Ebook Modeling and Analysis of Passive Vibration Isolation Systems PDF Full Chapter PDF

Modeling and Analysis of Passive
Vibration Isolation Systems - eBook

PDF
Visit to download the full and correct content document:
https://ebooksecure.com/download/modeling-and-analysis-of-passive-vibration-isolati
on-systems-ebook-pdf/
MODELING AND
ANALYSIS OF
PASSIVE VIBRATION
ISOLATION SYSTEMS
MODELING AND
ANALYSIS OF
PASSIVE VIBRATION
ISOLATION SYSTEMS
SUDHIR KAUL
Associate Professor, School of Engineering and Technology,
Western Carolina University, Cullowhee, NC, USA
Elsevier
Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States
Copyright © 2021 Elsevier Inc. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or any information storage
and retrieval system, without permission in writing from the publisher. Details on how to
seek permission, further information about the Publisher’s permissions policies and our
arrangements with organizations such as the Copyright Clearance Center and the
Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright
by the Publisher (other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and
experience broaden our understanding, changes in research methods, professional practices,
or medical treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in
evaluating and using any information, methods, compounds, or experiments described
herein. In using such information or methods they should be mindful of their own safety
and the safety of others, including parties for whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or
editors, assume any liability for any injury and/or damage to persons or property as a matter
of products liability, negligence or otherwise, or from any use or operation of any methods,
products, instructions, or ideas contained in the material herein.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress
ISBN: 978-0-12-819420-1
For Information on all Elsevier publications visit our website at

https://www.elsevier.com/books-and-journals
Publisher: Matthew Deans

Acquisitions Editor: Dennis McGonagle
Editorial Project Manager: Hilary Carr
Production Project Manager: Prasanna Kalyanaraman
Cover Designer: designer Christian Bilbow
Typeset by Aptara, New Delhi, India

Contents
Preface vii
Dedication ix
Acknowledgment xi
1. Vibration isolation—background 1
1.1 Introduction 1
1.2 Isolator materials 1
1.3 Common elastomeric isolator designs 4
1.4 Stiffness and damping 4
1.5 Single-degree-of-freedom system 9
1.6 Multiple-degree-of-freedom system 16
Review exercises 24
References 25
2. Viscoelastic modeling—passive vibration isolators 27

2.1 Viscoelasticity 27
2.2 Voigt or Kelvin–Voigt model 28
2.3 Zener model 30
2.4 Maxwell–Voigt model 33
2.5 Generalized Maxwell or Maxwell Ladder model 37
2.6 Voigt fractional model 40
2.7 MV fractional model 42
2.8 Hysteresis model 44
Review exercises 58
References 60
3. Vibration isolation system modeling 61

3.1 Planar isolation systems (three degrees-of-freedom) 61
3.2 Spatial isolation systems (six DOFs) 67
3.3 Vibration isolation system with displacement limiting design 70
3.4 Vibration isolation system with hysteresis 73
Review exercises 88
References 89
v
vi Contents
4. Vibration isolation systems—nonlinear models 91

4.1 Single degree-of-freedom isolator with stiffness nonlinearity 91
4.2 Single-DOF isolator with stiffness and damping nonlinearity 99
4.3 Planar isolation system with stiffness and damping nonlinearity 104
4.4 Other nonlinear models 107
Review exercises 116
References 117
5. Modeling elastomer characteristics 119

5.1 Mullins effect 119
5.2 Payne effect 124
5.3 Aging 128
5.4 Creep 131
5.5 Hyperelastic model 135
References 141
6. Modeling inertia effect 143

6.1 Inertia effect 143
6.2 Inertia effect—single-degree-of-freedom model 145
6.3 Inertia effect—two-degree-of-freedom model 153
6.4 Inertia effect—three-degree-of-freedom model 162
References 169
7. Elastomeric vibration isolator design 171

7.1 Example—single-degree-of-freedom isolator design 171
7.2 Example—planar isolation system design 178
7.3 Example—spatial isolation system design 184
7.4 Case studies 190
References 207
Appendix A 209
Appendix B 215
Index 219
Preface
Passive vibration isolators are widely used in areas such as automotive,

aerospace, manufacturing, heavy machinery, and civil structures. On the one
hand, the design and development of passive vibration isolators is a mature
technology. On the other hand, analytical modeling of such isolation systems
is still evolving due to the multifaceted intersection of different disciplines.
The aim of this book is to serve as a reference for engineers and researchers
involved in the design,development,modeling,analysis,and testing of passive
vibration isolation systems.
This book has been divided into seven chapters.Chapter 1 presents a brief
review of vibration analysis and terminology. The aim of this chapter is to
serve as a refresher; this chapter can be used in conjunction with Appendix
A and Appendix B to recapitulate the content from an introductory course
in vibration analysis. Chapter 2 presents several linear viscoelastic rheological
models that can be used for a single-degree-of-freedom analysis of vibration
isolation systems. The main attributes of each model are discussed in this
chapter along with the governing relationships between critical model and
design parameters. Chapter 3 presents linear viscoelastic models for planar
(two- and three-degree-of-freedom) and spatial (six-degree-of-freedom)
vibration isolation systems. Additional models for piecewise behavior and
hysteretic systems are also presented in this chapter. Chapter 4 presents
nonlinear models for single-degree-of-freedom systems as well as multiple-
degree-of-freedom systems that can be used for the analysis of passive
vibration isolation. Although nonlinearities can be attributed to multiple
sources, this chapter primarily focuses on analytical models for a few specific
nonlinearities associated with stiffness and damping characteristics. Chapter
5 presents models that are typically used for the analysis of elastomeric
vibration isolators. Models that can be used to represent such phenomena
as Mullins effect, Payne effect, hyperelasticity, aging, and creep have been
discussed in this chapter. Chapter 6 presents models that can be used to
account for the inertia effect that is typically observed in vibration isolation
systems that need to withstand very high excitation frequencies. These
models allow an evaluation of vibroacoustic characteristics well above 1
kHz while capturing internal resonance and wave effects. Chapter 7 presents
examples and case studies that integrate concepts from the models presented
in the previous chapters of the book while demonstrating the influence of
vii
viii Preface
the vibration isolation system on overall system dynamics. There are two
brief appendices that may be used as a refresher on ordinary differential
equations and matrix algebra. Each chapter has a few exercise problems that
can be solved to test the understanding of the content presented in the
chapter.
The models discussed in this book encompass a wide range that can
be useful for the analysis of passive vibration isolation systems. While some
of the models presented in this book have been used for quite some time,
others are relatively new and offer useful options for gaining an analytical
insight that can be used for design. Furthermore, some of the models are
phenomenological, while others are semi-empirical; therefore allowing a
design or analysis engineer to customize the models during the product
development process. Some of the models for elastomeric materials and
nonlinear behavior that have been discussed in this book are active areas
of research and continue to be discussed and investigated in the existing
literature. A surge in the use of electric powertrains has resulted in new
requirements for passive vibration isolation systems, a few models presented
in this book are possible options for the analysis of internal resonance in
such systems. I hope that the variety of models discussed in this book is
useful in the design and development of passive vibration isolation systems
by holistically accounting for vibration response, system dynamics, design
parameters, and isolator design.
Dedication
Dedicated to my mother, whose life was a beacon of perseverance, resilience,

humility, and kindness.
Dedicated to my father, whose life continues to inspire me with an insatiable
love for learning.
ix
Acknowledgments
I would like to acknowledge the help received from the entire publishing
team at Elsevier. I also want to acknowledge the support received from
Western Carolina University and all my colleagues, students, peers, and
collaborators over the years. While many individuals have helped in shaping
my understanding of the content of this book, two individuals merit
special mention. I want to express my gratitude to my graduate advisor
at the University of Wisconsin-Milwaukee, Dr. Anoop Dhingra, and my
supervisor at Harley Davidson Motor Company, Dr. Tim Hunter, for
introducing me to different aspects of passive vibration isolation. Finally, I
want to thank my wife for her love, patience, and understanding during this
endeavor.
I would be remiss in failing to acknowledge the contribution of many
individuals who have shaped my academic journey, this includes many
teachers from my school and professors from the universities I attended.
I also want to thank my sister for being my very first teacher and mentor.
xi
CHAPTER 1
Vibration isolation—
background
Contents
1.1 Introduction 1
1.2 Isolator materials 2
1.3 Common elastomeric isolator designs 4
1.4 Stiffness and damping 6
1.5 Single-degree-of-freedom system 9
1.6 Multiple-degree-of-freedom system 17
Review exercises 24
References 25
1.1 Introduction
The use of vibration isolators and vibration isolation systems is widely
prevalent in multiple applications such as automotive, railroad, aerospace,
heavy machinery, civil structures, etc. Some of the main reasons for using
a vibration isolator include mitigation of resonance peaks, reduction of
transmissibility, enhancement of fatigue life, improvement in ergonomics,
etc. in the presence of external or internal sources of dynamic excitation.
The design of a vibration isolator requires a close examination of multiple
considerations such as the source of dynamic excitation, range of excitation
frequency, excitation amplitude, allowable displacement, acceleration limits
of the isolated system, available design envelope, etc. Additionally, consider-
ations of environmental conditions, manufacturability, and material choice
are also important. All these considerations accentuate the importance of
a theoretical model that can reasonably predict the performance of the
isolation system before finalizing the design and before manufacturing
prototypes that can be used for testing. Therefore, it is critical to select a
suitable model that can be correlated to test results and eventually used to
finalize design details.
Modeling and Analysis of Passive Vibration Isolation Systems. Copyright © 2021 Elsevier Inc.
DOI: 10.1016/B978-0-12-819420-1.00007-8 All rights reserved. 1
2 Modeling and analysis of passive vibration isolation systems
1.2 Isolator materials

Vibration isolation can be achieved by using materials capable of providing
a combination of highly elastic behavior in conjunction with damping
properties. Pneumatic, hydraulic, elastic metal, and elastomeric designs are
commonly used in commercial vibration isolation applications. Elastomeric
materials are arguably most common and are extensively used in the industry
with a very commonly used design consisting of elastomeric material
bonded to metal plates or a metal core. Such isolators are typically called
elastomeric mounts. Natural rubber, neoprene, and butyl rubber are some of
the commonly used elastomers in commercial vibration isolators. Elastomers
provide a designer with a range of stiffness and damping characteristics
as well as an ability to withstand different environmental conditions. This
ability to satisfy performance requirements over a wide range of rugged
conditions along with the ease of manufacturing through a molding process
make elastomers a common choice for isolators during the design process.
Table 1.1 lists some of the commonly used elastomers for manufacturing
passive vibration isolators with a listing of some of their characteristics
that can be considered during design. In addition to the commonly used
elastomers, manufacturers often develop proprietary elastomeric recipes
to serve the needs of a specific design that may require a combination
of properties from different materials. Properties of elastomeric materials
can be changed significantly by changing their composition or by using
different blends.A typical manufacturing process of the raw material involves
vulcanization by adding sulfur and by the addition of accelerators, fillers, and
plasticizers (Mark, Erman, & Roland, 2013). The raw material is then used
in a molding process to produce a vibration isolator of the designed shape
and size to deliver the necessary stiffness and damping properties. While
there are many characteristics that are sought from the design of a vibration
isolator, some of the common technical properties that a designer seeks to
comprehend are damping, dynamic stiffness, environmental resistance, and
some of the inherent nonlinearities.
Metal springs have been commonly used for vibration isolation ap-
plications as they can be designed to offer a range of stiffness properties
in heavy machinery applications. Most of these designs do not allow
much flexibility with damping as most metal springs offer relatively low
material damping. Coil springs, disc springs, slotted springs, etc. are some
examples of metal springs commonly used in vibration isolation applications
(Rivin, 2003).
Vibration isolation—background 3
Table 1.1 Commonly used elastomeric materials.

