Professional Documents
Culture Documents
(Download PDF) System Dynamics For Engineering Students Concepts and Applications 2Nd Edition Nicolae Lobontiu Full Chapter PDF
(Download PDF) System Dynamics For Engineering Students Concepts and Applications 2Nd Edition Nicolae Lobontiu Full Chapter PDF
https://ebookmass.com/product/aerodynamics-for-engineering-
students-7th-edition-e-l-houghton/
https://ebookmass.com/product/chemistry-for-engineering-
students-4th-edition-lawrence-s-brown/
https://ebookmass.com/product/orbital-mechanics-for-engineering-
students-fourth-edition-edition-howard-d-curtis/
https://ebookmass.com/product/process-industry-economics-
principles-concepts-and-applications-2nd-edition-david-brennan/
System Dynamics 4th Edition William J. Palm
https://ebookmass.com/product/system-dynamics-4th-edition-
william-j-palm/
https://ebookmass.com/product/power-system-engineering-2nd-
edition-d-p-kothari-i-g-nagrath/
https://ebookmass.com/product/machine-learning-for-business-
analytics-concepts-techniques-and-applications-with-jmp-pro-2nd-
edition-galit-shmueli/
https://ebookmass.com/product/etextbook-pdf-for-kinematics-and-
dynamics-of-machines-2nd-edition/
https://ebookmass.com/product/intermediate-algebra-connecting-
concepts-through-applications-2nd-edition-mark-clark/
System Dynamics for
Engineering Students
Concepts and Applications
Second Edition
Nicolae Lobontiu
University of Alaska Anchorage
Academic Press is an imprint of Elsevier
125 London Wall, London EC2Y 5AS, United Kingdom
525 B Street, Suite 1800, San Diego, CA 92101-4495, United States
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom
Copyright © 2018 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.
ISBN: 978-0-12-804559-6
This text is a modern treatment of system dynamics and its relation to traditional me-
chanical engineering problems as well as modern microscale devices and machines.
It provides an excellent course of study for students who want to grasp the funda-
mentals of dynamic systems, and it covers a significant amount of material also
taught in engineering modeling, systems dynamics, and vibrations, all combined
in a dense form. The book is designed as a text for juniors and seniors in aerospace,
mechanical, electrical, biomedical, and civil engineering. It is useful for understand-
ing the design and development of micro- and macroscale structures, electric and
fluidic systems with an introduction to transduction, and numerous simulations us-
ing MATLAB and SIMULINK.
The creation of machines is essentially what much of engineering is all about.
Critical to almost all machines imaginable is a transient response, which is funda-
mental to their functionality and needs to be our primary concern in their design.
This might be in the form of changing voltage levels in a sensor, the deflection of
a spring supported mass, or the flow of fluid through a device. The phenomena
that govern dynamics are not simply its mechanical components but often involve
the dynamics of transducers as well, which are often electromechanical or fluidic
based. This text not only discusses traditional electromagnetic type actuators but
also ventures into electrostatics, which are the dominant form of actuators in micro-
electromechanical systems.
This book presents an opportunity for introducing dynamic systems to scientists
and engineers who are concerned with the engineering of machines both at the mi-
cro- and macroscopic scale. Mechanism and movement are considered from the
types of springs and joints that are critical to micromachined, lithographic-based de-
vices to traditional models of macroscale electrical, fluidic, and electromechanical
systems. The examples discussed and the problems at the end of each chapter
have applicability at both scales. In essence this is a more modern treatment of
dynamical systems, presenting views of modeling and substructures more consistent
with the variety of problems that many engineers will face in the future. Any univer-
sity with a substantive interest in microscale engineering would do well to consider a
course that covers the material herein. Finally, this text lays the foundation and
framework for the development of controllers applied to these dynamical systems.
xv
Preface
xvii
xviii Preface
2. It uses modeling and solution techniques that are familiar from other disciplines,
such as physics or ordinary differential equations, and subsequently introduces
new modeling and solution procedures.
3. It provides a rather even coverage (space) to each book chapter.
4. While various chapter structures are possible in a system dynamics text, this book
proposes a sequence that was intended to be systematic and consistent with the
logical structure and progression of the presented material.
As such, the book begins with an introductory Chapter 1, which offers an over-
view of the main aspects of a system dynamics course for engineering students. The
next four chaptersdChapters 2e5dare dedicated to mechanical (Chapters 2 and 3),
electrical (Chapter 4), and fluid and thermal (Chapter 5) system modeling. They
contain basic information on components, systems, and the principal physical and
mathematical tools enabling to model a dynamic system and determine its solution.
Dynamic systems modeling is performed both by means of Lagrange’s equations
and of methods designed for a particular system, such as Newton’s second law of
motion for mechanical system or Kirchhoff’s laws for electrical systems. These
chapters also define and illustrate analogies between the different systems.
Once the main engineering dynamic systems have been studied, Chapter 6 pre-
sents the Laplace transform technique, a mathematical tool that allows simplifying
the differential equation solution process for any of the individual systems. This
chapter is directly connected to the next segment of the book, containing Chapters
7e9. Chapter 7 introduces the transfer function approach, which facilitates
modeling a dynamic system directly in the Laplace domain, by expressing the output
as the product between the transfer function and the input. The complex impedance,
which is actually a transfer function connecting the Laplace-transformed input and
output of a specific system element, is also introduced and thoroughly treated in this
chapter. Chapter 8 studies the state space modeling and solution approach, which is
also related to the Laplace transform of Chapter 6 and the transfer function of
Chapter 7. Chapter 9 discusses modeling system dynamics in the frequency domain
by means of the sinusoidal (harmonic) transfer function. Chapter 10 analyzes
coupled-field (or multiple-field) dynamic systems, which are combinations of
mechanical, electrical, magnetic, piezoelectric, fluid, or thermal systems. In this
chapter, dynamic models are formulated and solved by means of the procedures
studied in previous chapters.
Because of the partial and natural overlap between system dynamics and
controls, the majority of textbooks on either of these two areas contain coverage
of material from the adjoining domain. Consistent with this approach, Chapters
11e13 cover basic concepts of feedback controls, as follows: Chapter 11 discusses
components and block diagrams for feedback control system modeling; Chapter 12
covers the stability of control systems; Chapter 13 presents notions of feedback
controls in the time and frequency domains.
The book also includes six appendixes: Appendix A is a refresher on complex
numbers, Appendix B is a review of matrix algebra, Appendix C gives a summary
Preface xix
to system dynamics problems are the ones provided by Simulink, the graphical user
interface program built atop MATLAB, and applications are included in almost all
the chapters of solved and proposed exercises that can be approached by Simulink.
Through a companion website, the book comprises more ancillary support ma-
terial, including companion book chapters with extensions to the printed book
(with more advanced topics, details of the printed book material, and additional
solved examples, this section could be of interest and assistance to both the
instructor and the motivated student).
Whenever possible, alternative solution methods have been provided in the text
to enable using the algorithm that best suits various individual approaches to the
same problem.
The ancillary material also comprises an instructor’s manual, an image bank of
figures from the book, MATLAB code for the book’s solved examples, and Power-
Point lecture slides. After publication and as a result of specific requirements or sug-
gestions expressed by instructors who adopted the text and feedback from students,
additional problems resulting from this interaction will be provided on the website,
as well as corrections of the unwanted but possible errors.
To make distinction between variables, small-cap symbols are used for the time
domain (such as f for force, m for moment, or v for voltage), whereas capital symbols
denote Laplace transforms (such as F for force, M for moment, or V for voltage).
With regard to matrix notation, the probably old-fashioned symbols { } for vectors
and [ ] for matrices are used here, which can be replicated easily on the board.
