = and the A's and are constants.
Evidently, each of the solutions above represents a mode of vibration as
discussed in 4-3. At a given mode, the system resembles an indepen-
dent one-degree-of-freedom system.
Now, assume the same system is described by the generalized coordi-
nates {q) and the equations of motion are coupled. From Eq. the
Coordinates 159
motions in the {q) coordinates are
= + +
where and are the modal vectors for the frequencies
and respectively.
Substituting Eq. in (4-32) and simplifying, we get
where is the inverse of the modal matrix [u], as defined in Eq.

(4-22). The transfornnation between the and the {q) coordinates in Eq.
(4-34) is identical to that shown in Eq.
The discussion implies that the equations of motion can be uncoupled
by means of a coordinate transformation. In other words, given the
coupled equations the {q) coordinates, the equations can be uncoupled
by substituting {p) for {q) as shown in Eq. (4-34). This can be done, but
the general theory 6-4 will be needed. In the mean time, we shall
illustrate with another example and then show the general in
the next section.
Example 7
Determine the principal coordinates for the shown in Fig. if
and
.
Solution:
From Example we have = 1 and Hence Eq. (4-33) becomes
The transformation above indicates that, if = + = -

the equations of motion in the coordinates are uncoupled. Let us further
examine this statement.
The equations of motion for the same system from Eq. are
+ 2 kx, - kx, =
* We assume that the matrices M and K in the generalized coordinates are symmetric. An
inherent symmetry in the system can he assumed. We shall not discuss nonsymmetric
matrices as illustrated in Example 6.
160 Systems with More One of Freedom CHAP. 4
Adding and subtracting the equations, we obtain

+ x,) + + =
or + =
Again, the equations are uncoupled if define = (x, + x,) and =

Since the amplitudes of oscillation are arbitrary, the factor
between the two of and in the problem is secondary.
4-6 MODAL ANALYSIS: TRANSIENT

VIBRATION OF UNDAMPED SYSTEMS
Consider the steps to solve the equations of motion of an undamped

system. From Eq. we get
(1)The equations can be uncoupled by means of the modal matrix [u]

and expressed in the principal coordinates { p )as shown in Example 7. (2)
Each of the uncoupled equations can be solved as an independent
degree-of-freedom system. (3) Applying the coordinate transformation in
Eq. the solution can be expressed in the or { q } coordinates as
desired. The steps enumerated are conceptually simple. Except for the
formula to uncouple the equations of motion, we have the necessary
information for the modal analysis of transient vibration of undamped
systems. For systems with more than two of freedom, however,
computer solutions, as illustrated in Chap. 9, are manditory to alleviate
the numerical computations.
The. modal matrix of undamped systems can be found by the method
described in the previous sections. From Eq. the equations of
motion of an unforced system are
A principal mode occurs if the entire system executes synchronous

.harmonic motion at a natural frequency Thus, the acceleration of is
- 0 2 q i , or { - 0 2 q )= Substituting for in Eq.
(4-36) and simplifying, we get
[ - o ZM + = (4-37)
Since the system is at a principal mode, the displacement vector { q ) is also
a modal vector at the natural o. In other words, in Eq.
(4-37) gives the relative amplitude of vibration of the masses of the
system at the given natural frequency.
Since Eq. (4-37) is a set of homogeneous algebraic equations, it
possesses a solution only if the characteristic determinant is zero;
SEC. 4-6 Modal Analysis: Transient of Systems 161
that is,
-
This is the characteristic of the frequency equation, which may be

compared with Eq. (4-13).Previously, the frequency equation was solved
by hand calculations as shown in Eq. (4-14). Computer solutions, how-
ever, are necessary for systems with more than two degrees of freedom.
Likewise, instead of solving for the modal vector by hand calculations as
shown in Eq. computers can be used to for {q) in Eq. (4-37).
A modal vector is found for each natural frequency. The modal matrix
[u] is formed from a combination of the modal vectors as shown in Eq.
(4-23).
Example 8
Find the coefficients of the frequency equation for the system shown in Fig.
( a ) Vibratory system
Excitations
(c) Transient response

FIG. 4-10. Transient vibration of undamped system: Examples 8 to 10.
162 Systems More One Degree of Freedom
Solution:
Applying Newton's second law to each of the masses of the system, we have
- - - - +
= - - x,) - - +
=- - - - - +
The program COEFF to find the coefficients of the frequency equation is

listed in Fig. The valuesof the matrices M and are entered (READ)
and verified (WRITE)in Par. #I. The computationsin Par. first change Eq.
(4-36) into the form
+ =
The subroutineSINVS is called to find the inverse of M. The

subroutine SMPLY performs the matrix multiplication H,
where H = is the dynamic matrix. The subroutine is called
to give the coefficients of the frequency equation.
The print-out is listed in Fig. We first give the command
MERGE COEFF,
to merge the main program COEFF with the necessary subroutines. The
subroutine is for the matrix substitutions in the calculations. When
the computer is READY, the command RUN is given to start the program.
The data entries are self-explanatory. The is
1.4071
Assume the roots of the frequency equation in Example 8 are = 33.23,

