new modeling and vibration analysis of a rocking–mass ...€¦ · in vibrating beam gyroscope,...
TRANSCRIPT
Modeling and Vibration Analysis of a Rocking–mass
Gyroscope System
by
Masoud Ansari
A Thesis Submitted in Partial Fulfillment
of the Requirements for the Degree of
Master of Applied Science Mechanical Engineering
in
The Faculty of Engineering and Applied Science
Mechanical Engineering
University of Ontario Institute of Technology
April 2008
©Masoud Ansari, 2008
iii
ABSTRACT
Gyroscopes are one of the most widely used devices for measuring the
angle of rotation or the rate of change of angular rotation. In the last few
years, the advent of micromachining technology has made it possible to
fabricate high performance Micromachined Electro-Mechanical Systems
(MEMS) gyroscopes at a relatively low cost. Of the various types of MEMS
gyroscopes, vibrating beam type, are the most commonly used, but they have
a main drawback (cross-axis effect) which significantly affects their
measurement and results in much error. In view of this, the present work
focuses on a gyroscope, called rocking-mass gyroscope, which does not have
that drawback at all and can give a very precise measurement.
The rocking-mass gyroscope consists of an assembly of four cantilever
beams with a rigid mass attached to them in the middle subjected to base
rotations. Due to the gyroscopic effect, the beams undergo coupled flexural-
torsional vibrations. The main goal of the research is to develop an accurate
model of such a system and along this line a detailed mathematical modeling
of the gyroscope is developed for a macro-scaled system. The equations of
motion clearly show the presence of the gyroscopic couplings in all cantilever
beams. In order to analyze the effectiveness of this type of gyroscope a
computer simulation model in its most general form has been developed.
iv
Before deriving the governing equations of motion for a rocking-mass
gyroscope, a detailed mathematical model of a single beam gyroscope was
considered to investigate the cross-axis effect in this type of gyroscopes.
The characteristic equation of a rocking-mass gyroscope, using an
exact method, is derived, and the fundamental frequency of the vibration and
the corresponding mode shape are obtained. Finally, the time history diagram
of this system is presented to illustrate the dynamic response of the system.
Simulation results show that the bending vibrations induced in the second and
the forth beams are directly proportional to the magnitude of the base rotation.
Therefore, the gyroscope can be widely used as a device for measuring the
base rotation velocity.
Keywords
Rocking-mass gyroscope; Vibrating beam gyroscope; Extended
Hamilton’s principle; Frequency analysis; Mathematical modeling.
v
Dedication
To my father Jamshid and my mother Shahnaz who influenced my life
the most.
vi
Acknowledgments
I would like to sincerely and wholeheartedly thank Prof. E. Esmailzadeh for
his guidance and kindness throughout this work. His patience as an advisor,
boundless energy while teaching, promptness while reviewing all my writing,
and passion for doing research are to be commended and worth emulating.
This thesis would not have been possible without his support.
I would also like to thank my advisory committee members Dr. Dan Zhang
and Dr. Greg Lewis for giving their valuable time and for serving on my
thesis committee.
I am grateful to Dr. Nader Jalili of Clemson University for his constant
guidance and motivation during this research work.
A special thanks to Ms. Jane Dubis and Ms. Kathryn Hanson in the university
library for all their cooperation and help during my course of study at UOIT.