Several solved examples and end-of-chapter problems in this book resulted from
exercises that I have used and tested in class over the years while teaching system
dynamics, and I am grateful to all the students who contributed to enhancing the
scope and quality of the original variants. I am indebted to the anonymous academic
reviewers who critically analyzed this text. They have made valid suggestions for
improvement, which were well taken and applied to this current second edition.
I am very thankful to Steven Merken, Senior Acquisition Editor at Elsevier
Engineering & Computing Textbooks, whose commitment to this project and quality
support has been instrumental in realizing this second edition. I am also grateful to
Nate McFadden, Senior Development Editor at Elsevier Engineering & Computing
Textbooks, and Mohanapriyan Rajendran, Production Project Manager, for kind and
efficient assistance, as well as for timely converting this project from its draft form to
its refined print state.
In closing, I would like to acknowledge and thank the unwavering support of my
wife, Simona, who painstakingly reviewed, checked, and edited the manuscript for
the second edition. She definitely made this project possible. As always, my
thoughts and profound gratitude for everything they gave me go to my daughters
Diana and Ioana and to my parents Ana and Nicolae.
Resources That Accompany
This Book
System dynamics instructors and students will find additional resources at textbooks.
elsevier.com
AVAILABLE TO ALL
Additional Online Content linked to specific sections of the book, extra content
includes advanced topics, additional worked examples, and more.
Downloadable MATLAB Code for the book’s solved examples.
xxiii
CHAPTER
Introduction
1
This chapter discusses the notion of modeling or simulation of dynamic engineering
systems as a process that involves physical modeling of an actual (real) system,
mathematical modeling of the resulting physical representation (which generates
differential equations), and solution of the mathematical model, followed by inter-
pretation of the result (response). Modeling in this text uses lumped (or pointlike)
parameters and is placed in the context of either analysis or design. The dynamic
system mathematical model is studied in connection to its input and output signals,
such that single-input, single-output (SISO) and multiple-input, multiple-output
(MIMO) systems can be formed. Linear systems are categorized depending on the
order of the governing differential equations as zero-, first-, second-, or higher-
order systems. In addition to the applications usually encountered in system
dynamics texts, examples of compliant (or flexible) mechanisms that are incorpo-
rated in micro- or nanoelectromechanical systems (MEMS or NEMS) are included
here. The nature of presentation is mainly descriptive in this chapter, as it attempts to
introduce a few of the concepts that are covered in more detail in subsequent
chapters.
FIGURE 1.1
Flow in the Dynamic System Modeling Process.
Car motion
l1 l2
Original position
x θ
CG
m, J
k2
Front suspension k1 Car body
Rear suspension
Road surface
FIGURE 1.2
Simplified Physical Model of a Car That Moves Over Even Terrain.
are fixed points. Under these simplifying assumptions, the parameters that define the
car’s properties are its mass, its mechanical moment of inertia about an axis passing
through the CG and perpendicular to the drawing plane, and the spring stiffness of
the two suspensions. What is the minimum number of variables fully describing the
state (or configuration) of this simplified system at any moment in time? If we attach
the system motion to the CG, it follows that the vertical motion of the CG (measured
by the variable y) and the rotation (pitch) of the rigid rod (which symbolizes the car
body) about a horizontal axis and measured by an angle q are sufficient to specify the
position of the car body at any time moment. Of course, we have used another
simplifying assumption that the rotations and vertical displacements are relatively
small, and therefore the motions of the suspensions at their joining points with
the car body (modeled here as a rigid rod) are purely vertical.
As a consequence, the system parameters are the car mass m and its moment of
inertia J, the suspension spring constants (stiffnesses) k1 and k2, as well as the dis-
tances l1 and l2, which position the CG of the car. Generally, all these parameters
have known values. The variables (unknowns or DOF) are y, the vertical motion of
the CG, and q, the rotation of the body car about its CG. The next step is deriving
the mathematical model corresponding to the identified physical model, and this
phase can be achieved using a specific modeling technique, such as Newton’s sec-
ond law of motion, the energy method, Lagrange’s equations, or the state-space
representation for this mechanical systemdall these modeling techniques are dis-
cussed in subsequent chapters. The result consists of two linear ODE containing
the system parameters m, J, k1, k2, l1, l2, and the unknowns y, q, together with their
time derivatives. Solving for y and q in terms of initial conditions (for this system,
these are the initial displacements when t ¼ 0, namely y(0), q(0), and the initial
_
velocities yð0Þ; _
qð0Þ) provides explicitly the functions y(t) and q(t), and this consti-
tutes the system’s response. The system behavior can be studied by plotting y and q
as functions of t.
1.2 Modeling Engineering System Dynamics 5
l1 l2
y θ
CG
ks2 cs2 y
m, J 2
y1 ks1 cs1 m2
m1 kt ct
kt u2
ct
u1
FIGURE 1.3
Simplified Physical Model of a Car Moving Over Uneven Terrain, With the Degrees of
Freedom of the Suspensions Shown.
More complexity can be added to the simple car physical model of Figure 1.2, for
instance, by considering that the wheels are separate from the mechanical suspen-
sion through the tire elasticity and damping. The assumption of an uneven terrain
surface can also be introduced. Figure 1.3 depicts the physical model of the car
when all these system properties are taken into accountdplease note that the masses
of wheels, tires, and suspensions are included and combined together (they are
denoted by m1 and m2 in Figure 1.3), and when the two wheels are considered iden-
tical. It can now be seen that two more DOF are added to the existing ones, so that
the system becomes a four-DOF system (they are y, q, y1, and y2), whereas the input
is formed by the two displacements applied to the front and rear tires, u1 and u2.
Dynamic modeling is involved in two apparently opposite directions: the analysis
and the design (or synthesis) of a specific system. Analysis starts from a given system
whose parameters are known. The dynamic analysis objective is to establish the
response of a system through its mathematical model. Conversely, the design needs
to find an actual dynamic system capable of producing a specified performance or
response. In analysis we start from a real-world, well-defined system, which we
attempt to characterize through a mathematical model, whereas in design (synthesis)
we embark with a set of requirements and use a model to obtain the skeleton of an
actual system. Figure 1.4 gives a graphical representation of the two processes.
(a)
(b)
FIGURE 1.4
Processes Utilizing Dynamic Models: (a) Analysis; (b) Design (Synthesis).
6 CHAPTER 1 Introduction
y y
meq
x
kt
(a) (b)
FIGURE 1.5
Cantilever Beam: (a) Actual, Distributed-Parameter Inertia and Stiffness; (b) Equivalent,
Lumped-Parameter Inertia and Stiffness.
Caution should be exercised when studying complex flexible systems, where the
lumping of parameters can yield results that are sensibly different from the expected
and actual results, as measured experimentally or simulated by more advanced (nu-
merical) techniques, such as the finite element method. However, for the relatively
simple compliant device configurations analyzed in subsequent chapters, lumped-
parameter modeling yields results with relatively small errors.
Modeling Methods
Several procedures or methods are available for deriving the mathematical model of
a specified lumped-parameter physical model. Some of them are specific to a certain
1.3 Elements, System, Input, and Output 7
system (such as the Newton’s second law of motion, which is applied to mechanical
systems or the Kirchhoff’s laws, which are used in electrical systems). Others can be
utilized more across the board for all dynamic systems, such as the energy method,
the Lagrange’s equations, the transfer function method, and the state-space
approach. These methods are detailed in subsequent chapters.
Solution Methods
Once the mathematical model of a dynamic system has been derived, and consists of
one or more differential equations (linear or nonlinear), the solution can be obtained
mainly using two methods. One method is the direct integration of the differential
equations, and the other method uses the direct and inverse Laplace transforms.