86.67, and 246.8. Find the matrix of the system.
Solution:
For the given values of and and the direct application of
Eq. (4-37) gives
It can shown that the solution of the homogeneous algebraic equation is

1.126 where is the modal vector for
SEC. 4-6 Vibration of Systems
.. COEFFICIENTS OF
REQUIRED: SCOEFF SINVS
OF
ENTER: N
. IS THIS YES; 2
IS
CO + + + + ..+
VALUES OF CO TO CN
WRITE
WRITE
N,
WRITE N
.
I-1.N
WRITE
READ IANS
IF 10
AND OUTPUT.
SINVS
CALL K,
CALL C,
(6.85:
WRITE
STOP
COEFF, SINVS,
RUDY
OF
N
RCW
BOW
3 0 0 1 0 0 0 2 140 -60 -20 -60 220 -80 -20 100
0.0 0.0
0.0 0.0
0.0 0.0
IS THIS YES; 2 YO.

71
CO
VALUES OF TO
. +
F I G . 4-11. Program to find coefficients of frequency equation: Example

8.
33.23. Similarly, for and 246.8, the modal vectors are

(-0.381 0.429 and = -3.755 respectively.
Hence the modal matrix is
164 Systems More One of CHAP. 4
The equations of motion of undamped systems be uncoupled by the

orthogonal relations, which will be derived in 6-4. These are
where the transpose of the modal matrix [u] and and

are diagonal matrices. Substituting from (4-34)
we have
Premultiplying this by the transpose of [u] gives
where is the excitation associated with the {p) coordi-

nates. Since and are (4-41) gives the uncoupled
equations.
Describe a procedure to find the transient response of the system in Fig.

Assume the excitation for each of the masses is as shown in Fig.
The initial conditions are 2 -1) and
4 1).
Solution:
We shall first show the orthogonal relations in Eq. (4-40) and then describe
the procedure. Substituting the values of and [u] from Example 9 in
(4-40) yields
Similarly, it can be shown that the K matrix gives

SEC. 4-7 165
Hence, from Eq. the uncoupled equations of motion are

+ =
=
From Eq. the initial conditions expressed in terms of the principal

coordinates { p ) are
and =
The uncoupled equations in (4-42) and the initial conditions from Eq.
(4-43) are used to solve the problem in terms of the principal coordinates
{p). The resppnse of the masses in the { x ) coordinates are obtained by
means of the transformation { x )= in Eq.
Since the excitations shown in Fig. are arbitrary, they are
quantized and assumed to have constant values for each time interval
At = 0.05. The solutions are plotted in Fig. Although the procedure
is conceptually simple, computer solutions are mandatory. The problem is
solved by the program TRESPUND shown in 9-8.
4-7 SYSTEMS
A special case of practical importance occurs when a root of the
frequency equation vanishes. When a natural frequency is zero, there
is no relative motion in the system. The system can move as rigid body
and is called semidefinite.
Two semidefinite systems are shown in Fig. 4-12. The
system consists of a number of masses coupled by springs. It be used
to represent the vibration of a train. The rotational system may represent
a rotating machine, such as a diesel engine for marine propulsion. One of
the disks may represent the propeller, another disk the flywheel, and the
remaining disks the rotating and the equivalent reciprocating parts of the
engine.
(a) Rectilinear system
(b)Rotational system
FIG. 4-12. Semidefinite systems.
Systems with More Than One Degree of Freedom 4
(a) Vibratory system Principal mode
FIG. 4-13. Schematic of a motor-generator set.
Let an electrical motor-generator set be represented by a two-disk

system shown in Fig. The rotors and are connected by a
shaft of spring constant Summing the torques for each rotor about the
shaft axis, the equations of motion are
This is of the same form as Eq. (4-9). Hence the frequency equation from
Eq. (4-13) is
Expanding the determinant and dividing by we get
The roots of the equation are w2 = and 0

2
= +
Following the procedure in Eq. the relative amplitude of the
disks at the principal modes are
for
where = + as indicated in Eq. (4-46). When = and

= 1, the two disks have identical angular displacements. Since there
is no relative displacement between the disks, the shaft is not stressed and
the assembly rotates as a rigid body. This is called the zero mode. When
w= the two disks oscillate in opposite directions and the mode shape
is = as illustrated in Fig.
To extend the theory to systems with more than two degrees of
freedom, consider the three-disk assembly in Fig. The system
could be used to represent, to the first approximation, a
freedom system, such as a diesel engine for marine propulsion. From
SEC. Systems
Shaft 1 Shaft 2
(a) Vibratory system
Principal modes:
4-14. A system.
Newton's second law, equations of motion of the disks are
(4-48)
Similar to Eq. frequency equation of the system is
The relative amplitudes at the modes can be obtained from

Eq. (4-48) and expressed as
and -=
kt2
When w 2 = 0, the relative amplitudes of the disks are = = 1.