Most importantly, I would like to thank my parents, Jamshid and Shahnaz,
and my sisters Maryam and Marzieh, for their unconditional support, love and
affection. Their encouragement and never-ending kindness made everything
easier to achieve.
vii
Table of Contents
Title Page……………………………………………………………...……….i
Abstract………………………………………………….…………………...iii
Dedication………………………………………………..……………….…...v
Acknowledgements……………………………………..……………….…...vi
Table of Contents……………….…………………………………….……..vii
List of Tables……………………………………………………...………….xi
List of Figures……………………………………………………..………...xii
List of Appendices…………………………………………….….......…….xiii
CHAPTERS
1. INTRODUCTION……………………………………………………….....1
Research Background….………………………………………………1
Thesis Contributions…….……………………………………………..4
Thesis Outline………….………………………………………………5
2. OVERVIEW OF GYROSCOPE TECHNOLOGY………………...………7
Introduction……………………………………………………………7
viii
Micromachined Vibratory Rate Gyroscopes and Their Operational
Principles……………………………………………………………..11
Specifications of MEMS Gyroscopes ……………………..…………13
Applications of MEMS Gyroscopes……………….…………………14
Different Types of MEMS Vibratory Gyroscopes …..………………16
Control Strategies of MEMS Gyroscopes ……..….…………………23
Summary………………………………………………………...……26
3. OVERVIEW OF SINGLE BEAM GYROSCOPE…………………….…27
Introduction……………………………………………..……………27
Governing Equations of Motion…………...…………………………31
Beam Kinematics……………………………………………………..32
Translational Motion……………………………………………..…..33
Rotational Motion……………………………………………….……37
Piezoelectric Modeling……………………………………………….40
Constitutive Equations...…………………………………………...…41
Piezoelectric Patch Actuator....………………...………………….….42
Effects of Piezoelectric Actuator………………………...…………...43
Equations of Motion…………………..……………………………...45
ix
Time Response………………...………………..……………………48
Cross-axis Effects..…………………………………………………...52
Summary………………………………………………………...……54
4. MODELING OF THE ROCKING-MASS GYROSCOPE……………….55
Introduction………………………………………………………..…55
Principle of Operation...…………………………………………….. 56
Governing Equations of Motion...……………………………………59
Translational Motion….……………………………………………...60
Rotational Motion………………………………………………….…61
Equations of motion………...………………………………………...63
Summary………………….…………………………………………..70
5. ROCKING-MASS GYROSCOPE ANALYSIS AND NUMERICAL
SIMULATIONS……………………………………………………………..71
Development of the Frequency Equation…….………………………72
Validation of the Utilized Method for Solving EVP………..………..83
Frequency Analysis of a Timoshenko Beam…………………………83
Time Response of the System………………..………………………89
Summary………………………………………..….…………………93
x
6. CONCLUSIONS AND FUTURE WORK………..………………………94
Recommendations for Future Work………………………………….95
APPENDICES………………………………………….…………………………96
A: Detailed Derivation of Equations of Motion for a Single Beam
Gyroscope…………...………………………………………………..97
B: Frequency Equation…….……………………………………..…101
C: Maple Code to Develop the Frequency Equation………………..104
REFERENCES.………………………………………………….…………110
xi
List of Tables
2.1 Performance requirements of different gyroscopes ..................................14
3.1 Physical parameters of the system ............................................................51
5.1 Physical parameters of the system………………………………...……..81
5.2 The first three natural frequencies of the fixed-free Timoshenko beam...88
xii
List of Figures
1.1 A conventional gyroscope…………………………………………….…..2
2.1 Spinning wheel gyroscope ......................................................................... 8
2.2 (a) Optical gyroscopes (Ring laser gyroscope)...........................................9
2.2 (b) Optical gyroscopes (Fiber optic gyroscope) ........................................9
2.3 The Coriolis effect.....................................................................................10
2.4 Principle of operation of MEMS gyroscopes............................................12
2.5 Applications of MEMS gyroscopes……………...………………………15
2.6 Tuning fork gyroscope ..............................................................................16
2.7 Vibrating beam gyroscope ........................................................................17
2.8 Cylinder gyroscope operating principle.....................................................18