The big advantage of the Laplace method, as will be shown in Chapter 6, consists
in the fact that the original, time-defined differential equations are transformed into
algebraic equations, whose solution can be found by simpler means. The Laplace-
domain solutions are subsequently converted back into the time-domain solutions
by means of the inverse Laplace transform. The transfer function and the state-
space methods are also employed to determine the time response in Chapters 7 and
8, respectively.
System Response
The majority of the systems studied in this text are linear and are modeled with
constant-coefficient ODE as initial-value problems. The solution to a linear ODE
that describes the system behavior is the sum of two parts: one is the complementary
(or homogeneous) solution (which is the solution when no input or excitation is
applied to the system), and the other is the particular solution (which is one solution
of the equation when a specific forcing or input acts on the system). The complemen-
tary solution is representative of the free response, and usually vanishes after a
period of time with dissipation present, so it is indicative of the transient response.
The particular solution, on the other hand, persists in the overall solution, and there-
fore defines the forced or steady-state response of the system to a specific input.
+ + i
R L C v v L
– –
C
(a) (b)
FIGURE 1.6
(a) Individual Electrical Elements; (b) Electrical System Formed of These Components.
y1 y2
k1
k2
m1 m2
f2
c f1
FIGURE 1.7
MIMO Mechanical System With Linear Motion.
u u u
0, t < 0
A u=
0, t < 0 0, t < 0 A ⋅ t 2, t ≥ 0
u= u=
A, t ≥ 0 A ⋅ t, t ≥ 0
t t t
(a) (b) (c)
u u u
0, t < 0
A ∞ u=
A A A ⋅ sin ( ω ⋅ t ) , t ≥ 0
τ , 0≤ t ≤ τ lim
u= τ u = τ→0 τ A
0, t < 0, t > τ 0, t < 0, t > τ
τ t 0←τ t t
reduced to lumped-parameter (most often) linear systems that are similar to other
well-established system dynamics examples.
Compliant (flexible) mechanisms are devices that use the elastic deformation of
slender, springlike portions, instead of classical rotation or sliding pairs to create,
transmit, or sense mechanical motion. The example of Figure 1.9 illustrates the rela-
tionship between a classical translation (sliding) joint with regular springs and the
corresponding compliant joint formed of flexure hinges (slender portions that
bend and enable motion transmission). The compliant device of Figure 1.9(a) is
Fixed support
Flexure hinges
y
k k
Body Motion direction m
k k
Flexure hinges
Fixed support
(a) (b)
FIGURE 1.9
Realizing Translation: (a) Compliant Mechanism With Flexure Hinges; (b) Equivalent
Lumped-Parameter Model.
10 CHAPTER 1 Introduction
Input Input
Amplified output
Piezoelectric actuator
(a) (b)
FIGURE 1.10
Flexure-Based Planar Compliant Mechanism for Motion Amplification: (a) Photograph of
Actual Device; (b) Schematic Representation With Pointlike Rotation Joints.
constrained to move horizontally because the four identical flexure hinges bend
identically (in pairs of two), whenever a mechanical excitation is applied along
the motion direction. The lumped-parameter counterpart is drawn in
Figure 1.9(b), where the four identical flexure hinges have been substituted by
four identical translation springs, each of stiffness k.
Another compliant mechanism example is the one of Figure 1.10(a), which pic-
tures a piezoelectrically actuated, displacement-amplification device. Figure 1.10(b)
is the schematic representation of the actual mechanism, where the flexure hinges
are replaced by classical pointlike rotation joints. The schematic shows that the input
from the two piezoelectric (PZT) actuators is amplified twice by means of two lever
stages. The mechanism is clamped to and offset above the base centrally, as indi-
cated in Figure 1.10(a), and is free to deform and move in a plane parallel to the
base plane.
As monolithic (single-piece) devices, compliant mechanisms present several ad-
vantages over their classical counterparts, such as lack of assembly, no moving parts,
and therefore no losses due to friction between adjacent parts, no need for mainte-
nance, and simplicity of fabrication (although at costs that are higher generally
compared to classical manufacturing procedures). Their main drawback is that the
range of motion is reduced because of the constraints posed by limited deformations
of their compliant joints. Compliant mechanisms are encountered in both macro-
scale applications (with dimensions larger than millimeters) and micro- or nanoscale
ones (when the device dimensions are in the micrometer or nanometer range
(1 mm ¼ 106 m, 1 nm ¼ 109 m)), particularly in microelectromechanical systems.
In many situations, compliant mechanisms are built as single-piece (monolithic) de-
vices with techniques such as wire electro-discharge machining (wire EDM),
through water jet machining or by microfabrication techniques (for MEMS), such
as surface or bulk micromachining.
1.4 Compliant Mechanisms and Microelectromechanical Systems 11
Electrostatic sensing
Thermal actuation
Input motion
Fixed armature
FIGURE 1.11
Top View of Compliant MEMS With Thermal Actuation and Electrostatic Sensing of
Motion.
Fixed support
Torsion hinge kr
Fixed plate
A
Mobile plate Mobile plate Jt
kr
Force
the name torsional mirror. Figure 1.12(c) shows the equivalent lumped-parameter
model of the actual torsional mirror. Several other MEMS devices are analyzed in
subsequent chapters as mechanical, electrical, or coupled-field systems.
where ai (i ¼ 0 to n) and b are constant factors, and the input function u(t) can also
include derivatives. The maximum derivation order of the output function y(t) in a
system of the nth order is n, such that a second-order system, for instance, is defined
by a maximum-order input derivative of 2, and so on.
y
k
f
k
Massless body
FIGURE 1.13
Massless Body With Springs.
1.5 System Order 13
Thermometer
θ
Bath
θb
FIGURE 1.14
Bath-Thermometer Thermal System.
As detailed in Chapter 5, the heat flow rate stored by the thermometer during
contact is
dqðtÞ
q ¼ m$c$ (1.5)
dt
where c is the specific heat of the thermometer, m is its mass, and q is the thermom-
eter temperature. At the same time, it is known that convective heat transfer between
the bath and thermometer is governed by the equation
q ¼ h $ A $ ½qb qðtÞ (1.6)
where h is the convection heat transfer (film) coefficient, and A is the thermometer
area in contact with the fluid. Equating the heat flow rates of Eqs. (1.5) and (1.6) re-
sults in
m $ c dqðtÞ
$ þ qðtÞ ¼ qb (1.7)
h $ A dt
which indicates that s ¼ m$c/(h$A) and K ¼ 1. The input is the bath temperature qb,
and the output is the thermometer temperature q(t). Figure 1.15 displays a typical
first-order system response for the particular case where s ¼ 10 s and qb ¼ 80 C.
For a first-order system, the response to a step input can be characterized by the
steady-state response, y(N), the rise time (time after which the response gets to 90%
14 CHAPTER 1 Introduction
FIGURE 1.15
Thermometer Temperature as a Function of Time (Typical First-Order System Time
Response).
of the steady-state response, but other definitions are also applicable), and the
settling time (time necessary for the response to stay within 2% of the steady-
state response values); more details on this topic are given in the website Chapter
13, which studies the dynamics of control systems.
y
k
m
f
c
FIGURE 1.16
Mechanical System With Mass, Spring, and Damper.
FIGURE 1.17
Mass Displacement as a Function of Time (Typical Second-Order System Time
Response).
16 CHAPTER 1 Introduction
Ra
Rotor armature
+ ia
va vb Jl
– cr
θ, ma
La Stator Load
FIGURE 1.18
Schematic of a DC Motor as an Electromechanical System.
response to reach its maximum value), the peak response (the maximum response),
and the settling time; all these parameters are studied in more detail in Chapter 13 in
the context of controls.