This indicates that the whole assembly may rotate as a rigid body. The
relative amplitudes of principal modes are shown in Fig.
Note that there is one sign change in the amplitudes for the first mode
and two sign changes the second mode.
Example Geared
Let a two-shaft system in Fig. be connected by a pair of gears. (a)
Neglecting the inertial effect of the gears, determine the frequencies of the
system. (b) Repeat part a but include the inertial effect gears.
Solution:
Since the two shafts are at different rotational speeds, it is expedient to find
an equivalent system referring to a common shaft. Let N, be the number of
with More Than One Degree of Freedom CHAP. 4
Equivalent system, refer to shaft 1,

neglect inertial effect of gears
Equivalent system, refer to shaft I ,

including inertial effect of gears
FIG. 4-15. Semidefinite geared system: Example
teeth on the pinion and that of the gear. shaft 1 as reference,

the equivalent inertia of is as shown in Example 1, Chap. 3.
Similarly, the equivalent spring constant of shaft 2 referring to shaft 1 is
=
(a) Referring to shaft 1 and neglecting the inertial effect of the gears, the
equivalent system is shown in Fig. The shafts are in series and
the equivalent spring constant is
1 1 1
+-
+
The: natural frequencies from Eq. (4-46) are
(b) Including the gears, the system has four disks and therefore four
equations of motion. of the pinion and of the gear can be
combined to give referring to shaft 1. Thus, the equival-
ent system consists of three disks and two shafts as shown in Fig.
The frequency equation would be identical to Eq. (4-49).
SEC. 4-8 Vibration-Humor& 169
Note that the natural frequencies are the frequencies of oscillation of

one disk relative to another, superposed on the rigid body rotation of
assembly. The natural frequencies can be calculated referring to one shaft
or the other. The proof of this statement is left as an exercise.
4-8 FORCED
The general form of the equations of motion of a two-degree-of-

freedom system is shown in Eq. (4-4). If the excitation is harmonic, the
equations can be solved readily by the impedance method developed in
2-6. The numerical solutions for systems with damping, however, are
tedious. Computers can be used to alleviate the calculations.
Applying the impedance method to Eq. (4-4), we substitute the har-
monic force vector for the generalized force where
and the 1 matrix is the phasor of All the
harmonic components in are assumed of the same frequency w. If not,
one frequency can be treated at a time and the resultant response
obtained by superposition. Let be the harmonic response, where
and is the phasor of Applying the impe-
dance method and factoring out Eq. (4-4) becomes
the matrices C, and K can be identified readily.

The equations above can be alternatively expressed as
where
= - + for i, 1,2
170 Systems with More One Degree of Freedom CHAP. 4
and = is the impedance matrix. In other words, Eqs. to

(4-55) are different forms of the same equation, all of which can be
summarized by Eq. (4-55).
The solution gives the amplitude and phase angle of the response
relative to the excitation Premultiplying both sides of Eq.
the inverse of gives
=
For a two-degree-of-freedom system this can be written explicitly as
= and (4-58)
Equations (4-55) to (4-57) are equally applicable to

freedom systems. The elements of the impedance matrix in Eq.
(4-55) become
= - + for i, 1 2 .. , (4-59)
and is of order n. Note that each is identical in form to the
mechanical impedance in Eq. (2-51) and is symmetric by the proper
choice of coordinates as discussed in 4-4.
Example Undamped dynamic absorber
Excessive vibration, due to near resonance conditions, is encountered in a

constant machine shown in Fig. The original system consists
of m, and k,. It is not feasible to change and (a) Show that a
dynamic absorber, consisting of and will remedy the problem. (b)
Plot the response curves of the system, assuming = 0.3. (c)
gate the effect of the mass ratio
(a) Vibratory system (b) Equivalent system
FIG. 4-16. dynamic absorber: Example 12.

SEC. Forced Vibration-Harmonic
Solution:
The equations of the equivalent system in Fig. are
= - - x,) + sin
=- -
The impedance method can be applied directly, since excitation is
harmonic. From we have
(a) Following Eq. the frequency equation is obtained by equating

the characteristic determinant of the coefficient matrix of to
zero, that is,
(4-60)
From Eq. 14-58), the phasors of the responses are
1
and
-
where the characteristic. determinant. that the amplitude
becomes at the excitation frequency = An undamped
dynamic is "tuned" for such that ap-
proaches zero at the resonance frequency of the original system.
(b) The frequency equation in (4-60) can be expressed as
Since = a frequency ratio is defined as