Material Key characteristics
Natural rubber or Good processability, high elongation, high
polyisoprene (NR) tensile strength, relatively low damping
ratio, good bonding to metals,
moderate-to-low oil resistance,
moderate-to-low chemical resistance,
operating temperature: −30 to 160 °F.
Neoprene or Good abrasion resistance, good tear strength,
polychloroprene mechanical properties similar to NR,
moderate oil resistance, moderate chemical
resistance, flame retarding, operating
temperature: −20 to 180 °F.
Nitrile or acrylonitrile Good abrasion resistance, high oil resistance,
butadiene resistance to swelling, conductor of
electricity, operating temperature: −30 to
250 °F
Styrene butadiene High abrasion resistance, moderate-to-high
oil resistance, moderate-to-high chemical
resistance, good electrical insulator,
operating temperature: −20 to 210 °F
Ethylene-propylene-diene High abrasion resistance, relatively higher
terpolymer damping ratio, moderate tear resistance,
low oil resistance, low chemical resistance,
operating temperature −40 to 300 °F
In some cases, it is common to use a separate damper to augment

damping of the vibration isolation system. Viscous dampers are designed
to offer resistance to relative motion between two surfaces that are typically
separated through a fluid film. Some of these dampers can exhibit nonlinear
behavior due to strong temperature dependence. Since the early 1990s,
magnetorheological (MR) dampers have been developed by researchers and
manufacturers to provide smart damping properties that can be controlled
through input current to an electromagnet that in turn governs the behavior
of the damper. MR fluids consist of micron-sized particles in a carrier
fluid, an MR damper allows control over the apparent viscosity of the
fluid by controlling the magnetic flux of the electromagnet. Such a damper
is considered to be a semi-active system that can be used for vibration
isolation and control (Choi & Wereley, 2008; Dominguez, Sedaghati, &
Stiharu, 2004). Friction dampers and electromagnetic dampers are other
examples of dampers that have been used in some vibration isolation

applications.
A hydraulic mount, also called a hydromount, is another vibration isola-
tor that has been used in automotive applications. Such an isolator provides
properties that are amplitude dependent as well as frequency dependent. The
isolator typically consists of two chambers connected through a channel that
allows fluid passage from one chamber to the other. This design allows the
vibration isolator to exhibit low stiffness and high damping for dynamic
excitations with large amplitude and low frequency while demonstrating
low damping at small amplitude and high frequency vibrations (Truong
& Ahn, 2010). Different designs of hydromounts have been used in some
automotive applications to provide dynamic characteristics that can be tuned
to provide a frequency-dependent behavior.
1.3 Common elastomeric isolator designs

Some of the common designs of passive vibration isolators involve elas-
tomeric material bonded to metal plates or a metal core with a static member
that is assembled to a rigid frame and a dynamic member that separates
the isolated components from the source of dynamic excitation. There are
some designs that consist of elastomeric materials without being bonded to a
metal plate or a metal core, such designs typically do not need to withstand
high static loads. Passive elastomeric isolators are generally designed to be
under compression loading or shear loading with circular or rectangular
cross sections being the most commonly used. Grommets, bushings, etc.
are also common examples of passive elastomeric isolators. Some of the
commonly used design configurations of elastomeric isolators are shown in
Table 1.2.
Most elastomeric isolators exhibit varying amounts of creep. Creep is ex-
hibited in the form of an increasing deformation even when the load is static.
Creep behavior is exponential and time dependent. Upon removal of the
static load, creep is mostly reversible but leaves some residual strain or per-
manent set in the elastomer. Dynamic loads can also result in an increasing
deformation in elastomeric isolators with every loading cycle,this property is
often referred to as the dynamic set. In elastomeric isolator design, it is com-
mon to allow for 10% deflection to account for creep and dynamic set. Some
specific characteristics of elastomeric isolators will be discussed in detail in
Chapter 5.
Table 1.2 Examples of passive elastomeric isolator designs.

1.4 Stiffness and damping

Stiffness and damping are two main characteristics sought from a vibration
isolator and are used for characterizing a dynamic model. Stiffness identifies
the ability of a material to withstand applied force. The stiffness constant
or the spring constant (k) of an isolator is defined as the force per unit
deflection, expressed as follows:
F
k= (1.1)
x
In Eq. (1.1), F is the force and x is the corresponding deflection. A
relatively higher stiffness indicates a higher value of k, in turn implying that a
higher force is necessary to obtain the same amount of deflection. For linear
elastic behavior, the stiffness constant is directly related to the modulus of
elasticity of the material. The modulus of elasticity of elastomeric materials
is significantly lower, and the material regains its original shape and size very
quickly after undergoing significant deformation. Eq. (1.1) identifies a linear
relationship, some of the passive isolator designs may not be represented by
this stiffness model. Such cases will be discussed in the chapter on nonlinear
vibration isolators. Often, an equivalent stiffness is computed for developing
a dynamic model when multiple isolators (or spring elements) are connected
together. An example of calculating the equivalent stiffness is presented in
Example 1.1.
Example 1.1: Find the equivalent stiffness for the system of springs in (A)
parallel and (B) series shown in Fig. 1.1.
For the two springs in parallel F = (k1 + k2 )x, this can be identified
from the free-body diagram as an external force causes the same deflection
in both the springs. This relationship can be used to write the ratio of force
A B
Figure 1.1 Springs in parallel and springs in series.
and deflection as Fx = k1 + k2 . Therefore, for the two springs in parallel,

the equivalent spring constant can be expressed as the sum of the spring
constants: keq = k1 + k2 , thereby expressing the force-deflection ratio as
F
x
= keq . It can be seen that the equivalent spring constant (keq ) of the
springs in parallel is higher than the spring constant of each spring.In general,
the equivalent spring constant for any number of springs in parallel can be
calculated as keq = ki .For the two springs in series,F = k1 x1 = k2 x2 = keq x
i
as the force in both the springs remains the same. The overall deflection, x,
is the sum of the individual deflection x1 and x2 of each spring, x1 + x2 = x.
Substitution from the force equation results in the following: kF1 + kF2 = kFeq .
These two equations can be used to calculate the equivalent spring constant
of the two springs in series as keq = 1 +1 1 . The overall equivalent spring
k1 k2
constant keq of the springs in series is lesser than the spring constant of each
spring. In general, the equivalent spring constant for any number of springs
in series can be calculated as keq = 1 1 . It may be noted that spring
ki
elements may be connected through a combination of series and parallel
configurations. However, the equivalent spring constant of the system can
be calculated by using a term-by-term reduction of the series and parallel
configurations.
As vibration isolation systems can be subject to a wide range of excitation
frequencies, it is common to represent the force-deflection relationship in
terms of complex stiffness. For an input displacement of x = Xejωt and an
output force of f = Fej(ωt + ϕ ) , complex stiffness is the ratio between the force
and the corresponding displacement and is expressed as

f F F F k
k∗ = = e jϕ = cos(ϕ) + j sin(ϕ) = k + jk = k 1 + j
x X X X k
(1.2)
In Eq. (1.2), k∗ is complex stiffness that is often expressed in terms of the
magnitude and phase from Eq. (1.2) as

|k∗ | = k 2 + k 2 (1.3)

η= k
k
= tan(ϕ) (1.4)
In Eq. (1.3), |k∗ | is the magnitude of complex stiffness, often called

dynamic stiffness, and η is called the loss factor that is directly related to
the phase angle, ϕ, as seen in Eq. (1.4). This form of stiffness is commonly
used in case of hysteretic damping, with a lag between the applied force and
the resulting deformation.
Damping is a complex phenomenon that can be used to indirectly
represent the energy dissipated by the vibration isolation system during
repeated or cyclic loading. Viscous damping is commonly used to model
the damping of many vibration isolators. In some cases, hysteretic damping is
more suitable to represent the damping of a vibration isolator, which will be
discussed further in subsequent chapters.Three variables used to characterize
viscous damping are the damping constant, the critical damping constant,
and the damping ratio (also called the damping factor). The damping
constant (c) is defined as applied force per unit relative velocity
F
c= (1.5)
v
In Eq. (1.5), v is the relative velocity corresponding to the applied force, F.

Critical damping is defined to identify the limiting case between oscillatory
and nonoscillatory motion, and the critical damping constant is defined
as
√
cc = 2 km (1.6)
In Eq. (1.6), k is the stiffness constant and m is the mass of the system.
The damping ratio is defined as the unitless ratio between the damping
constant and the critical damping constant.
c
ξ= (1.7)
cc
In Eq. (1.7), ξ is the damping ratio. A system is said to be undamped

when the damping ratio is zero, this is an idealization that is sometimes
used in preliminary modeling of an isolation system. A system is said to be
underdamped when the damping ratio is between zero and one, this is the
most common case in vibration isolation modeling. A system is said to be
overdamped when the damping ratio is above one. If the damping ratio is
equal to one, the system is said to be critically damped. Vibration isolators are
most commonly modeled as spring-damper units with stiffness and damping
characteristics that are used to isolate a system from an excitation source or
to isolate one subsystem from another.
For hysteretic damping, an equivalent damping ratio, ξ eq , and an equiva-

lent damping constant, ceq , are defined as
η
ξeq = (1.8)
2
k √
ceq = = η mk (1.9)
ω
In Eqs. (1.8) and (1.9), η is the loss factor defined in Eq. (1.4). It may
be noted that the equivalent damping ratio and the equivalent damping
constant assume a harmonic response, with frequency ω, to a harmonic

excitation. The variables k and k are the variables associated with complex
stiffness, as seen in Eqs. (1.2) and (1.3). It may be noted that some systems may
exhibit friction damping,details about such damping can be found in various
studies (Balachandran & Magrab, 2019; Inman, 2014; Thomson & Dahleh,
1998; Inman, 2014; Meirovitch, 1997; Rao, 2007 2017; Tongue, 2002).
1.5 Single-degree-of-freedom system