2.9 (a) Schematic of vibrating ring gyroscope ................................................19
2.9 (b) Flexural modes of vibrating ring gyroscope .......................................19
2.10 SEM image of the MEMS tuning fork gyroscope ..................................21
2.11 SEM image of a comb driven surface micromachined vibratory
gyroscope.........................................................................................................22
2.12 Dual axis rate gyroscope .........................................................................22
2.13 SEM image of ring gyroscope.................................................................23
xiii
3.1 Schematic of a single beam gyroscope…………………………………..27
3.2 Schematic of a flexural-torsional beam gyroscope………………...…….28
3.3 Cantilever beam kinematics……………………………………………...33
3.4 End mass kinematics…………………………………………………….35
3.5 Euler angle rotations……………………………………………….…….37
3.6 Piezoelectric patch actuator ……………………………………………..43
3.7 Schematic of a piezoelectric actuator attached on the beam…………….44
3.8 System response (torsional deflection θ(L,t)) to secondary base rotation;
(a) Ω1=0 rad/s , (b) Ω1 = 0.05 rad/s , (c) Ω1 = 0.1 rad/s , and (d) Ω1 = 0.5
rad/s……………………………………………………………………….…53
4.1 Schematic of a rocking-mass gyroscope………………………………....56
4.2 Primary rocking motion of the mass……………………………………..57
4.3 Secondary rocking motion of the mass…………………………………..58
4.4 Top view of the rocking-mass gyroscope with the coordinate systems…60
5.1 First mode shape of the system…………………………………………..82
5.2 Mode shapes of a cantilever Timoshenko beam (a) First mode (b) Second
mode (c) Third mode ……………………………………………………..…88
5.3 System response (a) bending deflection w1 and bending deflection w2 to
different base rotation rates: (b) Ω3=0, (c) Ω3=25 rad/s and (d) Ω3=50
rad/s………………………………………………………………………….91
xiv
List of Appendices
A Detailed Derivation of Equations of Motion for a Single Beam
Gyroscope……...…………………………………………………………….97
B Frequency Equation…………………………………………………...…101
C Maple Code to Develop the Frequency Equation…….………………….104
1
Chapter 1
INTRODUCTION
Research Background
A gyroscope is a device for maintaining orientation or measuring the
angle of rotation or the rate of change of angular rotation. In spite of the
conventional gyroscopes that use the conservation of angular momentum of a
spinning wheel for sensing the angular rate, a vibrating structure gyroscope,
which is a type of gyroscope that functions much like the halteres* of insects,
and can be used as a relatively inexpensive type of Attitude indicator†, uses
transfer of energy between two vibrating modes of a structure caused by
Coriolis acceleration, for measurement. The Coriolis acceleration is an
acceleration which arises as a result of motion of a particle relative to a
rotating reference. Only the components of motion in a plane parallel to the
equatorial plane are influenced. The effect is named after Gaspard-Gustave
Coriolis, a French scientist who described it in 1835, though the mathematics
appeared in the tidal equations of Pierre-Simon Laplace in 1778.
* Halteres, also known as balancers or poisers, are small knobbed structures found as a pair in some two-winged insects; they are flapped rapidly to maintain stability when flying. † An attitude indicator (AI), also known as gyro horizon or artificial horizon, is an instrument used in an aircraft to inform the pilot of the orientation of the airplane relative to earth.
2
Figure 1 shows a sample of conventional gyroscopes, and its
components. Since the conventional gyroscopes have a few major
disadvantages, namely their large size, high cost and limited life, they were
soon replaced by vibratory gyroscopes.
Figure 1. A conventional Gyroscope.
As mentioned before, the MEMS gyroscopes can be batch fabricated at
a very low cost (around US$30 per part in quantity as of late 2007). Since
there is no rotating part in this kind of gyroscope, they are commonly used in
different applications, to name a few, automotive active suspension, air bag
activation, consumer electronics and guided missiles. The vibrating element in
this type of gyroscope can have different shapes, but the most commonly used
elements are tuning forks, rings and beams. Among these, beams are being
used more commonly.