The frequency equation reduces to
From Eq. responses can be expressed as
The equations are plotted in Fig. 4-17 for = 0.3. The plus or
minus sign of the amplitude ratio denotes that the response is either
in-phase or out-of-phase with the excitation. Resonances occur at
= 0.762 and 1.311. Note that = when 0 2 m 2 = It can be
shown from Eq. (4-61) that this condition occurs when the excitation
sin is balanced by the spring force
(c) The frequency equation, is in Fig. 4-18 to show the

effect of the mass ratio When m is small, the resonant
172 Systems More One of Freedom CHAP. 4
4
0
0
-4
-6
0 2.0
Frequency ratio
FIG. 4-17. Typical harmonic response of a two-degree-of-freedom sys-

tem: Example 12.
frequencies are close together about the resonance frequency of the
original system. This means that there is little tolerance for variations in
the excitation frequency, although = when = Furthermore, it
is observed in Eq. (4-63) that the amplitude of the absorber at =
can be large for small values of When is appreciable, the
resonant frequencies are separated. For example, when
1.6
14
.
0
1.0
0.8
0.6
0 0.2 0.4 0.6 0.8
Mass ratio
FIG. 4-18. Undamped dynamic of mass ratio
Example 12.
SEC. 4-8 173
resonances occur at r equal to 0.73 and 1.37 times that of the original
system. The amplitude of the absorber mass is correspondingly
reduced at r 1 for larger mass ratios.
Example 13. Dynamic

Consider the dynamic absorber in 12 in which a viscous damper c
is installed in parallel with the spring k, as shown in Fig. Briefly
discuss the problem.
Solution:
From Eq. the equations of motion in phasor notations are
jwc - w 2m 2 +
From Eq. the corresponding frequency equation a
- - jwc w 2m 2+
From Eq. the phasors of the responses are
where is the characteristic determinant. The values of and can

be calculated using the programs in Chap. 9 .
The response curve of a properly tuned* dynamic absorber with approp-
riate damping is shown in Fig. Curve 1 is that of an undamped
system and curve 2 corresponding to Curve 3 of intermediate
damping must pass through the intersections of these curves.
Excitation frequency
(a) Dynamic absorber with damping Vibratory system
FIG. 4-19. Dynamic absorber with damping: Example 13.
J. D. Hartog, Mechanical Vibrations, 4th ed., Book Company, New

York, 1956, pp. 93-102. Note that the "tuned" condition of = in Example 12
is only for undamped absorbers. See 4-27 for of dynamic absorbers with
viscous damping.
Foundation
FIG. 4-20. Vibration isolation: Example 14.
14. Vibration
A constant speed machine is isolated as shown in Fig. and the
in the office complain of the annoying vibration transmitted from the
machine. It is proposed (1)to mount the machine on a cement block
as shown in Fig. 4-20, or (2) bolt rigidly- to Assume m 4,
= and the excitation frequency = = (a) the
magnitudes of and and the force transmitted to rigid
foundation. (b) Neglecting the damping in the system for the estimation,
would you approve proposal 1 or
Solution:
From the equations of motion in phasor notations are
The characteristic determinant can be identified from the equation

above. Equating to zero gives the frequency equation. Thus,
From Eq. the response of the system is
The force is transmitted to the foundation through the

and the damper Thus,
1
= ( k ,+ +
175
The solution and the force transmitted can be calculated. from

the equations above.
(b) Proposal 1: The frequency equation of the undamped system is
Substituting the given conditions = 4, etc., the frequency equation

be as
where r = = Correspondingly, we get
For the given values, resonance at and 1.14. At

or r = 2, we have = =
and =
Proposal 2: If is attached to the system has one degree of
freedom. equation of motion in phasor notations is
02(ml + =
Using the given data and the results by we get
and
1-5r2
Resonance occurs at At the excitation frequency =
or r = 2, = = and = = Hence
the force transmitted is higher than that in proposal 1.
4-9 INFLUENCE
The method of influence coefficients gives an alternative procedure to
formulate the equations of motion of a dynamic system. It is widely used
in the analysis of structures, such as an aircraft. A spring can be described
by its stiffness or its compliance, which is synonymous to the flexibility
influence coefficient. We shall first (1) show Maxwell's reciprocity
theorem, (2) relate the and flexibility matrices, and then (3)
illustrate the method of influence coefficients.
System More One of Freedom CHAP. 4
Station 1 7
................
.......
F I G . 4-21. Method of influence coefficients.
An influence defines the static elastic property of a

system. The quantity is the deflection at station i owing to a unit force
applied at station when this force is the only force applied. Consider the
beam shown in Fig. 4-21. The vertical force is applied at station 1 and
at station 2.
First, let be applied to station 1 and then to station 2 . When is
applied alone, the deflection at station 1 is The potential energy in
the beam, by virtue of its deflection, is Now, when is applied,
the additional deflection at station 1 due to is The by
corresponding to this deflection is Thus, the total potential
energy of the system due to and is
Secondly, let be applied to station 2 and then to station 1 . It can