Systems of orders larger than two are also encountered in engineering applica-
tions, as in the following example, which results in a third-order system model.
The electromechanical system of Figure 1.18 consists of a dc (direct-current) motor
and load. The dynamic model of this rotary system consists of equations that
describe the mechanical, electrical, and mechanicaleelectrical (coupled-field)
behavior. Essentially, the electromechanical system sketched in Figure 1.18 is
formed of a mobile part (the rotor armature), which rotates under the action of a
magnetic field produced by the electrical circuit of an armature (the stator). The
electrical circuit is formed of a resistor Ra, an inductor La, and a voltage source
va. As discussed in Chapters 4 and 10, the electrical system’s behavior is governed
by Kirchhoff’s second law, according to which
dia ðtÞ
Ra $ ia ðtÞ þ La $ ¼ va ðtÞ vb ðtÞ (1.13)
dt
where the subscript a indicates the armature, and vb is the back electromotive force
(voltage).
The mechanical part of the system is formed of the rotary load inertia Jl and a
rotary damper cr; according to Newton’s second law of motion and as detailed in
Chapters 3 and 10, the system is governed by the equation
d2 qðtÞ dqðtÞ
Jl $
2
¼ ma ðtÞ cr $ (1.14)
dt dt
where ma is the torque developed due to the statorerotor interaction. It is also known
that the following equations couple the mechanical and electrical fields:
8
>
< ma ðtÞ ¼ Km $ ia ðtÞ
dqðtÞ (1.15)
>
: vb ðtÞ ¼ Ke $
dt
1.6 Coupled-Field (Multiple-Field) Systems 17
+ Actuation circuit
vi
–
Sensing circuit
FIGURE 1.19
Coupled-Field System With Mechanical, Electrical, and Piezoelectric Elements. PZT,
piezoelectric.
18 CHAPTER 1 Introduction
K
Quantity to be measured
FIGURE 1.20
Sensing Characteristics in a Linear Measurement Process.
the unknown function (response) y(t) and when the unknown function and its deriv-
atives in the left-hand side of Eq. (1.1) are first-degree polynomial functions. Sys-
tems where the coefficients ai are not constant (are time variable, for instance)
still preserve their linear character.
Consider a linear system whose output y(t) is formally expressed in terms of the
input u(t) as O(u) ¼ y where O is a linear operator. For an input c$u (c being a con-
stant factor), the resulting output is O(c$u) ¼ c$O(u) ¼ c$y; this property of linear
systems is known as scalability or homogeneity. Linear systems are also additive,
which means that when the input is a sum of multiple inputs, such as u1 þ u2, the
corresponding output becomes O(u1 þ u2) ¼ O(u1) þ O(u2) ¼ y1 þ y2. Combining
scaling and additivity results in O(c1$u1 þ c2$u2) ¼ O(c1$u1) þ O(c2$u2) ¼
c1$O(u1) þ c2$O(u2) ¼ c1$y1 þ c2$y2, where c1 and c2 are constant factors. This
linear superposition feature allows calculating the output of a dynamic system to
a linear combination of multiple inputs, which is c1$u1 þ c2$u2 þ ., by superim-
posing the system outputs to individual inputs, namely c1$y1 þ c2$y2 þ ..
An important consequence of linearity of systems and of their mathematical
models is that of enabling direct and inverse Laplace transformations. For instance,
Laplace transforming the linear Eq. (1.1) results in Y(s) ¼ G(s)$U(s), where Y(s) and
U(s) are the Laplace transforms of the output y(t) and input u(t), while G(s)dthe
transfer functiondplays the role of a gain function. The output y(t) can simply be
found by inverse Laplace transforming Y(s).
In a SISO mechanical system, nonlinearity can be produced by several factors
connected to either mass, stiffness, damping, or motion features. Consider the
body of mass m of Figure 1.21, which is attached by two identical springs of stiffness
k and of undeformed length l.
When the body moved a distance x to the right from the equilibrium position, the
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
elongation of each of the two identical springs is equal to l2 þ y2 l. By applying
k
l
m
l
k
y
FIGURE 1.21
Mass With Two Springs in Deformed Position as a Nonlinear Mechanical System.
20 CHAPTER 1 Introduction
Newton’s second law of motion and projecting the two identical spring forces on the
horizontal motion direction, the following equation results:
d2 yðtÞ
þ 2keq $ yðtÞ ¼ 0
m$ (1.17)
dt2
with the equivalent stiffness being
0 1
B l C
keq ¼ k $ @1 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiA (1.18)
2
l2 þ yðtÞ
which indicates the stiffness is nonlinear, and therefore the whole mechanical system
is nonlinear. The stiffness increase is produced by a “hardening” effect, whereby the
stiffness and its slope increase with y. Another variant, where the slope decreases
with y, is generated by a “softening” effect. Both behaviors are nonlineardthey
are sketched in Figure 1.22, alongside the characteristic of a linear spring.
Hardening spring
Spring force
Linear spring
Softening spring
Deformation
FIGURE 1.22
Linear, Hardening, and Softening Spring Characteristics.
Voltage
Current
– i lin i lin
FIGURE 1.23
Saturation-Type Nonlinearity.
1.9 Feedback Control of Dynamic Systems 21
current increases past the linearity limits. As known from electromagnetism, the
voltage across an inductor is defined as vL ðtÞ ¼ L $ diðtÞ=dt. Obviously, when the
current is confined within the ilin and þilin bounds, the slope di/dt is constant,
and the voltageecurrent relationship is linear. This actually is the range utilized
in the majority of electrical system calculations.
Feedback
(a) (b)
FIGURE 1.24
(a) Uncontrolled Dynamic System; (b) Feedback-Control Dynamic System.
Mechanical Elements
2
OBJECTIVES
In this chapter you will learn about:
• Springs, heavy bodies, and viscous dampers, which are the main mechanical
elements of systems undergoing translation or/and rotation.
• Lumped-parameter properties of and energy associated with translatory and rotary
mechanical elements: stiffness, inertia, and damping.
• Transforming distributed-parameter inertia properties of elastic members such as
bars and beams into equivalent lumped-parameter properties.
• Transforming lumped-parameter mechanical properties by series/parallel connec-
tion, gears, and levers.
INTRODUCTION
This chapter introduces the basic mechanical elements that are used as building blocks
for translatory and/or rotary mechanical systems. These elements are the springs, which
describe elastic properties through stiffness (or spring constants), the heavy solids,
which carry inertial or mass properties, and the viscous dampers, which qualify a widely
used type of energy losses through damping coefficients. The stiffness, mass (or mass
moment of inertia), and damping coefficient are formulated as lumped-parameter (or
pointlike) element properties in terms of the related forces/moments, as well as in terms
of the associated energy. Flexible, distributed-parameter stiffness, and inertia of mem-
bers such as bars and beams are rendered into equivalent, lumped-parameter elements.
Examples are also offered of transferring basic lumped-parameter mechanical element
properties into equivalent ones through motion transformers such as levers and gears.
Translation Rotation
x1 x2 θ1 θ2
kt kr
FIGURE 2.1
Helical Spring With Symbols for Translatory and Rotary Motion.