Degree of freedom (DOF) is defined as the minimum number of inde-
pendent parameters needed to locate a rigid body in space. A single DOF
(SDOF) system is the simplest form of a vibrating system often used as a
preliminary model for vibration isolation. Such a model consists of a rigid
body connected to one or multiple springs or a spring-damper unit. The
system is said to be undamped if it is modeled without damping. This section
presents the solution of an SDOF system that will be used in subsequent
chapters. The equation of motion (EOM) for an undamped SDOF system
without any external excitation, shown in Fig. 1.2, can be expressed as
..
m x +kx = 0 (1.10)
In Eq. (1.10), m is the mass of the system and k is the stiffness constant,
while x is the deflection due to any external perturbation that the sys-
tem may be subject to (from its state of static equilibrium). Eq. (1.10) is
a homogeneous second-order differential equation with the following
solution:
x(t ) = c1 cos(ωnt ) + c2 sin(ωnt ) (1.11)
In Eq. (1.11), c1 and c2 are two constants that
can be determined from
the initial conditions of the system, and ωn = mk is the natural frequency
Figure 1.2 Undamped SDOF system-free vibration.
of the system. Eq. (1.11) provides the time (t) history of displacement for
the undamped system when it is perturbed from its static equilibrium. The
steps in the development of the solution in Eq. (1.11) are not presented here.
Appendix A provides a brief introduction to the solution of the second-
order differential equations.
The EOM for an undamped SDOF system in the presence of an
excitation input can be expressed as
..
m x +kx = f (1.12)
In Eq. (1.12), f is the excitation force. The solution to the EOM in
Eq. (1.12) is a combination of the homogeneous and the particular solution
for a general excitation force, f = f0 ejωt , with an excitation frequency of ω,
and can be written as
f0 /k
x(t ) = c1 cos(ωnt ) + c2 sin(ωnt ) + 2
(1.13)
1 − ωωn
The solution in Eq. (1.13) can be expressed in different forms depending

on the ratio between the excitation frequency and the natural frequency of
the system.
The EOM for a damped SDOF system with viscous damping and
without any external excitation, shown in Fig. 1.3, can be expressed as
.. .
m x +c x +kx = 0 (1.14)
The solution for the EOM of the damped system in Eq. (1.14) is
x(t ) = c1 eλ1t + c2 eλ2t (1.15)
Figure 1.3 Damped SDOF system—free vibration.
In Eq. (1.15), c1 and c2 are two constants that can be determined from
Also, the solution varies with the nature of λ1 and
the initial conditions.
2
λ2 , λ1,2 = − 2m c
± c
2m
− mk , to determine whether the system is
underdamped, overdamped, or critically damped. The steps needed for
solving the governing differential equation in Eq. (1.14) are provided in
Appendix A.
For vibration isolation, underdamping is the specific case of interest. For
underdamping, 0 < ξ < 1 and the solution in Eq. (1.15) can be modified as
follows:

x(t ) = e−ξ ωnt c1 cos(ωd t ) + c2 sin(ωd t ) (1.16)
In Eq. (1.16), c1 and c2 are two

constants that can be determined from
the initial conditions, and ωd = ωn 1 − ξ 2 is the damped frequency, while
ωn is the undamped natural frequency. It may be noted that for ξ = 0, the
solution in Eq. (1.16) becomes identical to the solution of the undamped
system in Eq. (1.11).
The EOM for a damped SDOF system in the presence of an excitation
input can be expressed as
..
m x +cx + kx = f (1.17)
In Eq. (1.17), f is the excitation force. The solution to the EOM in

Eq. (1.17) is a combination of the homogeneous and the particular solution
for a general excitation force, f = f0 ejωt , with an excitation frequency of ω,
and can be expressed as

f0
x(t ) = e−ξ ωnt c1 cos(ωd t ) + c2 sin(ωd t ) + e j(ωt−ϕ)
2 1/2
(k − mω2 )2 + (ωc)
(1.18)
In Eq. (1.18), ϕ is the phase angle between excitation force and the
resulting displacement and can be found as

−1 ωc
ϕ = tan (1.19)
k − mω2
Examples of the response of undamped and damped SDOF systems are
presented in Example 1.2.
Although the closed-form solution of Eq. (1.17) is readily available, it is
common to express the second-order equation as a system of the first-order
equations as follows:
.
x1 = x2 (1.20)
. c k f
x2 = − x2 − x1 + (1.21)
m m m
.
In Eqs. (1.20) and (1.21), x1 = x and x2 = x The system in Eqs. (1.20) and
(1.21) is equivalent to Eq. (1.17). The main advantage of expressing the
second-order differential equation as a system of the first-order differential
equations is the use of numerical methods. This is particularly beneficial for
a nonlinear system as a closed-form solution may not be readily available for
such a system. The system in Eqs. (1.20) and (1.21) can also be expressed in
a matrix form as
.
x1 0 1 x1 0
. = + 1 f (1.22)
x2 − mk − mc x2 m
The formulation in Eq. (1.22) is commonly called the state-space form

of the system.
The solution in this section is limited to harmonic excitation. The
excitation force may not always be harmonic. In such cases, alternative
methods are used to solve the governing EOM. Methods such as Fourier
series, convolution, Laplace transform, etc. can be found in introductory text
books such as (Rao, 2017) (Thomson & Dahleh, 1998). A brief review of
some of the properties of the Laplace transform and the Fourier transform
is provided in Appendix A.
Figure 1.4 Undamped SDOF—free vibration response.
Example 1.2a: For an undamped SDOF system, as shown in Figure 1.2,

with a mass (m) of 1 kg, a stiffness constant (k) of 10,000 N/m, and the
given initial conditions, plot the free vibration response. The given initial
.
conditions are as follows: x(0) = 0.01 m and x(0) = 0.1 m/s. Using the
solution from Eq. (1.11) and applying the initial conditions, the two constants
can be found as c1 = 0.01 and c2 = 0.001. Free vibration response can be
expressed as
x(t ) = 0.01 cos(100t ) + 0.001 sin(100t ) (E1.2.1)
The natural frequency of the system is 100 rad/s or 15.9155 Hz and the
natural time-period is the reciprocal of the natural frequency, 0.0628 s. The
first few cycles of free vibration response are shown in Fig. 1.4 by plotting
the response, x(t).
Example 1.2b: For a damped SDOF system, as shown in Fig. 1.3, with
a mass of 1 kg, a stiffness of 10,000 N/m, a viscous damping constant of
50 N-s/m, and the given initial conditions, plot the free vibration response.
.
The given initial conditions are as follows: x(0) = 0.01 m and x(0) =
0.1 m/s. The critical damping constant for the system is 200 N-s/m. This
Figure 1.5 Damped SDOF—free vibration response.
can be used to calculate the damping ratio of 0.25. As the damping ratio is
less than one, the solution for an underdamped system in Eq. (1.16) can be
used. The natural frequency of the system is 100 rad/s. Applying the initial
conditions, the free vibration response can be expressed as
x(t ) = e−25t [0.01 cos(96.82t ) + 0.0036 sin(96.82t )] (E1.2.2)
The undamped natural frequency of the system is 100 rad/s and the
undamped natural time-period is 0.0628 s. However, the damped natural
frequency is 96.82 rad/s and the corresponding time-period is 0.0649 s.
The free vibration response of the viscous damped system is shown in
Fig. 1.5.
In addition to using the time domain, it is often beneficial to use the
frequency domain for vibration analysis. Applying Fourier transform to the
EOM in Eq. (1.17) with zero initial conditions yields the following:
−mω2 X (ω) + jcωX (ω) + kX (ω) = F (ω) (1.23)
In Eq. (1.23), X(ω) and F(ω) are the Fourier transforms for x(t) and f(t),
respectively. The transformed equation in Eq. (1.23) is often expressed in
the form of the ratio between X(ω) and F(ω), called the frequency response
function (FRF), and is expressed as
X (ω) 1
= (1.24)
F (ω) (k − mω2 + jcω)
The ratio in Eq.(1.24) is commonly used to plot the magnitude and phase
of the frequency response at varying frequencies of the excitation force. It
may be noted that the frequency response is plotted in multiple formats,
typically in semi-log or log-log format. Also, it is common to plot velocity
or acceleration response instead of the displacement response shown in
Eq. (1.24).
Impulse response is another commonly used means of calculating the
response of an SDOF system. This can be computed by writing the transfer
function for the system using the EOM in Eq. (1.17). Using the Laplace
transform and zero initial conditions yields the transfer function, H(s), of
this system as
X (s) 1
H (s) = = (1.25)
F (s) (ms + cs + k)
2
In Eq. (1.25), X(s) and F(s) are the Laplace transformations for x(t) and f(t),
respectively. This form is typically called the displacement transfer function.
Eq. (1.25) can be alternately expressed as
X (s) 1/m
H (s) = = 2 (1.26)
F (s) s + 2ξ ωn s + ωn2
All variables in Eq. (1.26) are already defined earlier in this section. For
an impulse input force, F(s) = 1, and the inverse Laplace of the transfer
function in Eq. (1.26) is called the impulse response of the system and is
often useful in experimental characterization. The impulse response for the
transfer function in Eq. (1.26) can be expressed as

−1 −1 1/m
h(t ) = L [H (s)] = L (1.27)
s2 + 2ξ ωn s + ωn2
In Eq. (1.27), L−1 is the inverse Laplace transform and h(t) is the impulse
response in time domain. Example 1.3 provides an example of the frequency
response of a damped SDOF system.
Example 1.3: For a damped SDOF system, as shown in Figure 1.3, with a
mass of 1 kg, a stiffness of 10,000 N/m, and a viscous damping constant of
50 N-s/m, plot the magnitude and phase of the frequency response.
Figure 1.6 Magnitude response.
Using the form expressed in Eq. (1.24), the frequency response for the
given system can be expressed as
X (ω) 1
= (E1.3.1)
F (ω) 10,000 − ω2 + j50ω
It may be noted that the units of the frequency response in eq. (E1.3.1) are
m/N. The magnitude and phase for the frequency response can be expressed
as

X (ω) 1

F (ω) = (E1.3.1a)
(10,000 − ω2 )2 + (50ω)2

X (ω) −1 50ω
∠ = −tan (E1.3.1b)
F (ω) 10,000 − ω2
The magnitude and phase response is plotted in Figs. 1.6 and 1.7,
respectively.
Figure 1.7 Phase response.
1.6 Multiple-degree-of-freedom system

Most systems often consist of multiple DOF with different forms of coupling
between the rigid bodies whose motion is intertwined. This implies that
two or more independent parameters are needed to locate such a rigid
body in space. This section summarizes the general solution for a multi-
DOF (MDOF) system that is often expressed in a matrix form. A brief
introduction to matrix algebra is provided in Appendix B. The EOM
of an undamped MDOF system without any external excitation can be
formulated by combining the individual EOM as follows:
..
M X + KX = 0 (1.28)
In Eq. (1.28), M and K are the n × n mass and stiffness matrices, while
X is n × 1 for an MDOF system with n DOF. For a general solution of
X = X0 eλt , Eq. (1.28) can be written as
(λ2 I + M −1 K )X0 = 0 (1.29)
In Eq. (1.29), I is an identity matrix. A nontrivial solution of the system
in Eq. (1.29) can be obtained by solving for the values of λ, also called
the eigenvalues of the system. These eigenvalues can be found from the
determinant (det) of the matrix as
det(λ2 I + M −1 K ) = 0 (1.30)
The values of λ from the solution of Eq. (1.30) are the undamped natural
frequencies of the system. For each eigenvalue, λi , a corresponding vector of
size n × 1, ψ i , can be computed from Eq. (1.29), these are the corresponding
mode shapes or eigenvectors of the system. Some critical properties of mode
shapes are as follows:
• Mode shapes (ψ) can be scaled arbitrarily.
• Mode shapes have a weighted orthogonality property. This implies that
for two different mode shapes, ψ r and ψ s (where, r = s), ψrT Mψs = 0
and ψrT Kψs = 0.
• For r = s, ψrT Mψs = Mr and ψrT Kψs = Kr , where Mr and Kr are
diagonal matrices. The values of Mr and Kr are called the modal mass
and the modal stiffness, respectively.
• As mode shapes can be scaled arbitrarily, values of modal mass and modal
stiffness are not unique or well-defined numbers for a system.
• Specific scaling of mode shapes can be performed to obtain the modal
mass matrix as an identity matrix. This is often useful as the inverse
of an identity matrix is the identity matrix itself and its use will be
demonstrated in the subsequent part of this section.
The general EOM for a damped MDOF system with an external
excitation can be expressed as
.. .
M X + C X + KX = F (1.31)
In Eq. (1.31), M, C, and K are the mass, damping, and stiffness matrices,
respectively, each system matrix is n × n, where n is the number of DOF
of the MDOF system. Also, X, is an n × 1 matrix and F is the n × 1
external force or moment matrix. The nature of the system matrices can
vary significantly from one system to another. However, regardless of the
nature of the system matrices, the EOMs can be diagonalized and solved in
terms of modal coordinates, using some of the properties of mode shapes
discussed earlier in this section. For proportional damping, the damping
matrix can be expressed as a linear combination of the mass and stiffness
matrices, with C = aM + bK, where a and b are real constants. Using a
coordinate transformation of X = ψq, where ψ is the mode shape matrix,
and premultiplying Eq. (1.31) by ψ T , the following diagonalized system is
obtained:
.. .
ψ T Mψ q +ψ T Cψ q +ψ T Kψq = ψ T F (1.32)
It may be noted that ψ consists of all the mode shapes. The EOM yield
a system of equations that are uncoupled and can be solved individually
to solve for the values of q, these values can then be combined as per the
coordinate transformation to find the actual solution.The EOM in Eq.(1.31),
using the formulation in Eq. (1.32), can be expressed as
.. .
Mr q +Cr q +Kr q = F (1.33)
In Eq. (1.33), Mr , Cr , and Kr are diagonal matrices, each matrix is n × n,
while F = ψ T F is n × 1. The masses in Mr are called the modal masses
and the stiffness values in Kr are called modal stiffnesses. However, as mode
shapes are scaled arbitrarily, the modal mass and modal stiffness may not be
unique or well defined. This can be resolved by scaling such that all modal
mass values are 1, thereby yielding a modal mass matrix, Mr , that is an identity
matrix.
In case of nonproportional damping, it is not possible to decouple the
damping matrix. This is called general damping. In such cases, it is common
to convert the EOM of the system to a state-space formulation and solve the
eigenvalue problem. For general damping, the EOM for a damped MDOF
system, as shown in Eq. (1.31), is equivalently expressed as
.
C M X.. K 0n×n X. F
+ = (1.34)
M 0n×n X 0n×n −M X 0n×1
In Eq. (1.34), 0n × n is the n×n zero matrix and 0n × 1 is the n × 1 zero
matrix. Eq. (1.34) can be written as a first-order system as follows:
.
AZ +BZ = F (1.35)