be shown that the potential energy due to and is
The potential energies for the two methods of loading must be the
same, since the final states of 'the system are identical. Comparing the
expressions for we deduce that = for the system with two loads.
This is called Maxwell's reciprocity theorem.
For the general case, we have
for .... n (4-64)
which holds for all linear systems. When the force F is generalized to
represent a force or a moment, the influence coefficient correspond-
ingly represents a rectilinear or an angular displacement. Furthermore,
when the deflections due to the inertia forces are considered, we obtain
the equations of motion of the system.
During vibration, the inertia force associated with each mass is trans-
mitted throughout the system to cause a motion at each of the other
177
masses. For the undamped free vibration of a two-degree-of-freedom

system, we have
where {q} is a generalized displacement vector, the generalized

force vector of the inertia forces, and is the flexibility (influence
coefficient) matrix. For example, the deflection at station I is due to
the combined effect of the inertia forces and The total
deflection is +
From Eq. in the absence of dynamic coupling, the equations of
motion for the free vibration of an undamped system can be expressed as
where is the stiffnessmatrix. Eq. (4-66) by the inverse

of and rearranging, we get
Comparing the last two equations, it is evident that

= or
where I is a unit matrix. This is to say that is the inverse of and
vice versa.
Write the equations of motion for the system shown in Fig. by the
of influence and find the frequency equation.
To find the influence coefficients, let a unit static force be applied to m,.
The springs k and are in series and their combination is in parallel with
k,. Thus,
The deflection of is
Since the deflection of spring is inversely proportional to its stiffness and

the deflection of is it can be shown that the corresponding deflection
Systems of 4
Similarly, considering a unit static force at we get
Combining the influence coefficients yields
The equations of motion from Eq. (4-66) are
Note that the matrix from Eq. (4-9) is
It can be shown readily that = Moreover, the

of (4-70) by will give which are the equations of
motion of the same system by Newton's second law.
To find the frequency equation, we substitute for the second time
derivative and (4-69) in notation as
The frequency equation is obtained by the of

the coefficient matrix of to zero.
Substituting the values for expanding the determinant simp-

lifying, the frequency equation becomes
This is identical to the frequency equation in (4-14) by Newton's second

law for the same system.
SEC. Coefficients
Determination of influence coefficients
FIG. 4-22. Influencecoefficients due to force and moment; Example 16.
Example 16
Determine the natural frequency of the system shown in Fig.
Assume (1) the flexural stiffness of the shaft is (2) the inertia effect
of the shaft is negligible, (3) the shaft is horizontal in its static equilibrium
position, and (4) the mass moment of inertia of the disk is = mR 2 /4 where
R=
Solution:
The inertia forces are as shown in Fig. and the influence coefficients
are defined in Fig. From elementary beam theory, it can be shown
that the coefficients are
= = L2/2EZ
The equations of motion from Eq. are
Following the last example, this can be expressed in phasor notations as

shown in Eq. (4-71).
180 More of Freedom 4
The frequency is obtained by equating the determinant of the coeffi-

cients of to zero.
Substituting mR 2 /4 and R = and expanding we get

- + =
Hence
w= and
4-10 SUMMARY
The chapter introduces the theory of discrete systems from the
generalization of a two-degree-of-freedom system shown in Fig. 4-1. The
equations of motion in Eq. (4-1) through (4-4) are coupled, because the
equation for one mass is influenced by the motion of the other mass of
the system.
The modes of vibration are examined in 4-3 for undamped free
vibrations. The natural frequencies are obtained from the characteristic
equation in Eq. (4-13). A mode of vibration, called the principal mode, is
associated with each natural frequency. At a principal mode, (1) the
entire system executes synchronous harmonic motion at a natural fre-
quency and (2) the relative amplitudes of the masses are constant, as
shown in Eq. (4-16) and illustrated in Fig. The relative amplitudes
define the modal vector for the given mode. The general motion is the
superposition of the modes, as shown in Eq. (4-17).
A system can he described by more than one set of generalized
coordinates {q). In Eqs. (4-27) through it is shown that the
elements of the mass matrix and the stiffness matrix as well as the type of
coupling in equations of motion are dependent on the coordinates
for system description. Hence coordinate coupling is not an
property of the system. The coordinates that uncouple the
equations are called the principal coordinates The coordinates {p)
and {q} are related by the modal matrix [u] as shown in Eq.
A method for finding the modal matrix is shown in See. 4-6. The
equations of motion can be uncoupled by means of the modal matrix.
Thus, each uncoupled equation can be treated as an independent
degree-of-freedom system. The results can be expressed in the {p) or { q )
coordinates as desired. The technique is conceptually simple, but compu-
ters are necessary for the numerical solutions.
Many practical problems can be represented as semidefinite systems as
discussed in 4-7. A system is if it can move as a rigid
body. Correspondingly, at least one of its natural frequencies is zero.
Problems 181
The harmonic response of discrete systems can be found readily by the