3
In vibrating beam gyroscope, the beam with a mass attached to its tip
(vibrating resonator), when rotated, is subjected to Corioli's effect that causes
secondary vibration orthogonal to the original vibrating direction. By sensing
the secondary vibration, the rate of turn can be detected. For vibration exert
and detection the piezoelectric effect is often used, therefore vibrating gyros
are often called "piezo", "ceramic", or "quartz" gyro, although in fact
vibration and detection do not necessary use the piezo effect.
Although these gyroscopes offer many advantages, they are extremely
sensitive to variations in system parameters such as length of the beam and
magnitude of primary excitation. They are prone to cross-axis effects and
quadrature errors that require advanced control strategies for their elimination.
But usually even with very good control strategies, it is very hard to eliminate
the cross-axis effect, in single beam gyroscopes. Therefore, a rocking-mass
gyroscope will be studied in this research. This gyroscope consists of four
beams and never encounter with cross-axis effect, and will not face the errors
discussed before. So it would be one of the most precise gyroscopes, if not the
most precise one. The main challenge is the complexity of the modeling of
this gyroscope. In spite of this fact, a detailed mathematical analysis of a
rocking-mass gyroscope will be performed in this work.
4
Thesis Contributions
As mentioned before, due to the advantages of the vibrating mass
gyroscopes, they are being used in many applications, and therefore, most of
the researches have conducted different studies on them. But the main
problem is that not many researchers have considered the components of the
gyroscope as elastic and continuous parts, although this is not a good
assumption, and causes error in calculations. In one of the latest works [1]
which deals with the beam gyroscope as a continuous system comprising of
the vibrating mass attached to the rotating base through flexible beam, it has
been concluded that there is a significant error in this kind of gyroscope,
called cross-axis effect, caused by the secondary base rotations in the system.
The work outlined in this thesis deals with a special kind of vibrating
mass gyroscope called rocking-mass gyroscope. In the past, few work has
been done on the rocking-mass gyroscopes. They are mostly focused on the
manufacturing aspects of this kind of gyroscope, and no researcher has
worked on the detail mathematical modeling and analysis of such a system.
In the present work, we offer a detail mathematical modeling and
analysis of a rocking-mass gyroscope. The gyroscope consists of four beams,
undergoing flexural-torsional vibrations, attached to a rocking mass in the
middle. The research primarily focuses on developing an accurate
5
mathematical model of this type of gyroscope. Governing equations of the
system are derived and simulated to analyze the effectiveness of this type of
gyroscope. The fundamental frequency of the system and its corresponding
mode shape will be found as well. Finally the time response of the system will
be presented.
Thesis Outline
The present work has five other chapters, as follows:
In chapter 2, we describe the gyroscopic systems and their basic
principles; however the main focus is on the vibratory gyroscopes and their
operational principles. Different types of the Micro-machined Electro-
Mechanical Systems (MEMS) gyroscopes and their principle of operation will
be pointed out. Finally the application of the discussed gyroscopes as well as
the control strategies will be briefly discussed.
In order to study the main drawback of the single-beam gyroscopes, we
will study a single beam gyroscope in chapter 3. Detailed governing equations
of motion will be derived for a macro-scaled cantilever beam gyroscope
undergoing coupled bending and torsional vibrations. After modeling the
system in full detail, we will perform a time-domain analysis to investigate
the cross-axis effect in this type of gyroscope.
6
Chapter 4 covers the detailed mathematical modeling of the rocking-
mass gyroscope consisting of four flexible beams, undergoing coupled
bending and torsional vibrations, and a rocking-mass attached to them in the
middle. Equations of motion and the boundary conditions are presented, for a
macro-scaled system.
Chapter 5 deals with the development of the frequency equation and
simulation of the system equations. Eigenvalue problem is solved using an
exact method. Simulations are carried out to study the gyroscopic coupling
present in the system as well as the effects of base rotation on the magnitude
of the gyroscopic effect. The time-domain analysis will be carried out to
check the effectiveness of the rocking-mass gyroscope.