C M
In Eq. (1.35), Z = [X Ẋ ]T , F = [F 0n×1 ]T , A = and
M 0n×n

K 0n×n
B= . The eigenvalue problem for Eq. (1.35) can be solved to
0n×n −M
find the eigenvalues and eigenvectors by solving the following:
(A−1 B − λI )ψ = 0 (1.36)
In Eq. (1.36), λi are the eigenvalues and the corresponding ψ i are the
eigenvectors. An example of a proportionally damped and a nonpropor-
tionally damped two-DOF system is provided in Example 1.4.
Example 1.4: For a proportionally damped two-DOF system shown in

Fig. 1.8, find the undamped and damped natural frequencies and mode
shapes. Diagonalize the EOMs by using modal coordinates. The following
data are provided: m1 = 2 kg, m2 = 5 kg, k1 = k2 = k3 = 104 N/m. Use
proportional damping with a = 0.133 s−1 and b = 6.7 × 10−4 s.
How would the governing EOM be solved if the system is nonpro-
portionally
damped with the following general damping: matrix: C =
17 −1
N-s/m?
−1 27
The EOM of the system in Fig. 1.8 can be identified from the free-body
diagrams of the two masses as
.. . . .
m1 x1 = −k1 x1 − k2 (x1 − x2 ) − c1 x1 − c2 (x1 − x2 ) (E1.4.1)
.. . . .
m2 x2 = k2 (x1 − x2 ) − k3 x2 + c2 (x1 − x2 ) − c3 x2 (E1.4.2)
These EOM can be rewritten as follows:
.. . .
m1 x1 + (k1 + k2 )x1 − k2 x2 + (c1 + c2 )x1 − c2 x2 = 0 (E1.4.3)
.. . .
m2 x2 − k2 x1 + (k2 + k3 )x2 − c2 x1 + (c2 + c3 )x2 = 0 (E1.4.4)
Using the governing EOM in Eqs. (E1.4.3)
and (E1.4.4), the mass matrix
2 0
of the system can be expressed as M = , and the stiffness matrix
0 5

2 × 104 −1 × 104
of the system can be expressed as K = . Using
−1 × 104 2 × 104
the given constants
for proportional damping, the damping matrix can be
13.66 −6.7
written as C =
−6.7 14.065
As seen in Eq. (1.30), evaluating M−1 K will assist in the identification of
the natural frequencies and mode shapes, this matrix is as follows:

−1 10000 −5000
M K=
−2000 4000
Figure 1.8 Two degree-of-freedom system - proportional damping.

The eigenvalues of this matrix yield the following: λ21 = 2641 and
λ22= 11,359.The eigenvectors corresponding to λ1 and λ2 are ψ 1 = [0.5620
0.8271]T and ψ 2 = [0.9650 − 0.2623]T. The eigenvectors are combined
0.5620 0.9650
to form the mode shape matrix as ψ = . The mode
0.8271 −0.2623
shape matrix can be used for the transformation of X = ψq to diagonalize
the mass, damping, and stiffness matrices. For this system, X = [x1 x2 ]T and
q = [q1 q2 ]T . The transformed matrices are as follows:

4.0522 0
Mr =
0 2.2065

1.0702 × 104 0
Kr =
0 2.5063 × 104

7.7075 0
Cr =
0 17.08
The diagonalized elements of Mr and Kr can be used to find the natural
frequencies as

k1r
ω1 = = 51.39 rad/s = 8.18 Hz
m1r

k2r
ω2 = = 106.58 rad/s = 16.96 Hz
m2r
The damping ratios can also be computed from the diagonalized ele-
ments of Mr , Cr , and Kr as
c1r
ξ1 = √ = 0.0185
2 m1r k1r
c2r
ξ2 = √ = 0.0363
2 m2r k2r
The damped natural frequencies of the system are ωd1 = 8.17 Hz and
ωd2 = 16.95 Hz. The transformed EOM can be expressed as
.. .
m1r q1 + c1r q1 + k1r q1 = 0 (E1.4.5)
.. .
m2r q2 + c2r q2 + k2r q2 = 0 (E1.4.6)
Each transformed EOM can be solved independently by using the
solution for an SDOF system, one at a time. These solutions can then be
combined by using the transformation equation to provide the solution for

x1 and x2 .
If the system is nonproportionally damped, the EOM can be expressed
as
.. .
m1 0 x1 c1 + c2 −c2 x1 k 1 + k2 −k2 x1 0
.. + . + =
0 m2 x2 −c2 c2 + c3 x2 −k2 k 2 + k3 x 2 0
(E1.4.7)
For ⎡this system,
⎤ the state vectors are combined in the following form:
x1 (t )
⎢x2 (t )⎥
y(t ) = ⎢ ⎥
⎣x. 1 (t )⎦
.
x2 (t )
The system matrices can be combined as follows:

C M
A= (E1.4.8)
M 02×2

K 02×2
B= (E.1.4.9)
02×2 −M
.
The system can now be expressed as A y +By = 0. The eigenvalue
problem for this system can now be expressed as follows: (A−1 B − λI)
ϕ= 0, where λ represents the eigenvalues and ϕ represents the eigenvectors.
For this nonproportionally damped system, using the general damping
matrix yields the following eigenvalues: −4.12 ± j106.49 and −2.83 ±
j51.32. It may be noted that these eigenvalues are in the form of −ξ i ωi ±
jωdi . The corresponding mode shapes for the general damped system with
nonproportional damping are:
[−0.0004 ± j0.009 0.0002 ± j.00025 0.9649 −0.2623 ± j0.008]T
and
[− 0.0008 ± j0.0109 −0.009 ± j0.0161 0.5621 ± j0.0083 0.8268]T
For the system in Eq. (1.31), transformation into the frequency domain
by using the Fourier transform will result in the following:
−ω2 MX (ω) + jωCX (ω) + KX (ω) = F (ω) (1.37)
It may be noted that X(ω) and F(ω) are n × 1, where n is the number
of DOF of the system. The transformed system in Eq. (1.37) consists of a
system of equations corresponding to the DOF of the MDOF system, this
can be used to express the frequency response as

−1
X (ω) = (−ω2 M ) + ( jωC) + K F (ω) (1.38)
For the frequency response in Eq. (1.38), the output is typically obtained
by exciting one DOF at a time to determine the response of all the DOF
across a frequency sweep. An example of the frequency response of an
MDOF system is provided in Example 1.5.
Example 1.5: For the proportionally damped two-DOF system shown in
Fig. 1.8, plot the frequency response. Use the following data: m1 = 2 kg,
m2 = 5 kg, k1 = k2 = k3 = 104 N/m, C = 0.133M + 6.7 × 10−4 K.

2 0
The system matrices for this system are as follows: M = ,K =
0 5

2 × 104 −1 × 104 13.66 −6.7
,C = . Using the formulation
−1 × 104 2 × 104 −6.7 14.065
in the frequency domain, as shown in Eq. (1.38), with a unit excitation
amplitude of 1 N at each DOF, the magnitude of the frequency response
is plotted in Fig. 1.9.
Figure 1.9 Frequency response magnitude.

It may be noted that it is common to express the frequency response in

alternative forms such as the velocity frequency response or the acceleration
frequency response. These forms are particularly useful for experimental
characterization. Furthermore, it may be noted that the frequency response
is used in conjunction with modal decomposition properties to express
the response in terms of modal superposition. This has not been discussed
in this chapter but can be found in any introductory text on vibration
analysis. A few such books are listed in the references of this chapter
(Balachandran & Magrab, 2019; Inman, 2014; Meirovitch, 1997; Rao, 2007;
Rao, 2017; Rivin, 2003; Thomson & Dahleh, 1998; Tongue, 2002).
Review exercises
1. For an SDOF system attached to a spring-damper unit, as shown in
Fig. 1.3, find the undamped and damped natural frequency as well as the
damping ratio. Also, write the EOM and the transfer function for this system.
The following data can be used: m = 1 kg, c = 100 Ns/m, k = 106 N/m.
2. For the proportionally damped system with two-DOF system shown in
Fig. 1.10, find the natural frequencies (damped and undamped) and mode
shapes. Use the following data: m1 = 2 kg, m2 = 5 kg, k1 =k2 =k3 = 104 N/m.
Use proportional damping with C = 5M+0.0001 K. Use the mode shape
matrix to perform coordinate transformation and decouple the EOMs. Use
the decoupled EOMs to determine the natural frequencies and damping
ratios.
3. Plot the magnitude and phase of the frequency response for the SDOF
system from review exercise 1. Justify the response and discuss the results.
4. Plot the magnitude and phase of the frequency response for the two-
DOF system from review exercise 2 (refer to Fig. 1.10). Discuss the results in
the context of the results from review exercise 2.
Figure 1.10 Two degree-of-freedom system.