mechanical impedance method. Using phasor notations, the equations of
motion can be expressed as = in Eq. (4-55) and the response
as = in Eq. (4-57).
The method of influence coefficients in 4-9 gives an alternative
procedure to formulate the equations of motion. From Maxwell's recip-
rocity theorem, the flexibility matrix is symmetric. The inverse
of the flexibility matrix is the stiffness matrix Thus, except for the
technique in obtaining the equations of motion and certain advantages in
its application, the concepts of vibration in the previous sections can be
applied readily in this method.
PROBLEMS
Assume the systems in the figures to follow are shown in their static
equilibrium
4-1 Consider the system in Fig. Let = 10 kg, = = 40
and k = 60 Nlm. (a) Write the equations of motion and the frequency
equation. (b) Find the natural frequencies, the principal modes, and the
modal matrix. (c) Assume = {1 0 ) and = {0 1). Plot and
and their harmonic components. 0 ) and
- 1). Find and
4-2 Repeat Prob. 4-1 if m, = = 10 kg, k , = 40 = 140 and k =
6b Are the periodic?
4-3 A 200-kg uniform bar is supported by springs at the ends as illustrated in
Fig. 4-5. The length is L = 1.5 m, = 18
L
and = 22
Write the equations of motion and the frequency equation. (b) Find the
natural frequencies, the principal modes, and the modal matrix. (c) If
1, = = = 0 , find the motions and ( d ) Illustrate
the principal modes, such as shown in Fig.
4-4 For the three-degree-of-freedom system in Fig. 4-7, if =
= and find the motions and
For each of the systems shown in Fig. P4-1, specify coordinates to
describe the write the equations of motion, and find the frequency
equation.
(a) A double pendulum.

(b) The arm is horizontal in its static equilibrium
(c) Three identical pendulums.
( d ) A double compound pendulum.
A schematic of an overhead crane.
(f) The system is constrained to move in plane of the paper.
The bar and the shaft are horizontal. The shaft deflects verti-
cally. The bar moves vertically as well as rotates in a vertical, plane.
182 Systems More Degree of Freedom CHAP. 4
Uniform bar
FIG. P4-1. Vibratory systems.

A spherical pendulum.
(i) The airfoil moves vertically and pivots about cg.
Assume that there is no friction between m and
The pendulums are constrained to move in the plane of the paper.
4-6 For the double pendulum in Fig. let = and = (a) If

is the horizontal displacement of and that of m,, write the equations
of motion in terms of and find the natural frequencies. (b) If and
are the angular displacements of the pendulums, write the equations of
motion in terms and find the natural frequeneies.
4-7 Referring to Fig. 4-9 on coordinate coupling, (a) convert Eq. (4-28) to
using the relations = - = - e, = + e, and =
and convert Eq. (4-29) to using the relations
4-8 Referring to Fig. write the equations of motion of the system if the
vertical displacement of the bar is measured from: (a) the mass center cg;
(b) the point the point A ; (d) the point B.
4-9 Show that the frequency equation for the case of non-symmetrical matrices
in Example 6 is identical to
4-10 A company crates its products for shipping as shown in Fig. The
skid is securely mounted on a truck. Experience indicates that this method
of crating is satisfactory. To cut the shipping coat, it is proposed to put two
items in a crate as shown in Fig. Would you approve this proposal?
Crate
Crate
Skid Skid
FIG. P4-2.
4-11 Consider an undamped three-degree-of-freedom system
where is a vector of transient excitations. Find the frequency

184 Systems More One Degree of Freedom 4
equation and the natural frequencies. Determine the modal vectors and
the modal matrix. (c) Verify that the modal vectors are orthogonal relative
to the matrices M and K as shown in Eq. (4-40). (d) Write the uncoupled
equations as indicated in Eq.
4-12 Repeat Prob. 4-11 for the equations
4-13 Determine the motions and of the semidefinite system in Fig.

4-13 for the initial conditions: (a) = and = {0
= and =
4-14 For the semidefinite system shown in Fig. 4-14, if = 1.2
= 25 and = find the natural frequencies and
the relative amplitudes at the principal modes.
4-15 For the system in Prob. 4-14, find the motions if the initial
are: (a) = (0.1 and = (0); = and =
0
4-16 Neglecting the inertial effect of the pinion and the gear in Fig. 4-15, let
= 0.2 = k,, = 60 = and the gear
ratio = 3: 1. Find the natural frequencies of the system: (a) referring to shaft
1; (b) referring to shaft 2.
4-17 Assume a variable speed engine with four impulses per revolution is
attached to of the gear system described in Prob. 4-16. Find the
resonance speed the gear system. What would be the resonance speed if
the engine is attached to
4-18 Assume the inertial effect of the pinion and gear in Prob. 4-16 is not
negligible. Repeat Prob. 4-16 if = 0.02 and =
4-19 Find the motions and of the semidefinite system shown in Fig.
where is a unit impulse. Assume zero initial conditions.
Stopper
P4-3.
4-20 A semidefinite system a stopper as shown in Fig. Find the