Free end
fy, uy
E, Iz
kt ¼ 3El3$ Iz
l l
fy, uy
Fixed-guided beam
kt
f y, u y
Gudied end
Fixed end
E, I z
kt ¼ 12El3 $ Iz
l
Fixed end
fy, uy
E, I z
kt kt ¼ 192E
l3
$ Iz
l/2
l l
Fixed-free bar
Fixed end Free end
E, A
fx, ux
kt ¼ E $l A
l
kt fx, ux
Fixed-free helical spring
G, I t l
Free end
Fixde end
4
kt ¼ 8N
G$d
$ D3
l fx, ux
26 CHAPTER 2 Mechanical Elements
Free end
mz, θz
E, Iz
l
4
kr ¼ p E$ N$ I$z D ¼ 64N
E $d
$D
kr mx, θx
mx, θx
l
l
Fixed-free bar in torsion
mx, θx
Fixed end
Free end
G, I p G $ Ip
kr ¼ l
I
z-axis (axis perpendicular on the xy plane of drawings), and Ip is the polar moment of
area of the circular cross section.
Note that the rotary stiffness of a helical spring is identical to that of a spiral
spring in Table 2.3 because the total wire length of a helical spring is lt ¼ N$p$D.
x1 x2 x x
k1
k1 k2
f f
k1 k2 f
k2
(a) (b) (c)
FIGURE 2.2
Translatory Spring Combinations: (a) Series; (b) and (c) Parallel.
identical, whereas the sum of individual spring forces equals the force f. The spring
combination of Figure 2.2(c) is a parallel one with respect to the reference x because
the two springs undergo the same deformation at that point. The equivalent series
stiffness ks and the parallel stiffness kp corresponding to n spring elements of stiff-
nesses k1, k2, ., kn are derived in the companion website Chapter 2 as:
1 1 1 1
¼ þ þ / þ ; kp ¼ k1 þ k2 þ / þ kn (2.1)
ks k1 k2 kn
nn
Example 2.1
Four identical translatory helical springs are combined in two arrangements, and in each of the two
combinations there are both series and parallel connections. When the same force is applied sepa-
rately to each spring arrangement at the free end (the other one being fixed), the free-end displace-
ment of one arrangement is maximum, whereas the free-end displacement of the other arrangement
is minimum. Identify the two spring configurations, determine the ratio of their free-end displace-
ments, and calculate the equivalent stiffness for each when d ¼ 1 mm, D ¼ 12 mm, N ¼ 10, and
G ¼ 160 GPa.
Solution
The largest displacement is obtained when all four springs are coupled in series because the stiffness
is minimum, namely ks ¼ k/4dsee Eq. (2.1). Conversely, the smallest displacement corresponds to a
full parallel spring connection when the stiffness is maximum, kp ¼ 4k, as also indicated in Eq. (2.1).
However, these connections are not allowed in this example. Two spring configurations (1 and 2) are
sketched in the top row of Figure 2.3, which are candidates satisfying this example’s requirement to
mix series and parallel pairs. Configuration 1 results in the maximum stiffness among all possible
designs of four springs, with at least two connected in series. Similarly, Configuration 2 yields the
minimum stiffness of all four-spring variants with at least two springs coupled in parallel.
The two springs of the top branch in Figure 2.3(a) are combined in series and the two springs in
the portion at the free end at the top in Figure 2.3(b) are connected in parallel; as a consequence, the
original designs on the top row of Figure 2.3 become the equivalent configurations shown in the bot-
tom row of the same Figure.
The equivalent stiffnesses of the spring connections shown in Figure 2.3 are:
8
>
> k 5
>
< k1 ¼ 2 þ k þ k ¼ 2 $ k
(2.2)
>
> 1 1 1 1 5 2
>
: ¼ þ þ ¼ or k2 ¼ $ k
k2 k k 2k 2k 5
28 CHAPTER 2 Mechanical Elements
Configuration 1 Configuration 2
x1
k k x2
k
k f k k f
k
k
x1
k/2 x2
k f k k 2k f
k
(a) (b)
FIGURE 2.3
Translatory Spring Combinations With: (a) Three Parallel Branches; (b) Three Series
Portions.
where the subscripts 1 and 2 identify configurations 1 and 2, respectively. Because f ¼ k1$x1 ¼ k2$x2,
Eq. (2.2) yield:
x2 k1 ð5kÞ=2 25
¼ ¼ ¼ (2.3)
x1 k2 ð2kÞ=5 4
By using the equation of a translatory helical spring given in Table 2.2 and the specified numerical
values, the following results are obtained: k ¼ 1157.4 N/m, k1 ¼ 5k/2 ¼ 2893.5 N/m, and
k2 ¼ 2k/5 ¼ 462.96 N/m. nn
nn
Example 2.2
The symmetric shaft-disk system of Figure 2.4(a) consists of identical and mirrored portions, each
formed of two shaft segments that are anchored at their distant ends and connected to the middle
rigid gear at their opposite ends. The fixed-end shafts have a diameter d1 ¼ 0.004 m and a length
l1 ¼ 0.1 m, whereas the other shafts have a diameter d2 ¼ 0.005 m and a length l2 ¼ 0.14 m. The
shear modulus is G ¼ 80 GPa.
(i) Calculate the torsion moment mx that has to be applied at O to produce a maximum rotation angle
qx,max ¼ 5 degrees at the same point.
(ii) Calculate the number of active turns N of a rotary helical spring with fixed-free ends to replace all
the actual shafts of Figure 2.4(a). The single spring produces the angle qx,max of (i) under the ac-
tion of the same moment mx. The spring is formed of a wire with diameter d ¼ 0.005 m, Young’s
modulus E ¼ 210 GPa, and mean diameter cannot exceed Dmax ¼ 0.03 m.
Solution
(i) As illustrated in Figure 2.4(b), the two shaft segments to the left of point O are serially connected,
and the same applies for the two mirrored and identical shafts to the right of O; the stiffness of
2.1 Spring Elements 29
θx, mx
Anchor
d2 d2 d θx, mx keq θx, mx
d1 k1 k1
Anchor
O x x x
k2 k2
Gear O O
l1 l2 l2 l1
each of these two segments is calculated with respect to O based on Eq. (2.1) as:
ks ¼ k1 $ k2 =ðk1 þ k2 Þ. Because these two serial portions are jointed at O, they undergo the
same rotation and therefore are coupled in parallel with respect to O. A single-spring model is
shown in Figure 2.4(c) and the equivalent stiffness with respect to O is:
k1 $ k2
keq ¼ 2ks ¼ 2 $ (2.4)
k1 þ k2
Based on Table 2.3 (the particular design of a fixed-free bar in torsion), the rotary stiffnesses of the
two flexible shaft segments are expressed as k1 ¼ G $ Ip1 l1 ; k2 ¼ G $ Ip2 l2 where the polar
area moments of the two bars are (see Appendix F): Ip1 ¼ pd14 32; Ip2 ¼ pd24 32. As a
consequence, the equivalent spring stiffness of Eq. (2.4) becomes:
As per Table 2.1, the maximum torque to be applied at the midpoint O is maximum because the
rotation angle at that point is maximum, namely:
E $ d4 E $ d4
N¼ or Nmin ¼ (2.7)
64D $ keq 64Dmax $ keq
With D ¼ Dmax and the numerical value of keq obtained at (i), the minimum number of active turns
is Nmin ¼ 2.675. The first integer larger than Nmin is 3; as a consequence, N ¼ 3 is a value that
satisfies the requirements. nn
30 CHAPTER 2 Mechanical Elements
nn
Example 2.3
The microscale mechanism of Figure 2.5(a) is formed of a serpentine springdwhose details are
provided in Figure 2.5(b)dand a bridge (a beam that is fixed at both ends). The two components
are initially separated by a gap Dy.
(i) Find the minimum force fmin acting at A along the y axis, which needs to be applied to the serpen-
tine spring to close the gap Dy to the midpoint B of the bridge.