References
Balachandran, B., Magrab, E.B., 2019. Vibrations, third ed. Cambridge University Press,
Cambridge, UK.
Choi, Y.-T., Wereley, N.M., 2008. Shock isolation systems using magnetorheological dampers.
J. Vib. Acoust., 130, 024503-1–6.
Dominguez, A., Sedaghati, R., Stiharu, I., 2004. Modelling the hysteresis phenomenon of
magnetorheological dampers. Smart Mater. Struct., 13, 1351–1361.
Inman, D.J., 2014. Engineering Vibration, fourth ed. Pearson, Hoboken, NJ, USA.
Mark, J., Erman, B., Roland, M., 2013. The Science and Technology of Rubber, fourth ed.
Academic Press, Waltham, MA, USA.
Meirovitch, L., 1997. Principles and Techniques of Vibrations. Pearson, Hoboken, NJ, USA.
Rao, S.S., 2007. Vibration of Continuous Systems. Wiley, Hoboken, NJ, USA.
Rao, S.S., 2017. Mechanical Vibrations, sixth ed. Pearson, Hoboken, NJ, USA.
Rivin, E.I., 2003. Passive Vibration Isolation. ASME Press, New York, NY, USA.
Thomson, W.T., Dahleh, M.D., 1998. Theory of Vibration with Applications, fifth ed. Prentice
Hall, Upper Saddle River, NJ, USA.
Tongue, B.H., 2002. Principles of Vibration, second ed. Oxford University Press, New York,
NY, USA.
Truong, T.Q., Ahn, K.K., 2010. A new type of semi-active hydraulic engine mount using
controllable area of inertia track. J. Sound Vib., 329, 247–260.
CHAPTER 2
Viscoelastic modeling—
passive vibration isolators
Contents
2.1 Viscoelasticity 27
2.2 Voigt or Kelvin–Voigt model 28
2.3 Zener model 30
2.4 Maxwell–Voigt model 33
2.5 Generalized Maxwell or Maxwell Ladder model 37
2.6 Voigt fractional model 40
2.7 MV fractional model 42
2.8 Hysteresis model 44
Review exercises 58
References 60
2.1 Viscoelasticity
The exhibition of a combination of viscous and elastic behavior is defined
as viscoelasticity. Viscoelastic behavior is often modeled as a combination
of linear behavior of elastic solids as per Hooke’s law and linear behav-
ior of viscous fluids as per Newton’s law. The theory of viscoelasticity
allows accommodating material behavior that involves storage of mechanical
energy as well as dissipation of energy (Christensen, 2003). Most of the
passive vibration isolators exhibit viscoelastic behavior as they are designed
to provide stiffness and damping characteristics to isolate a system from
dynamic forces while allowing for a means of dissipating energy. The nature
of viscoelasticity can vary significantly from one vibration isolator to another
as the behavior can depend on the design, material selection, geometrical
attributes, environmental conditions, etc. As a result, an appropriate model
that represents key characteristics of the isolator needs to be judiciously
selected. For isotropic materials, it is common to represent the constituent
stress–strain relationship of viscoelastic materials in the form of a governing
differential equation. Such a constituent relationship can be adjusted to allow
incorporating a varying number of spring-damper elements and model any
time lag in resistance to deformation upon the application of a load while
also modeling some aspects of time-dependent behavior. Furthermore, these
Modeling and Analysis of Passive Vibration Isolation Systems. Copyright © 2021 Elsevier Inc.
DOI: 10.1016/B978-0-12-819420-1.00003-0 All rights reserved. 27
constituent relationships can be directly linked to mechanical models to

identify critical characteristics of the vibration isolator design through testing
and model validation. Many viscoelastic materials exhibit strong influence
of temperature and frequency on their dynamic properties. While the extent
of this influence varies from one material to another, it is generally observed
that the influence of these two variables can be divided into three regimes
called the glassy (at a relatively low temperature), transition, and rubbery
(at a relatively high temperature) regions due to molecular mobility at
the microscopic level. While the storage modulus of viscoelastic materials
reduces with increasing temperature (going from glassy to rubbery region),
the loss factor typically increases as the temperature increases from the glassy
region to the transition region and then reduces as the temperature further
increases from the transition region to the rubbery region. An increase in
excitation frequency generally results in an increase in the storage modulus
of viscoelastic materials. Some of these characteristics can be described by
some of the models discussed in this chapter and the subsequent chapters.
2.2 Voigt or Kelvin–Voigt model

The Voigt or Kelvin–Voigt model is the simplest viscoelastic model that is
very commonly used to model a vibration isolator. This model consists of a
spring element and a damping element that are parallel to each other. This
model superposes the elastic behavior and the viscous behavior of an isolator,
a representation of this model is shown in Fig. 2.1.
As can be seen from Fig. 2.1, this model is the same as the general spring-
damper system discussed in Chapter 1. Therefore, the governing equation
of motion (EOM) of this system is the differential equation in Eq. (1.17)
Figure 2.1 Voigt or Kelvin–Voigt model.

Viscoelastic modeling—passive vibration isolators 29
in Chapter 1. For an isolator designed to be under shear, the stress–strain

relationship for the Voigt model of the isolator can be expressed as
dγ
τ (t ) = G0 γ + G1 (2.1)
dt
In Eq. (2.1), τ is the shear stress and γ is the shear strain, and G0 and
G1 are positive coefficients that can be defined in terms of geometrical and
other related parameters of the isolator. The reaction force of the spring-
damper element as a result of any perturbation from static equilibrium can
be expressed as
.
f (t ) = c y + ky (2.2)
In Eq. (2.2), f(t) is the time history of the reaction force in the loading
direction, while y is the corresponding displacement about static equi-
librium. Using zero initial conditions and applying Laplace transform to
Eq. (2.2) results in the following:
F (s) = (k + cs)(s) (2.3)
In Eq. (2.3), F(s) is the Laplace transform of f(t), (s) is the Laplace
transform of y(t), and s is the Laplace operator.
For an isolator under double shear, τ = 2A f
and γ = δh , where δ is the
deflection in the isolator corresponding to the applied force, f. The isolator
under double shear has a thickness h with the area A being subject to shear.
It may be noted that δ and yare being used interchangeably, that is, δ = y,
and (s) represents the Laplace transform of both the variables. Substitution
of these variables in Eq. (2.1) yields the following results:
f δ 1 dδ
= G0 + G1 (2.4)
2A h h dt
Applying Laplace transform to Eq. (2.4) with zero initial conditions
results in the force-displacement relationship in the Laplace domain as
2A
F (s) = (G0 + G1 s)(s) (2.5)
h
Comparison of the results of Eqs. (2.3) and (2.5) provides the coefficients
in Eq. (2.1) in terms of design variables of the isolator as G0 = 2A h
k and
G1 = 2A c. Also, Eq. (2.5) can be used to write the stress-strain transfer
h
function of the isolator as

F (s)/2A
= (G0 + G1 s) = G(s) (2.6)
(s)/h
In Eq. (2.6), G(s) is called the complex modulus. The complex modulus
is defined as the ratio between stress and strain. In the frequency domain,
the complex modulus can be expressed as

G(ω) = (G0 + jωG1 ) = G + jG (2.7)

In Eq. (2.7), G and G are the real and imaginary components of
the complex modulus, called the storage modulus and the loss modulus,

respectively. For the complex modulus in Eq. (2.7), G = G0 and G = ωG1 .
These moduli can be determined from experimental characterization of a
vibration isolator. Furthermore, the coefficients of the complex modulus
can be directly related to the design characteristics of the isolator such as its
geometry, stiffness constant, and damping constant. The ratio between the

loss modulus, G , and the storage modulus, G , is called the loss factor, and
is also commonly referred to as the loss tangent.
The governing EOM of a single-degree-of-freedom (SDOF) system
supported by an isolator with a Voigt model, as shown in Fig. 2.1, can be
expressed as
mÿ + c ẏ + ky = f (t ). (2.8)
In Eq. (2.8), m is the mass of the system supported by the vibration
isolator and f(t) is the external dynamic force. The EOM is written about
the state of static equilibrium, this implies that the EOM does not account
for the static deflection due to the weight of the isolated system.
It may be noted that the governing equations have been derived for
the Voigt model in this section with the isolator designed to be under
double shear. A similar sequence of steps can be carried out for any other
configuration.Furthermore,the viscoelastic model assumes that the behavior
of the isolator exhibits a lag between the applied force and the resulting
deformation. This assumption is typically valid for many vibration isolator
designs due to inherent damping.
2.3 Zener model

The Zener model is a modified form of the Voigt model that introduces
a Maxwell element that is parallel to the spring element, instead of the
damping element in the Voigt model.A Maxwell element consists of a spring
element and a damper in series. This model is shown in Fig. 2.2.
For an isolator under shear, the stress–strain relationship for the Zener
model of the isolator can be expressed as
dτ dγ
τ (t ) + D1 = G0 γ + G1 . (2.9)
dt dt
Viscoelastic modeling—passive vibration isolators 31
Figure 2.2 Zener model.
In Eq. (2.9), D1 , G0 , and G1 are the coefficients of the Zener model. For
an isolator under double shear, Eq. (2.9) can be expressed as
f D1 df δ 1 dδ
+ = G0 + G1 . (2.10)
2A 2A dt h h dt
Applying Laplace transform to Eq. (2.10) with zero initial conditions
results in the following:
2A
F (s)(1 + D1 s) = (G0 + G1 s)(s). (2.11)
h
The variables in Eq. (2.11) are similar to the variables defined in the
previous section for the Voigt model.
The reaction force of the isolator using the Zener model can be expressed
as
f (t ) = ky + c1 (ẏ − ẏ1 ). (2.12)
As there is an additional nodal degree-of-freedom (DOF) due to the
node in the Maxwell element between the spring and the damper, the
equilibrium equation at the node is
c1 (ẏ − ẏ1 ) = k1 y1 . (2.13)
In Eqs. (2.12) and (2.13), y1 is the displacement at the node between
the spring and the damper. Applying Laplace transform with zero initial
conditions to Eq. (2.12) results in the following:
F (s) = (k + c1 s)(s) − c1 sY1 (s). (2.14)
In Eq.(2.14), Y1 (s) is the Laplace transform of y1 (t).Application of Laplace
transform with zero initial conditions to Eq. (2.13) is used to determine Y1 (s)
in terms of (s) as
c1 s
Y1 (s) = (s). (2.15)
k1 + c1 s
Y1 (s) from Eq. (2.15) is substituted in Eq. (2.14) to eliminate Y1 (s) and
express the governing relationship between F(s) and (s) as

c1 kc1
1 + s F (s) = k + s + c1 s (s). (2.16)
k1 k1
The polynomial coefficients of Eq. (2.16) can be directly compared to
Eq. (2.11) to determine the coefficients of the Zener model as D1 = kc11 ,
G0 = 2A h
k,G1 = 2A ( k1 +c1 ).Furthermore,the stress–strain transfer function
h kc1
for the isolator can be expressed in the following form:

F (s)/2A G0 + G1 s
= . (2.17)
(s)/h 1 + D1 s
The force-displacement transfer function for the Zener model can be
expressed as