maximum force transmitted to the base of the stopper. Assume the velocity
is constant and the springs are initially unstressed. Assume = and
= 2k.
4-21 A branched-geared system is shown in Fig. P4-4. Assume the inertial effect
of the shafts and the coupling is negligible. The gear ratio of the gears
: = 1: 2 and : = 1: The data as shown are in the units. (a)
Problems 185
Gears
P4-4. Branched-geared system.
Specify the diameters of the shafts 1 and 2 such that the system has only two
numerically distinct nonzero natural frequencies. (b) Find the natural fre-
quencies.
4-22 Assuming harmonic excitations, find the steady-state response of each of the
systems in Fig. P4-5.
4-23 An air compressor of 270 kg mass is mounted as shown in Fig. 4-16. The
normal operating speed is 1,750 rpm. (a) If the resonant frequencies should
be at least percent from the operating speed, specify k , , and
What is the amplitude of at the operating speed?
4-24 A torque T sin is applied to of the torsional system in Fig. If
= 0.5 = 560 226 and =
specify and k,, of the absorber such that the resonant frequencies are 20
percent from the excitation frequency.
186 More One Degree of CHAP. 4
4-25 Repeat Prob. 4-24 if the absorber is as shown in Fig.

4-26 A horizontal force F sin is applied to the mass of the system shown in
Fig. Find the condition for which is stationary.
4-27 A dynamic absorber with damping is shown in Fig. 4-19. For optimum
design, the amplitudes of are equal at the of curves 1 and
2. Show that the relation = + is satisfied for this op-
timum.
4-28 Find the influence coefficients for the system shown in Fig. 4-5. Write the
dynamic equations. Show that the frequency equation can be reduced to
that obtained from Newton's second law.
4-29. Repeat Prob. 4-28 for the system shown in Fig. 4-7. Assume = = and
4-30 the influence wefficients and the frequency equations for each of the
systems shown in Fig. to (c).
4-31 Find the influence coefficients for each of the systems shown in Fig. P4-7.
Assume that the beams are of negligible mass.
(a) Uniform cantilever (b) Simply supported beam

FIG. P4-7.
4-32 A shaft carrying two rotating disks is shown in Find the
influence coefficients and the critical speeds of the assembly. Assume that
(1) the deflections of the bearings and the gyroscopic effect of the disks are
negligible, and (2) L, = 150 mm and L = 600 mm.
4-33 A with bending and carrying three rotating disks, is shown

in Fig. Assume that the mass of the shaft and the gyroscopic effect
of the disks are negligible. Find the critical speeds of the assembly: (a) if
and (b) if and
4-34 A shaft of negligible mass and carrying two disks is shown in Fig.
P4-9. Determine the influence and the critical speeds.
Computer problems:
Remarks: Transient response and frequency response are examined in the
problems. The modal analysis for the transient response of positive-definite
undamped systems with distinct frequencies can be examined by parts or by a
combined program. Problems 4-35 to 4-39 show the parts of the modal analysis,
following the theory developed in the chapter. Using the program TRESPUND,
Probs. 4-40 to 4-42 show the combined program, which is a collection of
subroutines. The remaining problems deal with the response of discrete
systems.
4-35 Characteristic equation. Use the program COEFF listed in Fig. 4-11 to find
the coefficients of the characteristic equation of each of the following
systems:
2 0 0
(a) 1 + -1 2 -1 =
4-36 Natural frequencies. The roots of the characteristic equation in Eq.

(4-38) yield the natural frequencies of the system. Write a program to
find the roots of the characteristic equation in Prob. 4-35. Hint: Use the
subroutine $ROOT in Fig. C-7.
4-37 Modal The steps to obtain a modal matrix are: (1)substituting an
eigenvalue A(=w2) in Eq. (2) finding the
solutions of the homogeneous equations, as shown in Example 9, to obtain a
modal vector; (3) combining the modal vectors to form a modal matrix. For
the data in Prob. 4-35, write a program to
(a) convert the equations of motion to the form