(ii) Calculate the deflection at the bridge midpoint when a force f1 ¼ 3$fmin is applied at A on the
serpentine spring along the y direction with the device in its original (detached) position. All flexible
segments are of identical circular cross section with diameter d and of identical material proper-
ties. Known are: d ¼ 2 mm, l ¼ 100 mm, Young’s modulus E ¼ 160 GPa, and Dy ¼ 5 mm.
2l y
y
l fmin y
l
Rigid connector A
B ks
Long beam
Δy f Bridge
kl
A ks
Short beam
Serpentine spring O
l
Anchor
Solution
(i) The three beams of the serpentine unit of Figure 2.5(b) act as springs in series and the lumped-
parameter model of the serpentine spring alone is that of Figure 2.5(c). The stiffness of the
serpentine spring is:
1 1 1 1 2 1
¼ þ þ ¼ þ (2.8)
kse ks kl ks ks kl
where ks and kl are the stiffnesses of the short and long beams, respectively. It should be noted
that in terms of their end-point deformations, the three beams are fixed-guided, and therefore,
their stiffnesses are calculated based on Table 2.2 and Appendix F as:
E $ Iz E p $ d 4 3p $ E $ d 4 E $ Iz
ks ¼ 12 $ ¼ 12 $ $ ¼ ; kl ¼ 12 $
l 3 l 3 64 16l3
ð2lÞ3
E p $ d 4 ks
¼ 12 $ 3
$ ¼ (2.9)
ð2lÞ 64 8
By combining Eqs. (2.8) and (2.9), the stiffness of the serpentine spring becomes:
1 2 8 10 ks 6E $ Iz 3p $ E $ d4
¼ þ ¼ or kse ¼ ¼ ¼ (2.10)
kse ks ks ks 10 5l3 160l3
The force fmin to be applied at the free end of the serpentine spring (as shown in Figure 2.5(a)) to
close the gap Dy is fmin ¼ kse $ Dy . The numerical values of the serpentine spring stiffness and
minimum force are kse¼ 0.151 N/m and fmin ¼ 0.754 mN.
2.2 Inertia Elements 31
f1 kb y N
f1
Applied at A A B
kse M
fmin P
O Δy Δy1 Δy
(a) (b)
FIGURE 2.6
(a) Lumped-Parameter Model of Coupled Serpentine Spring and Bridge; (b) Spring
Characteristics Before and After Contact.
(ii) Once point A on the serpentine has been moved to coincide with the midpoint B of the bridge, the
resulting spring system is shown in Figure 2.6(a).
From this stage on, the force f1 works against both the serpentine spring and the bridge (because
the two springs are coupled in parallel with respect to point A h B) and therefore the resulting
equivalent stiffness is:
6E $ Iz 192E $ Iz 126E $ Iz
keq ¼ kse þ kb ¼ þ ¼ ¼ 21 $ kse (2.11)
5l3 ð2lÞ3 5l3
where kb is the bending stiffness of the bridge with respect to its midpoint and is provided in
Table 2.2. The two spring characteristics, corresponding to the serpentine alone (deformation
between 0 and Dy) and then to the coupled serpentine-bridge springs (deformation between
Dy and Dy1), are illustrated in Figure 2.6(b). The total deformation Dy1 is calculated as:
fmin f1 fmin fmin 2fmin 1 2
Dy1 ¼ Dy þ kMPk ¼ þ ¼ þ ¼ fmin $ þ
kse keq kse keq kse kse þ kb
2kse 23
¼ 1þ $ Dy ¼ $ Dy
kse þ kb 21
(2.12)
which is Dy1 ¼ 5.476 mm. Eq. (2.12) took into account that the extra deformation equal to the
length of segment MP is produced by a force equal to f1 fmin ¼ 3fmin fmin ¼ 2fmin. nn
Bearings
Central axis
l
x θ d
θ
Δ
dm r Offset parallel axis
m m J
Mass moment JO ¼ m $ l 2 JO ¼ 12
1 $ m $ l2 Jx ¼ 12 $ m $ R2 Jx ¼ 12
1 $ m $ w 2 þ h2
of inertia
2.2 Inertia Elements 33
Table 2.5 Inertia Force/Moment and Kinetic Energy for Translatory and
Rotary Motion
Translation Rotation
Inertia force/moment fi ¼ m $ a ¼ m $ x€ €
mi ¼ J $ a ¼ J $ q
Kinetic energy T ¼ 12 $ m $ v2 ¼ 12 $ m $ x_2 T ¼ 12 $ J $ u2 ¼ 12 $ J $ q_
2
For translation, the mass is included in the inertia force, which is proportional to
the acceleration (the second time derivative of displacement) and is calculated as
shown in Table 2.5 where fi has the same direction as x in Figure 2.7(a).
Similarly, an inertia moment (or inertia torque) is defined in rotary motion as a
function of the mass moment of inertia and the angular acceleration (the second time
derivative of rotation angle)dsee Table 2.5 where mi has the same direction as q in
Figure 2.7(b) and 2.7(c). Bodies in motion possess kinetic energy, denoted by T,
which is expressed for translation and rotation as a function of the corresponding
velocity in the same Table 2.5.
The mass moment of inertia is an additive/subtractive amount, and this property
is used in calculating the mass moment of inertia of compound bodies formed of
several elementary bodies whose moments of inertia are readily available.
nn
Example 2.4
(i) Calculate the mass moment of inertia about the centroidal (symmetry) axis of the solid right cir-
cular cone frustum shown in Figure 2.8(a) in side view and in a cross-section. The frustum is
defined by d1, d2, and l. The material is homogeneous with a mass density r. Use the obtained
result to also calculate the mass moment of inertia of a cylinder, both about its centroidal axis and
about a parallel axis that is offset at a distance d ¼ d2 from the centroidal axis.
l
A Section A– A
d1 dx d2 r d22 d21 d3 d1
rx
dr
x
dx l2 l1
A
(a) (b)
FIGURE 2.8
(a) Frustum of Right Circular Conical Solid in Side View and Cross-Section; (b) Simplified
Longitudinal Section of a Bevel Gear.
34 CHAPTER 2 Mechanical Elements
(ii) Calculate the mass moment of inertia of the simplified bevel gear of Figure 2.8(b) with respect to
its longitudinal axis in terms of the given dimensions. Numerical application: d1 ¼ 0.015 m,
d21 ¼ 0.016 m, d22 ¼ 0.024 m, d3 ¼ 0.008 m, l1 ¼ 0.01 m, l2 ¼ 0.015 m, and r ¼ 7800 kg/m3.