F (s) k + kc1
k1
+ c1 s
= . (2.18)
(s) 1 + kc11 s
It may be noted that Eq. (2.17) is written in the form of a ratio between
stress and strain to express the ratio as the complex modulus for the Zener
model. Furthermore, it may be noted that the force-displacement ratio in
frequency domain is often referred to as dynamic stiffness in the literature.
The governing EOM of an SDOF system supported by an isolator with
a Zener model, as shown in Fig. 2.2, can be expressed as
mÿ + c1 (ẏ − ẏ1 ) = f (t ). (2.19)
In Eq. (2.19), m is the mass of the system supported by the vibration

isolator.Also,the EOM in Eq.(2.19) will need to be used in conjunction with
the nodal equation in Eq. (2.13) to determine the system response in time
domain. It may be noted that although the Zener model has been developed
for an SDOF isolator in this section, the entire model consists of two DOFs
due to the node in the Maxwell element. The frequency response for the
Zener model can be derived by using Fourier transform for Eq. (2.19) and
eliminating nodal deflection by using the equilibrium equation at the node.
Another random document with
no related content on Scribd:
destroy or shut up his fleet. Having accomplished the great feature of
the task by getting hold of the most decisive position, further effort
must be directed towards, possibly not upon, those points which may
serve him still for bases. In so doing, your fleet must not be divided,
unless overwhelmingly strong, and must not extend its lines of
communication beyond the power of protecting them, unless it be for
a dash of limited duration.
If compelled to choose between fortified ports of the enemy and
his fleet, the latter will be regarded as the true objective; but a
blockade of the ports, or an attack upon them, may be the surest
means of bringing the ships within reach. Thus, in the War of
American Independence, the siege of Gibraltar compelled the British
fleet on more than one occasion to come within fighting reach of the
enemy’s blockading fleet, in order to throw in supplies. That the
allies did not attack, except on one occasion, does not invalidate the
lesson. Corbett in his “Seven Years’ War” points out very justly, in
Byng’s celebrated failure, which cost him his life, that if he had
moved against the French transports, in a neighboring bay, the
French admiral would have had to attack, and the result might have
been more favorable to the British. Such movements are essentially
blows at the communications of the enemy, and if aimed without
unduly risking your own will be in thorough accord with the most
assured principles of strategy. A militarily effective blockade of a
base essential to the enemy will force his fleet either to fight or to
abandon the theater of war. Thus, as has been pointed out elsewhere,
in Suffren’s campaign in Indian Seas, so long as Trincomalee was in
possession of the British, a threat at it was sure to bring them out to
fight, although it was not their principal base. The abandonment of
the theater of war by the navy will cause the arsenal to fall in time,
through failure of resources, as Gibraltar must have fallen if the
British fleet had not returned and supplied it at intervals. Such a
result, however, is less complete than a victory over the enemy’s
navy, which would lead to the same end, and so be a double success,
ships and port.
9. The Value of the Defensive[28]
It is true that in certain respects the defensive has advantages, the

possession of which may even justify an expression, which has been
stated as a maxim of war, that “Defense is a stronger form of war
than Offense is.” I do not like the expression, for it seems to me
misleading as to the determinative characteristics of a defensive
attitude; but it may pass, if properly qualified. What is meant by it is
that in a particular operation, or even in a general plan, the party on
the defense, since he makes no forward movement for the time, can
strengthen his preparations, make deliberate and permanent
dispositions; while the party on the offensive, being in continual
movement, is more liable to mistake, of which the defense may take
advantage, and in any case has to accept as part of his problem the
disadvantage, to him, of the accumulated preparations that the
defense has been making while he has been marching. The extreme
example of preparation is a fortified permanent post; but similar
instances are found in a battle field carefully chosen for advantages
of ground, where attack is awaited, and in a line of ships, which by
the solidarity of its order, and deployment of broadside, awaits an
enemy who has to approach in column with disadvantage as to train
of guns. In so far, the form taken by the defense is stronger than the
form assumed for the moment by the offense.
If you will think clearly, you will recognize that at Tsushima the
Japanese were on the defensive, for their object was to stop, to
thwart, the Russian attempt. Essentially, whatever the tactical
method they adopted, they were to spread their broadsides across
the road to Vladivostok, and await. The Russians were on the
offensive, little as we are accustomed so to regard them; they had to
get through to Vladivostok—if they could. They had to hold their
course to the place, and to break through the Japanese,—if they
could. In short, they were on the offensive, and the form of their
approach had to be in column, bows on,—a weaker form,—which
they had to abandon, tactically, as soon as they came under fire.
In our hostilities with Spain, also, Cervera’s movement before
reaching Santiago was offensive in character, the attitude of the
United States defensive; that is, he was trying to effect something
which the American Navy was set to prevent. There being three
principal Spanish ports, Havana, Cienfuegos, and Santiago, we could
not be certain for which he would try, and should have been before
two in such force that an attempt by him would have assured a
battle. We were strong enough for such a disposition. The two ports
thus to be barred were evidently Havana and Cienfuegos. The
supposed necessity for defending our northern coast left Cienfuegos
open. Had Cervera made for it, he would have reached it before the
Flying Squadron did. The need for keeping the Flying Squadron in
Hampton Roads was imaginary, but it none the less illustrates the
effect of inadequate coast defenses upon the military plan of the
nation.
The author whom I quote (Corbett, “Seven Years’ War,” Vol. I, p.
92), who himself quotes from one of the first of authorities,
Clausewitz, has therefore immediately to qualify his maxim, thus:
“When we say that defense is a stronger form of war, that is, that it
requires a smaller force, if soundly designed, we are speaking, of
course, only of one certain line of operations. If we do not know the
general line of operation on which the enemy intends to attack, and
so cannot mass our force upon it, then defense is weak, because we
are compelled to distribute our force so as to be strong enough to
stop the enemy on any line of operations he may adopt.”
Manifestly, however, a force capable of being strong enough on
several lines of operation to stop an enemy possesses a superiority
that should take the offensive. In the instance just cited, of Cervera’s
approach, the American true policy of concentration would have had
to yield to distribution, between Cienfuegos and Havana. Instead of a
decisive superiority on one position, there would have been a bare
equality upon two. Granting an enemy of equal skill and training, the
result might have been one way or the other; and the only
compensation would have been that the enemy would have been so
badly handled that, to use Nelson’s phrase, he would give no more
trouble that season, and the other American division would have
controlled the seas, as Togo did after August 10, 1904. From the
purely professional point of view it is greatly to be regretted that the
Spaniards and Russians showed such poor professional aptitude.
The radical disadvantage of the defensive is evident. It not only is
the enforced attitude of a weaker party, but it labors under the
further onerous uncertainty where the offensive may strike, when
there is more than one line of operation open to him, as there usually
is. This tends to entail dissemination of force. The advantages of the
defensive have been sufficiently indicated; they are essentially those
of deliberate preparation, shown in precautions of various kinds. In
assuming the defensive you take for granted the impossibility of your
own permanent advance and the ability of the enemy to present
himself before your front in superior numbers; unless you can harass
him on the way and cause loss enough to diminish the inequality.
Unless such disparity exists, you should be on the offensive. On the
other hand, in the defensive it has to be taken for granted that you
have on your side a respectable though inferior battle fleet, and a sea
frontier possessing a certain number of ports which cannot be
reduced without regular operations, in which the armed shipping can
be got ready for battle, and to which, as to a base, they can retire for
refit. Without these two elements there can be no serious defense.
10. Commerce-Destroying and Blockade[29]
It is desirable to explain here what was, and is, the particular specific
utility of operations directed toward the destruction of an enemy’s
commerce; what its bearing upon the issues of war; and how, also, it
affects the relative interests of antagonists, unequally paired in the
matter of sea power. Without attempting to determine precisely the
relative importance of internal and external commerce, which varies
with each country, and admitting that the length of transportation
entails a distinct element of increased cost upon the articles
transported, it is nevertheless safe to say that, to nations having free
access to the sea, the export and import trade is a very large factor in
national prosperity and comfort. At the very least, it increases by so
much the aggregate of commercial transactions, while the ease and
copiousness of water carriage go far to compensate for the increase
of distance. Furthermore, the public revenue of maritime states is
largely derived from duties on imports. Hence arises, therefore, a
large source of wealth, of money; and money—ready money or
substantial credit—is proverbially the sinews of war, as the War of
1812 was amply to demonstrate. Inconvertible assets, as business
men know, are a very inefficacious form of wealth in tight times; and
war is always a tight time for a country, a time in which its positive
wealth, in the shape of every kind of produce, is of little use, unless
by freedom of exchange it can be converted into cash for
governmental expenses. To this sea commerce greatly contributes,
and the extreme embarrassment under which the United States as a
nation labored in 1814 was mainly due to commercial exclusion from
the sea. To attack the commerce of the enemy is therefore to cripple
him, in the measure of success achieved, in the particular factor
which is vital to the maintenance of war. Moreover, in the
complicated conditions of mercantile activity no one branch can be
seriously injured without involving others.
This may be called the financial and political effect of “commerce
destroying,” as the modern phrase runs. In military effect, it is
strictly analogous to the impairing of an enemy’s communications, of
the line of supplies connecting an army with its base of operations,
upon the maintenance of which the life of the army depends. Money,
credit, is the life of war; lessen it, and vigor flags; destroy it, and
resistance dies. No resource then remains except to “make war
support war;” that is, to make the vanquished pay the bills for the
maintenance of the army which has crushed him, or which is
proceeding to crush whatever opposition is left alive. This, by the
extraction of private money, and of supplies for the use of his troops,
from the country in which he was fighting, was the method of
Napoleon, than whom no man held more delicate views concerning
the gross impropriety of capturing private property at sea, whither
his power did not extend. Yet this, in effect, is simply another
method of forcing the enemy to surrender a large part of his means,
so weakening him, while transferring it to the victor for the better
propagation of hostilities. The exaction of a pecuniary indemnity
from the worsted party at the conclusion of a war, as is frequently
done, differs from the seizure of property in transit afloat only in
method, and as peace differs from war. In either case, money or
money’s worth is exacted; but when peace supervenes, the method of
collection is left to the Government of the country, in pursuance of its
powers of taxation, to distribute the burden among the people;
whereas in war, the primary object being immediate injury to the
enemy’s fighting power, it is not only legitimate in principle, but
particularly effective, to seek the disorganization of his financial
system by a crushing attack upon one of its important factors,
because effort thus is concentrated on a readily accessible,
fundamental element of his general prosperity. That the loss falls
directly on individuals, or a class, instead of upon the whole
community, is but an incident of war, just as some men are killed
and others not. Indirectly, but none the less surely, the whole
community, and, what is more important, the organized government,
are crippled; offensive powers impaired.
But while this is the absolute tendency of war against commerce,
common to all cases, the relative value varies greatly with the
countries having recourse to it. It is a species of hostilities easily
extemporized by a great maritime nation; it therefore favors one
whose policy is not to maintain a large naval establishment. It opens
a field for a sea militia force, requiring little antecedent military
training. Again, it is a logical military reply to commercial blockade,
which is the most systematic, regularized, and extensive form of
commerce-destruction known to war. Commercial blockade is not to
be confounded with the military measure of confining a body of
hostile ships of war to their harbor, by stationing before it a
competent force. It is directed against merchant vessels, and is not a
military operation in the narrowest sense, in that it does not
necessarily involve fighting, nor propose the capture of the
blockaded harbor. It is not usually directed against military ports,
unless these happen to be also centers of commerce. Its object, which
was the paramount function of the United States Navy during the
Civil War, dealing probably the most decisive blow inflicted upon the
Confederacy, is the destruction of commerce by closing the ports of
egress and ingress. Incidental to that, all ships, neutrals included,
attempting to enter or depart, after public notification through
customary channels, are captured and confiscated as remorselessly
as could be done by the most greedy privateer. Thus constituted, the
operation receives far wider scope than commerce-destruction on the
high seas; for this is confined to merchantmen of belligerents, while
commercial blockade, by universal consent, subjects to capture
neutrals who attempt to infringe it, because, by attempting to defeat
the efforts of one belligerent, they make themselves parties to the
war.
In fact, commercial blockade, though most effective as a military
measure in broad results, is so distinctly commerce-destructive in
essence, that those who censure the one form must logically proceed
to denounce the other. This, as has been seen, Napoleon did; alleging
in his Berlin Decree, in 1806, that war cannot be extended to any
private property whatever, and that the right of blockade is restricted
to fortified places, actually invested by competent forces. This he had
the face to assert, at the very moment when he was compelling every
vanquished state to extract, from the private means of its subjects,
coin running up to hundreds of millions to replenish his military
chest for further extension of hostilities. Had this dictum been
accepted international law in 1861, the United States could not have
closed the ports of the Confederacy, the commerce of which would
have proceeded unmolested; and hostile measures being
consequently directed against men’s persons instead of their trade,
victory, if accomplished at all, would have cost three lives for every
two actually lost.
It is apparent, immediately on statement, that against commerce-
destruction by blockade, the recourse of the weaker maritime
belligerent is commerce-destruction by cruisers on the high sea.
Granting equal efficiency in the use of either measure, it is further
plain that the latter is intrinsically far less efficacious. To cut off
access to a city is much more certainly accomplished by holding the
gates than by scouring the country in search of persons seeking to
enter. Still, one can but do what one can. In 1861 to 1865, the
Southern Confederacy, unable to shake off the death grip fastened on
its throat, attempted counteraction by means of the “Alabama,”
“Sumter,” and their less famous consorts, with what disastrous
influence upon the navigation—the shipping—of the Union it is
needless to insist. But while the shipping of the opposite belligerent
was in this way not only crippled, but indirectly was swept from the
seas, the Confederate cruisers, not being able to establish a blockade,
could not prevent neutral vessels from carrying on the commerce of
the Union. This consequently suffered no serious interruption;
whereas the produce of the South, its inconvertible wealth—cotton
chiefly—was practically useless to sustain the financial system and
credit of the people. So, in 1812 and the two years following, the
United States flooded the seas with privateers, producing an effect
upon British commerce which, though inconclusive singly, doubtless
co-operated powerfully with other motives to dispose the enemy to
liberal terms of peace. It was the reply, and the only possible reply, to
the commercial blockade, the grinding efficacy of which it will be a
principal object of these pages to depict. The issue to us has been
accurately characterized by Mr. Henry Adams, in the single word
“Exhaustion.”[30]
Both parties to the War of 1812 being conspicuously maritime in
disposition and occupation, while separated by three thousand miles
of ocean, the sea and its navigable approaches became necessarily
the most extensive scene of operations. There being between them
great inequality of organized naval strength and of pecuniary
resources, they inevitably resorted, according to their respective
force, to one or the other form of maritime hostilities against
commerce which have been indicated. To this procedure combats on
the high seas were merely incidental. Tradition, professional pride,
and the combative spirit inherent in both peoples, compelled fighting
when armed vessels of nearly equal strength met; but such contests,
though wholly laudable from the naval standpoint, which under
ordinary circumstances cannot afford to encourage retreat from an
equal foe, were indecisive of general results, however meritorious in
particular execution. They had no effect upon the issue, except so far
as they inspired moral enthusiasm and confidence. Still more, in the
sequel they have had a distinctly injurious effect upon national
opinion in the United States. In the brilliant exhibition of enterprise,
professional skill, and usual success, by its naval officers and
seamen, the country has forgotten the precedent neglect of several
administrations to constitute the navy as strong in proportion to the
means of the country as it was excellent through the spirit and
acquirements of its officers. Sight also has been lost of the actual
conditions of repression, confinement, and isolation, enforced upon
the maritime frontier during the greater part of the war, with the
misery and mortification thence ensuing. It has been widely inferred
that the maritime conditions in general were highly flattering to
national pride, and that a future emergency could be confronted with
the same supposed facility, and as little preparation, as the odds of
1812 are believed to have been encountered and overcome. This
mental impression, this picture, is false throughout, alike in its
grouping of incidents, in its disregard of proportion, and in its
ignoring of facts. The truth of this assertion will appear in due course
of this narrative, and it will be seen that, although relieved by many
brilliant incidents, indicative of the real spirit and capacity of the
nation, the record upon the whole is one of gloom, disaster, and
governmental incompetence, resulting from lack of national
preparation, due to the obstinate and blind prepossessions of the
Government, and, in part, of the people.
Command of the Sea Decisive[31]
It is not the taking of individual ships or convoys, be they few or