+ = {0), where H = is a dynamic matrix,
(b) use the subroutine $ROOT in Fig. C-7 to find the eigenvalues,
(c) use the subroutine in Fig. C-9 to solve the homogeneous
equations in order to find the modal vectors, and
combine the modal vectors to obtain the modal matrix.
4-38 Modal matrix. Repeat Prob. 4-37 by using the subroutine listed in
Fig. C-10. The program gives the roots of the characteristic equation and
the modal matrix.
4-39 Principal modes. The equations of motion + = can be un-
coupled and expressed in terms of the as shown in Eq.
(4-41). Assuming appropriate initial conditions for the systems in Prob.
4-35, (a) write a program to uncouple the equations, (b) compute the
188 Systems More One Degree Freedom CHAP. 4
vibrations in the original coordinates and the principal coordinates, and (c)
plot the results for where the period of the first mode.
4-40 Modal analysis. Use the program TRESPUND in Fig. to find the
transient response of the in Fig. 4-10 for Choose the
appropriate initial conditions and consider the problem three parts as
follows:
Verify from the computer print-out that the value of and from
part c is 'the sum of that of parts a and b.
4-41 Transient response plot.
(a) Modify the program TRESPUND in Fig. such that the values of
the displacement are stored in one file and the velocity in
another.
(b) Execute the program for the system in Fig. 4-10 and use the program
in Fig. to plot the results.
4-42 absorber. The undamped dynamic absorber shown in Fig. 4-16
was analyzed in Example 12. (a) Write a program to implement (4-64)
for the harmonic response and store the results in a data file. Use the
program in Fig. to plot the results as illustrated in Fig.
4-17.
Dynamic absorber with damping was described Example 13. Assume
= as for undamped absorbers. Let = 0.3. Select values
for and five values for the damping coefficient c. Write a
program to store the data in a file for amplitude versus frequency,
with c as a and (b) amplitude versus frequency with c as a
parameter. Use the program in Fig. to plot the results.
4-44 Repeat Prob. 4-43, but for the optimum condition
as shown in Prob. 4-27.
445 Vibration isolation. Vibration isolation for the system shown in Fig. 4-20
was discussed in Ekample 14. Let m, = 180 kg, = 162 =
= k,, = c,, and the mean excitation frequency f = Hz. Assume
c, Calculate the amplitudes and the force trans-
mitted to the foundation with as a parameter for Hz. (a)
Write a program to store the calculated values of versus f in a data
versus f in a second file, and versus in a third file. Use the
program in Fig. to plot the results.
Frequency response. Consider an n-degree-of-freedom system with viscous
damping
Problem 189
where M is the mass matrix, the damping matrix, K the stiffness matrix,
and the excitation vector. Assume each element of is harmonic
and of the same frequency as discussed in 4-8. (a) Write a program
for n to store the information of the amplitude versus and phase
angles versus of the response in separate data files. (b) Use the program
in Fig. to plot the results. For purpose of illustration,
assume 0.1 and the equations in the SI units are

Mechanical Vibrations: Theory and Applications

Uploaded by

Copyright:

Available Formats

Mechanical Vibrations: Theory and Applications

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Mechanical Vibrations: Theory and Applications

Uploaded by

Copyright:

Available Formats

FRANCIS S.

Allyn and Bacon,

rights reserved. Printed in the United States of America.

Library of Congress Cataloging in Publication Data

Tse, Francis Sing

CHAPTER 2 SYSTEMS WITH ONE DEGREE

2-6 Frequency Response Method 45

CHAPTER 3 SYSTEMS WITH ONE DEGREE

CHAPTER 4 SYSTEMS WITH MORE THAN ONE DEGREE

CHAPTER 5 METHODS FOR

6-14 Summary 245

CHAPTER 7 CONTINUOUS SYSTEMS

CHAPTER 8 NONLINEAR SYSTEMS

8-8 Free Vibration 323

CHAPTER 9 SOLUTIONS BY DIGITAL COMPUTERS

A Elements of Matrix Algebra

Vibration is study of oscillatory motions. The ultimate goals of

time and frequency domain analysis is introduced in Chapter 2 through the

FIG. 1-8. Displacement, velocity, and acceleration vectors.

Thus, each differentiation is equivalent to the multiplication of the vector

1-2 ELEMENTS OF A VIBRATORY SYSTEM

The elements that constitute a vibratory system are illustrated in Fig.

1-1. Elements of a vibratory system.

FIG. 1-2. A periodic motion.

viscous damping c is measured in force per unit velocity.

T A BLE Comparison of Rectilinear and

Spring force= kx Spring torque =

Inertia force= Inertia =

To illustrate different types of vibratory motion, let us choose various

( b ) Damped free vibration Forced vibration

FIG. 1-3. Simple vibratory systems.

equilibrium position. Hence natural frequency describes the rate of

with each subsequent cycle of oscillation, and the frequency of vibration

F I G . 1-4. Free vibration of systems shown in Figs. and ( b ) .

conditions or the natural frequency of the system. It will be shown in

1-4 SIMPLE HARMONIC MOTION

Simple harmonic motion is the simplest form of periodic motion. It will

where t = time, = constant, and X = constant, the motion of P about the

* A sine, a cosine, or their combination can be used to represent a simple harmonic

FIG. 1-5. Simple harmonic motion: X cos

is called the circular frequency measured in

where a constant. When the acceleration of a particle with recti-

= cos + wt cos a -sin sin a )

=2X cos- t cos

FIG. 1-6. Graphical representation of beats.

consecutive maximum amplitudes, is

(a) Vectorial representation

(b) Harmonic motions

F I G . 1-7. Harmonic motions represented by rotating vector.

where X is the length of the vector or its magnitude and = is called

It should be remembered that a harmonic motion is a reciprocating

The equation = cos 0 j sin 0 is called Euler's formula.

(a) Vector addition ( b ) Vector addition with phasor notation (wt 0)

F I G . 1-9. Addition of harmonic functions: vectorial method.

The quantity = is a complex number and is called the complex

In representing harmonic motions by rotating vectors, it is often

Example 1. Manipulation of Complex Numbers

(a) = 1 + = j sin 60") = =

The last two examples indicate that the multiplication of a vector by j