Solution
(i) For a homogeneous cone frustum of mass density r, such as the one of Figure 2.8(a), the
mechanical moment of inertia is expressed as:
0 1
Z Z h Z
J ¼ r$ r 2 dV ¼ r $ @ r 2 dAAdx (2.15)
V 0 A
As shown in Figure 2.8(a), the area of an elementary circular strip of width dr and inner radius r is
dA ¼ 2prdr. By substituting dA into Eq. (2.15), the mass moment of inertia of the cone frustum
becomes:
0 1
Z Z Z
l
@
rx
pr l
J ¼ 2pr $ r 3 dr Adx ¼ $ rx4 dx (2.16)
0 0 2 0
which returns:
pr $ l 4
J¼ $ d1 þ d13 $ d2 þ d12 $ d22 þ d1 $ d23 þ d24 (2.18)
160
When d1 ¼ d2 ¼ d, the cone frustum becomes a cylinder and Eq. (2.18) simplifies to:
2
pr $ l 4 1 d4 1 pd2 d 1 1
J¼ $ 5d ¼ $ pr $ l $ ¼ $ l $ r $ ¼ $ m $ d2 ¼ $ m $ r2
160 2 16 2 4 4 8 2
(2.19)
with m being the mass of the cylinder. The cylinder’s mass moment of inertia about an axis
situated at d ¼ d2 from its centroidal axis is found from Eq. (2.19) by means of the
parallel-axis theorem, Eq. (2.14), as:
1 9
JD ¼ J þ m $ d22 ¼ $ m $ d2 þ m $ d2 ¼ $ m $ d2 (2.20)
8 8
2.2 Inertia Elements 35
(ii) The bevel gear sketched in Figure 2.8(b) is geometrically formed by adding a cylinder to a conical
frustum and subtracting another cylinder. Similarly, the mass moment of inertia of the gear shown
in Figure 2.8(b) is calculated by algebraic addition of the moments of inertia defining the three
abovementioned elements, namely:
J ¼ J1 þ J2 J3
1 d12 pr $ l2 4
¼ $ r p $ $ l1 $ d12 þ d21 þ d21
3
$ d22 þ d21
2 2
$ d22 þ d21 $ d22
3
þ d22
4
8 4 160
1 d2
$ r p $ 3 $ ðl1 þ l2 Þ $ d32
8 4
(2.21)
and its numerical value is J ¼ 2.29$106 kg m2. nn
nn
Example 2.5
The mechanical system of Figure 2.9(a) uses a piezoelectric actuator in the form of a fixed-free, cir-
cularecross-section flexible bar (diameter is d1 ¼ 0.008 m) of length l1 ¼ 0.03 m and a pointlike ball
of mass m3 ¼ 0.005 kg at its tip. By linear expansion, the bar actuator pushes the midpoint of a bridge
with rectangular cross section (dimensions are h ¼ 0.001 m and w ¼ 0.014 m) and of length
l2 ¼ 0.04 m. Find the lumped-parameter mass meq that is inertially equivalent to the actual system
and that is placed at A when considering the x-axis translatory motiondsee Figure 2.9(b). Determine
the error in this equivalent mass when ignoring the inertia contribution from the bridge. Known are also
the mass densities of the bar r1 ¼ 7500 kg/m3 and bridge r2 ¼ 7800 kg/m3.
Another random document with
no related content on Scribd:
Indian’s own game, they knew, a tremendously difficult one for mere
whites to succeed at undetected. To make a big detour and come
down by way of the sheep corral would be fatal, for the dogs there
would surely give a warning. Niltci would undoubtedly be guarded. If
they overpowered the guards, it would be known who had kidnaped
the prisoner. That would never do!
“As I see it, fellows, the only stunt for us is for one of us to entice
Niltci’s guards away for a time, while the other two get to him and cut
him loose,” said Sid in low tones as they crept through the sage.
“Anyone got any notion of what might attract a Navaho’s curiosity?”
he inquired.
There was a silence as the others shook their heads. Then Big John
spoke up. “I got a better layout than that, Sid. These Injuns are
superstitious, an’ thar’s one thing they’re scairt of wuss than anythin’
else. You-all got plenty of matches?” he asked suddenly.
The boys stopped and got out their emergency kits. Out of each they
produced a bundle of about fifty matches. “Ef these was the good old
brimstone sticks they’d go better,” said Big John enigmatically, taking
them in his big paw, “but I’ll make out with them, all right. You boys
run along, now, an’ git as near Niltci’s tree as ye kin. Then lay low
and watch my smoke.”
He disappeared into the sage without a further word. Sid and Scotty
looked at each other, puzzled, but they knew Big John had a mighty
good head on him, and he would be unlikely to try anything that
would not work. As they drew near Neyani’s hogan they redoubled
their caution, crawling through the sage and taking advantage of
every little ridge of rock. At length they were quite near Niltci’s tree
and well concealed in a patch of scrub bear grass. They could make
out the Indian boy’s blurred form, now, bound upright to the tree.
Beside him on the ground squatted two Navaho guards, silent,
motionless. They might be dead, for all the movement either made;
but that they were quite awake and on watch the boys could not
doubt.
For a long time nothing happened. All the desert was still as death,
shrouded in the gloom of a faint mist that hung over the barren
ground. Overhead the stars swung in their great courses, but their
light penetrated but feebly through the dust and mist and haze of
smoke that drifted over from the ashes of the great fire.
Then—“Hist!” whispered Scotty, at length, gripping Sid’s arm.
“What’s that?”
There, quite near,—over by a little nest of bowlders—something
moved! Dim, dark, hardly discernible, a round, black head peered
out, advancing fearsomely, with two phosphorescent disks of fire
glaring at them, balefully!—Then a wild, mortal shriek rent the night!
It was the Black Panther, himself!
The boys tugged feverishly at their revolvers, their hair standing on
end with fright. As for the two Navahos, one wild leap, one terror-
stricken grunt of dismay, and they were off like the wind, bolting out
into the sage. It was not until they were some distance away that
they raised terrified screams of fright, running like deer as they
yelled.
The panther crept nearer, and, just as Sid was about to fire, Big
John’s low voice called out to them reassuringly.
“Don’t shoot!—Up an’ git young Niltci, boys!” he gritted.—“How did
my fox-fire look?”
Sid went nearly hysterical with the reaction. “Was that you, John? It
was great! Simply great!” he chuckled, springing forward to wring the
cowman’s hand. Bubbling with delight, they all dashed for Niltci.
Swiftly his thongs were cut, and then, each grabbing an arm, they
started for the canyon.
“Run, fellers—I’m a good kitty, all right, but we ain’t got no time to
lose!” whispered Big John, hoarsely.—“Listen!”
Voices and yells came from hogans all about them. The night was
hideous with uproar—then it suddenly ceased and all was silent
again.
“Well, I’ll be derned!” grunted Big John, as they hustled along, “what
d’ye make of that?—I’m wise, boys!” he laughed. “Them brave bucks
has told the hogans that their friend, Mister Black Panther, has come
and took Niltci—an’ they’re all stickin’ close to home, scairt to death!
Hep, boys—hep!”
CHAPTER VIII
SILENT PINES AND YELLOW CRAGS
“I ’VE got to turn back, Sid,” gasped Colonel Colvin, as they halted at
the foot of the vast slope that topped the second rim of the canyon
like a house roof. “Climbing up out of here is a job for a young heart;
mine would need half-soling before we’d ever make the rim again!”
From safety holds on tough pinyons that overhung the precipice of
the second rim, they peered down at the chase far below. This
second rim was an ungodly wall, perhaps a thousand feet sheer, and
it no doubt cut a noticeable figure as viewed from El Tovar, where
tourists at that moment were raving in absurd sentimentalities over
the canyon. To the Colonel it meant a terminus, for him, of that
particular cougar chase, for to add its weight to the labors already in
store on the climb back would be foolhardy to one of his age. They
watched the tiny black dots weaving slowly across the lower slope,
that must have been the dogs in hot chase of the cougar, and after
them came two oval specks that were Scotty and Big John as
viewed perpendicularly from their height. Then the whole business
disappeared over a ledge and nothing but the baying of invisible
hounds came floating up from the far depths.
Colonel Colvin shook his head. “It’s their meat, Sid. But there’s room
enough in this country for two or three hunts to be going at the same
time. They’ve got the dogs, but we’ve got Niltci, who’s better than a
dog at forest hunting, I’ll warrant. We’ll climb back to the rim and
start something of our own back in the breaks.”
Sid felt that his place was with his father, anyway, and he did not
care much about being a tailender in a hunt that had already
distanced them.