many, that strikes down the money power of a nation; it is the
possession of that overbearing power on the sea which drives the
enemy’s flag from it, or allows it to appear only as a fugitive; and
which, by controlling the great common, closes the highways by
which commerce moves to and from the enemy’s shores. This
overbearing power can only be exercised by great navies, and by
them (on the broad sea) less efficiently now than in the days when
the neutral flag had not its present immunity.[32] It is not unlikely
that, in the event of a war between maritime nations, an attempt may
be made by the one having a great sea power and wishing to break
down its enemy’s commerce, to interpret the phrase “effective
blockade” in the manner that best suits its interests at the time; to
assert that the speed and disposal of its ships make the blockade
effective at much greater distances and with fewer ships than
formerly. The determination of such a question will depend, not
upon the weaker belligerent, but upon neutral powers; it will raise
the issue between belligerent and neutral rights; and if the
belligerent have a vastly overpowering navy he may carry his point,
just as England, when possessing the mastery of the seas, long
refused to admit the doctrine of the neutral flag covering the goods.
11. Strategic Features of the Gulf of Mexico and
the Caribbean[33]
In the special field proposed for our study, there are two principal
points of such convergence—or divergence: the mouth of the
Mississippi River, and the Central-American Isthmus. At the time
when these lectures were first written the opinion of the world was
hesitating between Panama and Nicaragua as the best site for a canal
through the Isthmus. This question having now been settled
definitively in favor of Panama, the particular point of convergence
for trade routes passing through the Caribbean for the Pacific will
continue at Colon, whither it for so long has been determined
because there is the terminus of the Panama Railroad.
These two meeting points or cross-roads have long been, and still
are, points of supreme interest to all mankind. At the one all the
highways of the Mississippi valley, all the tributaries and sub-
tributaries of the great river, meet, and thence they part. At the other
all highways between the Atlantic and Pacific focus and intersect.
The advancing population and development of the Mississippi valley,
and the completion of the Panama Canal, will work together to cause
this international interest to grow proportionately in the future.
Among the great Powers of the world, no one is concerned so vitally
in this progress as is the United States; because of her possession of
one of these centers, the mouth of the Mississippi with its huge back
country, and because of her geographical nearness to the other. This
peculiar interest, which is natural and inevitable in virtue of
proximity, is emphasized by the national policy known as the
Monroe Doctrine; and still more by the particular result of the
Doctrine which has involved the control, administration, and
military protection of that belt of Isthmian territory called the
Panama Canal Zone.
[In the intervening pages, it is shown that the triangle drawn on
the map (p. 101) includes all points of strategic importance, these
being indicated by black squares. Cuba is the key to the Gulf of
Mexico, and also controls three entrances to the Caribbean—the
Yucatan, Windward, and Mona Passages. The entrances, the chief
points of destination (Jamaica and the Isthmus), and the routes
thither, constitute the main objects of military control in the
Caribbean.—Editor.]
... Taking all together, control over transit depending upon
situation only, other conditions being equal, is greatest with Jamaica,
next with Cuba, least with the Lesser Antilles.
Accepting these conclusions as to control over transit, we now
revert to that question to which all other inquiries are subsidiary,
namely, Which of the three bases of operations in the Caribbean—
one of the Lesser Antilles, Jamaica, or Cuba with its sphere of
influence—is most powerful for military control of the principal
objective points in the same sea? These principal objectives are
Jamaica and the Isthmus; concerning the relative importance of
which it may be remarked that, while the Isthmus intrinsically, and
to the general interest of the world, is incomparably the more
valuable, the situation of Jamaica gives such command over all the
approaches to the Isthmus, as to make it in a military sense the
predominant factor in the control of the Caribbean. Jamaica is a pre-
eminent instance of central position, conferring the advantage of
interior lines, for action in every direction within the field to which it
belongs.
Military control depends chiefly upon two things, position and
active military strength. As equal military strength has been assumed
throughout, it is now necessary only to compare the positions held by
other states in the field with that of the occupant of Cuba. This
inquiry also is limited to the ability either to act offensively against
these objective points, or, on the contrary, to defend them if already
held by oneself or an ally; transit having been considered already.
Control by virtue of position, over a point external to your
territory, depends upon nearness in point of time and upon the
absence of obstacles capable of delaying or preventing your access to
it.
Both Santiago (or Guantanamo) and Cienfuegos are nearer to the
Isthmus than is any other one of the first-class strategic points that
have been chosen on the borders of the Caribbean Sea, including
Samana Bay and St. Thomas. They are little more than half the
distance of the British Santa Lucia and the French Martinique. The
formidable island and military stronghold of Jamaica, within the sea,
is nearer the Isthmus than Guantanamo is, by one hundred and fifty
miles, and than Cienfuegos by yet more.
Taking into consideration situation only, Jamaica is admirably
placed for the control of the Caribbean. It is equidistant from Colon,
from the Yucatan Passage, and from the Mona Passage. It shares
with Guantanamo and Santiago control of the Windward Passage,
and of that along the south coast of Cuba; while, with but a slight
stretching out of its arm, it reaches the routes from the Gulf of
Mexico to the Isthmus. Above all, as towards Cuba, it so blocks the
road to the Isthmus that any attempt directed upon the Isthmus
from Cuba must first have to account with the military and naval
forces of Jamaica.
There are, however, certain deductions to be made from the
strength of Jamaica that do not apply as forcibly to Cuba. Leaving to
one side the great and widely scattered colonial system of Great
Britain, which always throws that empire on the defensive and
invites division of the fleet, owing to the large number of points open
to attack, and confining our attention strictly to the field before us, it
will be observed that in a scheme of British operations Jamaica is
essentially, as has been said before, an advanced post; singularly well
situated, it is true, but still with long and difficult communications.
Its distance from Antigua, a possible intermediate base of supplies, is
over nine hundred miles; from Santa Lucia, the chief British naval
station in the Lesser Antilles, over one thousand miles, not less than
three days’ economical steaming. Great Britain, if at war with a state
possessing Cuba, is shut out from the Windward Passage by
Guantanamo, and from the Gulf of Mexico by Havana. The Mona
Passage, also, though not necessarily closed, will be too dangerous to
be relied upon. For these reasons, in order to maintain
communications with Jamaica, an intermediate position and depot,
like Santa Lucia, will be urgently needed. Supplies coming from
Bermuda, Halifax, or England would probably have to be collected
first there, or at Antigua, and thence make a more secure, but still
exposed, voyage to Kingston. The north coasts of Cuba and Haiti
must be looked upon as practically under the control of the Cuban
fleet, in consequence of the command which it exercises over the
Windward Passage, by virtue of position.
The possessor of Cuba, on the contrary, by his situation has open
communication with the Gulf of Mexico, which amounts to saying
that he has all the resources of the United States at his disposal,
through the Mississippi Valley. Cruisers from Jamaica attempting to
intercept that trade would be at a great disadvantage, especially as to
coal, compared with their enemy resting upon Havana. Cruisers from
Havana, reaching their cruising ground with little or no
consumption, can therefore remain longer, and consequently are
equivalent to a greater number of ships. On the other hand, cruisers
from Santiago could move almost with impunity by the north side of
Haiti as far as the Mona Passage, and beyond that without any other
risk than that of meeting and fighting vessels of equal size. If they
stretch their efforts toward the Anegada Passage, they would feel the
same disadvantage, relatively to cruisers from Santa Lucia, that
Jamaica cruisers in the Gulf would undergo as compared with those
from Havana; but by inclining their course more to the northward, to
or about the point Q (see map, page 101), they would there be
equidistant from Guantanamo and Santa Lucia, and so on an
equality with the latter, while at the same time in a position gravely
to endanger supplies from any point in North America. If it be
replied that Bermuda can take care of these cruisers at Q, the answer
is plain: on the supposition of equal forces, it can do so only by
diminishing the force at Santa Lucia. In short, when compared with
Jamaica, in respect of strategic relations to Bermuda, Halifax, and
Santa Lucia, Cuba enjoys the immense advantage of a central
position, and of interior lines of communication, with consequent
concentration of force and effort.
It is not easy to see how, in the face of these difficulties, Great
Britain, in the supposed case of equal force in this theater of war,
could avoid dividing her fleet sufficiently to put Jamaica at a
disadvantage as to Cuba. In truth, Cuba here enjoys not only the
other advantages of situation already pointed out, but also that of
being central as regards the enemy’s positions; and what is, perhaps,
even more important, she possesses secure interior land lines of
supply and coal between the points of her base, while covering the
sea lines in her rear, in the Gulf of Mexico. For Guantanamo and
Santiago have communication by rail with Havana, while the island
itself covers the lines from Havana to the Gulf coast of the United
States; whereas Jamaica depends wholly upon the sea, by lines of
communication not nearly as well sheltered.
Contrasted with Cuba, Jamaica is seen to be, as has been more
than once said, a strong advanced post, thrust well forward into the
face of an enemy to which it is much inferior in size and resources,
and therefore dependent for existence upon its power of holding out,
despite uncertain and possibly suspended communication. Its case
resembles that of Minorca, Malta, Gibraltar, the endurance of which,
when cut off from the sea, has always been measurable. The question
here before us, however, is not that of mere holding out on the
defensive, which would be paralysis. If Cuba can reduce Jamaica to a
passive defensive, Jamaica disappears as a factor in the control of the
Caribbean and Isthmus—no obstacle then stands in the way of Cuba
using her nearness to Panama. If Cuba can bring about a scarcity of
coal at Kingston she achieves a strategic advantage; if a coal famine,
the enemy’s battle fleet must retire, probably to the Lesser Antilles.
The case of Jamaica, contrasted with Cuba, covers that of all
strategic points on the borders of the Caribbean Sea, east, west,
north, or south. Almost on the border itself, although within it,
Jamaica has in nearness, in situation, in size, and in resources, a
decisive advantage over any of the ports of Haiti or of the smaller
islands. If Jamaica is inferior to Cuba, then is each of the other
points on the circumference, and, it may be added, all of them
together....
[It is shown that, while Santa Lucia is essential to Jamaica, the two
are too far apart to work together in concert. As for the Lesser
Antilles, they may be said to control the approaches from Europe,
while Cuba controls those from North America; but the Antilles are
twice as far from the Isthmus as Cuba is, and much weaker in
resources.—Editor.]
As to resources, those of all the West India islands for war will
depend mainly upon the policy and preparation of the governments.
Except Cuba, they are deficient in natural resources adequately
developed. Outside of direct governmental action it can only be said
that the much greater population of Cuba will draw more supplies
and furnish more material for troops and garrisons. At present, as
already noted, the resources of the United States are in effect also the
resources of Cuba.
As between the three possible bases for attempted control of the
Caribbean, no doubts can remain that Cuba is the most powerful,
Jamaica next, and the Antilles least. Jamaica being where it is, Cuba
cannot put forth her power against the Isthmus or against the lines
of transit in the Caribbean, until she has materially reduced, if not
neutralized, the offensive power of her smaller opponent. Upon the
supposition of equal fleets, if the Cuban fleet move against the
Isthmus, or into the Caribbean, it uncovers its communications; if it
seeks to cover these, it divides its force. Jamaica exactly meets the
case supposed in a previous chapter: “If, in moving upon the coveted
objective you pass by a strategic point held by the enemy, capable of
sheltering his ships—a point from which he may probably intercept
your supplies of coal or ammunition, the circle of influence of that
point will require your attention and reduce your force.”
In that case it was laid down that, if you cannot observe the port
without reducing your fleet below that of the enemy, you must not
divide it; either the intermediate point must be taken, or, if you think
you can accomplish your special aim with the supplies on board, you
may cut loose from your base, giving up your communications.
Undoubtedly, the same difficulty would be felt by the Jamaica fleet, if
it moved away from home leaving the Cuban fleet in port in Santiago
or Guantanamo; but, of the two, Jamaica has the inside track. It is
not so with operations based upon the Lesser Antilles only, and
directed against the Isthmus, or against any position in the western
basin of the Caribbean, Cuba being hostile; the line of
communication in that case is so long as to be a very serious
comparative disadvantage.
Upon the whole, then, Jamaica, though less powerful than Cuba,
seems to deserve the title of the “key to the Caribbean.” Only when
Cuba has mastered it can she predominantly control the positions of
that sea. But if Jamaica in this sense be the key, Cuba has the grip
that can wrest it away. Secure as to her own communications, in the
rear, towards the Gulf of Mexico, Cuba has it in her power to impose
upon her enemy a line so long and insecure as to be finally
untenable. First a scarcity of coal, then a famine, lastly the retreat of
the Jamaica fleet to the most available coal station. Such is the
solution I believe possible to the military problem of the Caribbean
as dependent upon geographical conditions,—that is, upon positions;
concerning which Napoleon has said that “War is a business of
positions.” The instant the Cuban fleet has gained a decided
superiority over that of Jamaica, it can take a position covering at
once the approaches to that island and the Windward Channel,
keeping all its own ships in hand while cutting off the enemy’s
supplies and reinforcements. The converse is not true of the Jamaica
fleet, in case it gains a momentary superiority, because the southern
ports of Cuba should be able to receive supplies by land, from the
Gulf of Mexico through Havana.
The general discussion of the strategic features of the Gulf of
Mexico and Caribbean ends here; but the treatment of the subject
will not be complete, unless there be some further specific
consideration of the bearing which the conclusions reached have
upon the facilities of the United States for naval action in the region
studied.
[The political developments between 1887 and 1911 are here
considered, including the growth of the American Navy; the
construction of the Panama Canal; the acquisition by the United
States of strategic points along the line from Key West to Culebra
Island, centering at Guantanamo and “most effectual for military and
naval action in the Caribbean;” and, finally, our increased
responsibilities arising from the growth of the German Navy and the
consequent limitation of England’s co-operation in support of the
Monroe Doctrine.—Editor.]
... The Caribbean Sea and the Isthmus of Panama furnish the
student of naval strategy with a very marked illustration of the
necessity of such cohesion and mutual support between military
positions assumed; as well as between those positions and the army
in the field,—that is, the navy. It affords therefore a subject of the
first importance for such a student to master, and that in fuller detail
than is expedient for a series of lectures, the object of which should
be to suggest lines of thought, rather than to attempt exhaustive
treatment. For an American naval officer, the intimate relation of the
Isthmus and its coming canal to the mutual support of the Atlantic
and Pacific coasts renders the subject doubly interesting. This
interest is yet farther increased by the consideration that the general
international importance to commerce of such a point as the Canal
can scarcely fail to make the conditions of its tenure and use a source
of international difference and negotiation, which often are war
under another form; that is, the solution depends upon military
power, even though held in the background. There are questions
other than commercial dependent upon the tenure of the Isthmus, of
which I will not here speak explicitly. To appreciate them fully there
must be constant reading and reflection upon the general topics of
the day.
One thing is sure: in the Caribbean Sea is the strategic key to the
two great oceans, the Atlantic and Pacific, our own chief maritime
frontiers.
12. Principles of Naval Administration
Opposing Elements[34]
Administration being a term of very general application, it will be