The conclusions and suggestions for the future work are presented in
Chapter 6.
7
Chapter 2
OVERVIEW OF GYROSCOPE TECHNOLOGY
Introduction
Gyroscopes or gyro rate sensors are widely used for navigation,
stabilization, general rate control, pointing, autopilot systems, missile
guidance control, etc. A typical example is the application of yaw rate sensors
to automobiles to provide input to the control systems for suspension, braking
and steering. During recent years there have been attempts to develop low
cost gyroscopes suitable for mass production. A promising concept of such a
device is the vibratory gyroscope which can be fabricated using surface-
micromachining technology.
Conservation of momentum is used in gyroscopes to measure the
angular velocity or acceleration. In fact they use the principle of precession
which is actually Newton’s third law of motion. This principle expresses if an
unbalanced force is applied to a stationary object, the object will resist motion
in that direction. We can measure this force, to state the angular velocity or
acceleration of that object.
8
Gyroscopes can be categorized in different ways. Based on the
principle of operation, they fall into three groups: (a) Spinning mass
gyroscopes, (b) Optical gyroscopes and (c) Vibrating mass gyroscopes.
The spinning mass gyroscope is based on the principle that the spin
axis of a spinning mass will remain in a fixed direction in space unless acted
upon by an external influence. It has a mass spinning steadily with free
movable axis, called gimbals. In order to measure the angle of rotation, the
precession principle is used. When the gyroscope is tilted, gyroscopic
effect causes precession (motion orthogonal to the direction tilt sense) on
the rotating mass axis, hence gives the angle moved by the mass. A typical
spinning mass gyroscope is illustrated in Figure 2.1.
Figure 2.1: Spinning wheel gyroscope.
In optical gyroscopes, laser ray reflects round around many times
within the enclosure. If the enclosure rotates, the duration between the
moments of laser emittance to eventual reception will be different, so it is
9
based on the Sagnac effect [2]. In an RLG (Ring Laser Gyro), the laser
go-around is done by mirrors inside the enclosure, and in a FOG (Fiber Optic
Gyro) the laser go-around is done by a coil of optical-fiber. Laser emitter
deteriorates with time, and the fiber has its life fragile. Figure 2.2 shows the
two types of optic gyroscopes and their components.
(a) (b)
Figure 2.2: Optical gyroscopes (a) Ring laser [3] and (b) Fiber optic [4].
The next type of gyroscopes is vibrating gyroscopes. Among all the
mentioned types, this type is the most commonly used. A vibrating element
(vibrating resonator), when rotated, is subjected to Coriolis effect that causes
secondary vibration orthogonal to the original vibrating direction. By sensing
the secondary vibration, the rate of turn can be detected.
Almost all reported micromachined gyroscopes use vibrating
mechanical elements to sense rotation. They have no rotating parts that
require bearings, and hence they can be easily miniaturized and batch
fabricated using micromachining techniques. All vibratory gyroscopes are
10
based on the transfer of energy between two vibration modes of a structure
caused by Coriolis acceleration. To understand the Coriolis effect, imagine a
particle traveling in space with a velocity vector V. An observer sitting on the
x-axis of the xyz coordinate system, shown in Figure 2.3, is watching this
particle. If the coordinate system along with the observer starts rotating
around the z-axis with an angular velocity Ω, the observer thinks that the
particle is changing its trajectory toward the x-axis with an acceleration equal
to 2V×Ω. Although no real force has been exerted on the particle, to an
observer, attached to the rotating reference frame an apparent force has
resulted that is directly proportional to the rate of rotation. This effect is the
basic operating principle underlying all vibratory structure gyroscopes [5].
Figure 2.3: The Coriolis effect [5].