“Shrewd guess of Big John’s that was, Dad,” he replied. “It cut off at
least half a mile for them. If the cat had gone north, along under the
first rim wall, they’d have been out of it instead of us. Let’s ride back
into the gulches and box canyons of the country behind the rim, and
see what we can see.”
They looked up, seeking a place to climb back. A thousand feet
above them towered the rim of rock, dented with columnar pinnacles,
crowned with dwarfed pines that they knew were themselves at least
a hundred feet high. The Colonel was winded and panting before
they had climbed for fifteen minutes. The crumbly soil slid down
underfoot; even zigzagging was slow and laborious toil,—not at all
eased by ledge after ledge of rock outcroppings that called for
hands, knees and feet to scramble up them. Niltci and Sid pulled and
pushed the Colonel up, but an hour of dizzy, sweating work had
gone by and all were spitting cotton before they stood at the base of
the rim rock precipice.
Five hundred feet sheer it rose above them. For comparison Sid
imagined that if he were looking up the Woolworth building to its very
top and if, at the same time, he were standing on a narrow shelf of
yellow and rotten rock with a slope three times as deep below,
ending in blue nothingness—he would have some of the sensations
that now overcame him as he looked up for some possible chimney
up which to climb and wondered how they were ever going to get his
father up it if they did find one. If he could only manage to stand off
from this thing a little, so as to get some idea of its surface, it would
be easy to find the break in the rim where they had come down.
Which way did it lie, north or south? They discussed it, finally
yielding to Niltci, who was sure that it lay north.
Along under the rim wall they crept. The narrow path was worn deep
with cougar tracks. It was a regular runway for them, for they lived
down here in the canyon and came over the rim at night to hunt in
the deer and wild horse country of the hinterland. At any point they
might come upon a cougar cave, here, and the Colonel, who was in
advance, never passed around a pinnacle base without stepping
warily, with his rifle poised for instant use.
“By George, Niltci, you’re wrong—we should have turned south!”
barked out Sid after perhaps half an hour of this gingerly progress.
“Look at these young Matterhorns coming up out of the canyon
below us! I never noticed them before!”
The party stopped to take bearings. Certainly the lookout was new
and unfamiliar. The canyon jutted out here in a great cape, and on its
slope Nature had dropped, casually, three or four red and yellow
mountains that rose below like pyramids. Anywhere else they would
be objects of wonder and bear grandiloquent names.
But Niltci shook his head vigorously and led on without a word. The
rim cliff ended abruptly a little further on in a huge tower of stone,
and, rounding it, they found themselves in a vast amphitheater, a
mile deep, and a mile across a valley of illimitable depths to its
opposite point. All around it the cliffs rose sheer. Surely they never
came down here! Niltci had to acknowledge that much, himself, but
instead of turning back to retrace their steps he grunted impetuously
and led them on, following the rim into that enormous basin.
“Aw, rats!—what’s the use, Niltci, you’re crazy!” exploded Sid, as
both he and the Colonel balked at going any further. For answer the
Indian boy pointed to a thin fissure that cleft the rim from top nearly
to bottom, up near the head of the basin. It was about half a mile
away. How Niltci could know that that crevice could be practicable for
ascent, Sid could not conjecture, but the red men were wise in the
ways of Nature, so he followed on, albeit incredulously. But he had
no idea what impassable obstacle might await them to the south if
they turned back. This, at least, looked possible!
Arrived at it, they peered up to where the last of it ended in a broken,
jagged path, showing where water had come down during the rains.
For fifty feet this rose up the cliff,—an absurd trail for anything but a
fly to attempt; then began the in-cutting of the fissure.
Niltci started up it, amid a howl of protest from Sid and the Colonel.
Like a creeping cat the Indian lad climbed steadily up until he had
reached the fissure, where he turned with a whoop of triumph.
“Wow! We can’t let him get away with that, Father!” chuckled Sid.
“Shut your eyes and climb! Forget mother, forget your insurance, and
all the rest of it—it’s the only way!—I’ll be right behind you.”
The Colonel groaned, whimsically, and started up. Niltci came down
again by some incredible feat of legerdemain—as they looked back
upon that climb afterwards—and gave him a strong lift over the worst
places, and so they all reached the bottom of the fissure. It was dark
and gloomy, and it curved around a bend above, so that they had no
idea how it was all going to turn out—most likely in some sheer wall,
thought Sid. But the only way to get over these things was to go
ahead and do them, so they climbed up into it. Part of the time Niltci
was straddling both walls of it; part of the climb crawling up vertical
ledges higher than his head. The curve mercifully hid from them the
frightful depths below, should anyone fall. It grew better, once around
it, cutting deeper and deeper into the rim wall and becoming less and
less vertical. Masses of pine roots fringed it overhead, and finally
their feet found a narrow bottom of yellow, crumbled rocks, which led
up in a steep slant to the forest above.
“Great work, Niltci! That’s mountaineering for you!” laughed the
Colonel as they dropped panting on the forest duff. “I suppose you
could visualize this whole water crack, having once seen the fissure
in the rim wall, eh?”
Niltci grunted happily. He had no idea what that speech was all
about, but evidently his adored Lord Colonel was pleased! Sid
rejoined them, a moment later, and all sat and looked ruefully at their
clothing. Their bleeding knees peeped through frayed and torn riding
breeches, their buckskin gloves were out at the fingers; Sid had a
scraped thigh, caked with blood; all the uppers of their cruiser
moccasins gapped open in rent seams. Niltci, in his light cottons and
buckskin leggings, seemed the least frazzled of the three, but his
bare toes poked out from thin moccasins worn through on the rocks.
It was half an hour before the Colonel sat up again.
“And now, where are we?” he queried, briskly. “We’ve got all the rest
of the day, so we’ll find the horses and go hang up a buck for camp
meat.”
They all rose and started off through the forest. A short walk through
the high pines that covered the plateau brought the blue haze
beyond, of the canyon again, and presently they came out on a rock
pinnacle that commanded the whole prospect below. A sheer fall
about a thousand feet lay below them. Beyond that smoky, purple
depths showed beyond over the second rim.
“The place where we came down must lie to the west of here, boys,”
declared the Colonel. “Big John and Scotty are somewhere down in
this valley—they’ll be all day getting back! We’ll start west for
another look-see.”
A second outlook from another point showed them the steep slope
down which they had first come. Up in the ravine at the head of it
would be the horses, for there they had first started the cougar. Soon
they were in it and had remounted.
“Sore and tattered, but still in the game!” ejaculated the Colonel as
he put spurs to the roan and led back up the ravine into the
hinterland.
Back in here they found the huge flanks of Buckskin all cut up with
rocky glades grown up with yellow pines, and gulches which led to
high, walled canyons, all leading out to a discharge into the Grand
Canyon, somewhere. Great pines grew heavily in the swales. It was
a wonderful, rich, plentiful game country! Again and again Niltci
grunted, to point down at deer tracks, wolf tracks, and the round
hoof-prints of wild horses, and there seemed to be a cougar after
every deer, judging from their frequent footprints!
“Hist!” called the Colonel, suddenly, stooping down to whip out his
.35 from its scabbard. The bushes shook, up in a densely grown
ravine that lost itself somewhere in the upper flanks of the mountain.
Gray shapes bounded across it, stiff-legged, flashing into sight
occasionally, to disappear as quickly.
The Colonel’s rifle barked, followed by Sid’s. One of the gray shapes
plunged, and there was a mad scramble in the timber. They were
mule deer, a whole drove of them! Sid fired at another, running
bewilderedly up a bush-strewn slope in full flight. Then all was still
again.
“I nailed mine,” said the Colonel. “We’ll wait a while. Once he lies
down he’ll never get up.—Lord, boys, there must have been thirty in