expected that that of the navy should present close analogies, and
even points of identity, with other forms of administration; for
instance, that in it, as elsewhere, efficiency of result will be better
secured by individual responsibility than by collective responsibility.
But, along with general resemblance, naval administration is very
clearly and sharply differentiated by the presence of an element
which is foreign to almost all other activities of life in countries like
Great Britain and the United States. The military factor is to it not
merely incidental, but fundamental; whatever other result may be
achieved, naval administration has failed unless it provides to the
nation an efficient fighting body, directed by well-trained men,
animated by a strong military spirit. On the other hand, many of the
operations connected with it differ from those common to civil life
only in a certain particularity of method. This is true in principal
measure of the financial management, of the medical establishment,
and to a considerable though much smaller degree of the
manufacturing processes connected with the production of naval
material. The business routine of even the most military department
of a naval administration is in itself more akin to civil than to
military life: but it by no means follows that those departments
would be better administered under men of civil habits of thought
than by those of military training. The method exists for the result,
and an efficient fighting body is not to be attained by weakening the
appreciation of military necessities at the very fountain head of their
supply in the administration. This necessary appreciation can be the
result only of personal experience of good and bad through the
formative period of life.
We find, therefore, at the very outset of our inquiry two
fundamental yet opposing elements, neither of which can be
eliminated. Nor can they be reconciled, in the sense of becoming
sympathetic. In its proper manifestation the jealousy between the
civil and military spirits is a healthy symptom. They can be made to

Ebook Modeling and Analysis of Passive Vibration Isolation Systems PDF Full Chapter PDF

Uploaded by

Copyright:

Available Formats

Ebook Modeling and Analysis of Passive Vibration Isolation Systems PDF Full Chapter PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ebook Modeling and Analysis of Passive Vibration Isolation Systems PDF Full Chapter PDF

Uploaded by

Copyright:

Available Formats

Modeling and Analysis of Passive

Vibration Isolation Systems - eBook

Copyright © 2021 Elsevier Inc. All rights reserved.

Library of Congress Cataloging-in-Publication Data

For Information on all Elsevier publications visit our website at

Publisher: Matthew Deans

Typeset by Aptara, New Delhi, India

2. Viscoelastic modeling—passive vibration isolators 27

3. Vibration isolation system modeling 61

4. Vibration isolation systems—nonlinear models 91

5. Modeling elastomer characteristics 119

6. Modeling inertia effect 143

7. Elastomeric vibration isolator design 171

Passive vibration isolators are widely used in areas such as automotive,

Dedicated to my mother, whose life was a beacon of perseverance, resilience,

1.2 Isolator materials

Table 1.1 Commonly used elastomeric materials.

In some cases, it is common to use a separate damper to augment

examples of dampers that have been used in some vibration isolation

1.3 Common elastomeric isolator designs

Table 1.2 Examples of passive elastomeric isolator designs.

1.4 Stiffness and damping

and deflection as Fx = k1 + k2 . Therefore, for the two springs in parallel,

In Eq. (1.3), |k∗ | is the magnitude of complex stiffness, often called

In Eq. (1.5), v is the relative velocity corresponding to the applied force, F.

In Eq. (1.7), ξ is the damping ratio. A system is said to be undamped

For hysteretic damping, an equivalent damping ratio, ξ eq , and an equiva-

1.5 Single-degree-of-freedom system

Figure 1.2 Undamped SDOF system-free vibration.

The solution in Eq. (1.13) can be expressed in different forms depending

Figure 1.3 Damped SDOF system—free vibration.

In Eq. (1.16), c1 and c2 are two 

In Eq. (1.17), f is the excitation force. The solution to the EOM in

and can be expressed as

The formulation in Eq. (1.22) is commonly called the state-space form

Figure 1.4 Undamped SDOF—free vibration response.

Example 1.2a: For an undamped SDOF system, as shown in Figure 1.2,

Figure 1.5 Damped SDOF—free vibration response.

Figure 1.6 Magnitude response.

Figure 1.7 Phase response.

1.6 Multiple-degree-of-freedom system

Example 1.4: For a proportionally damped two-DOF system shown in

Figure 1.8 Two degree-of-freedom system - proportional damping.

combined by using the transformation equation to provide the solution for

−ω2 MX (ω) + jωCX (ω) + KX (ω) = F (ω) (1.37)

can be used to express the frequency response as

Figure 1.9 Frequency response magnitude.

It may be noted that it is common to express the frequency response in

Figure 1.10 Two degree-of-freedom system.

constituent relationships can be directly linked to mechanical models to

2.2 Voigt or Kelvin–Voigt model

Figure 2.1 Voigt or Kelvin–Voigt model.

in Chapter 1. For an isolator designed to be under shear, the stress–strain

function of the isolator as

the complex modulus can be expressed as

2.3 Zener model

Figure 2.2 Zener model.

for the isolator can be expressed in the following form:

mÿ + c1 (ẏ − ẏ1 ) = f (t ). (2.19)

In Eq. (1.16), c1 and c2 are two