11
Micromachined Vibratory Rate Gyroscopes and Their
Operational Principles
Even though an extensive variety of micromachined gyroscope designs
and operation principles exists, almost all of the reported micromachined
gyroscopes use vibrating mechanical elements to sense angular rate. The
concept of utilizing vibrating elements to induce and detect Coriolis force
involves no rotating parts that require bearings, and have been proven to be
effectively implemented and batch fabricated in different micromachining
processes [6]. They are based on the transfer of energy between two modes of
vibration of a structure [7]. Various elements such as tuning forks, beams,
shells, rings, discs and cylinders are used as the proof mass in MEMS
gyroscope.
The operation principle of the vast majority of all existing
micromachined vibratory gyroscopes relies on the generation of a sinusoidal
Coriolis force due to the combination of vibration of a proof-mass and an
orthogonal angular-rate input. The proof mass is generally suspended above
the substrate by a suspension system consisting of flexible beams. Figure 2.4
illustrates an EMS gyroscope consists of a vibrating proof mass suspended
over a substrate via elastic beams. A force is applied to the proof mass to
vibrate along the x-axis (drive mode). When the gyroscope is subjected to an
12
angular rotation, a sinusoidal Coriolis force is induced in the direction
orthogonal to the drive-mode oscillation at the driving frequency. The
magnitude of this force is given as
Ω×= VmFc 2 (2.1)
where m is the vibrating mass, V is the velocity in the primary direction and Ω
is the applied rotation rate. As mentioned before, due to the Coriolis force, the
mass starts vibrating in the secondary direction (sense mode), and the
magnitude of the sense mode vibration, which is proportional to the rate of
rotation, can be measured to determine the rate of rotation. Different methods
can be utilized for actuating and sensing the vibrations, such as electrostatic,
piezoelectric or electromagnetic.
Figure 2.4: Principle of operation of MEMS gyroscopes [8].
Ideally, it is desired to utilize resonance in both the drive and the sense
modes in order to attain the maximum possible response gain, and hence
13
sensitivity. This is typically achieved by designing and electrostatically tuning
the drive and sense resonant frequencies to match. Alternatively, the
sense-mode is designed to be slightly shifted from the drive-mode to improve
robustness and thermal stability, while intentionally sacrificing gain and
sensitivity. However, the limitations of the photolithography-based
micromachining technologies define the upper-bound on the performance and
robustness of micromachined gyroscopes [6].
Specifications of MEMS Gyroscopes
The performance of a gyroscope is determined through three important
parameters of resolution, drift, zero-rate output (ZRO), and scale factor. In the
absence of rotation, the output signal of a gyroscope is a random function that
is the sum of white noise and a slowly varying function [9]. The white noise
defines the resolution of the sensor and is expressed in terms of the standard
deviation of equivalent rotation rate per square root of bandwidth of detection
[(o/s)/√Hz or (o/h)/√Hz]. The so-called “angle random walk” in o/√h may be
used instead. The short- or long-term drift of the gyroscope corresponds to the
peak-to-peak value of the slowly varying function [9]. The last very important
factor or the gyroscope is the Zero Rate Output (ZRO), which represents the
output of the device in the absence of a rotation rate [5]. In another way of
classification, gyroscopes fall into three categories: inertial-grade, tactical-
14
grade, and rate grade devices. Table 2.1 summarizes the requirements for each
of these categories ([10], [11]).
Table 2.1: Performance requirements of different gyroscopes [5]
Parameter Rate Grade Tactical Grade Inertial Grade
Angle Random Walk, (o/ h ) >0.5 0.5-0.05 <0.001
Bias Drift, (o/h) 10-1000 0.1-10 <0.01 Scale Factor Accuracy, (%) 0.1-1 0.01-0.1 <0.001 Full Scale Range, (o/sec) 50-1000 >500 >400 Max. Shock in 1 msec, (g’s) 103 103-104 103 Bandwidth, (Hz) >70 ~100 ~100
Applications of MEMS Gyroscopes
Applications for MEMS gyroscopes are very broad. Some example for
these applications are; automotive; vehicle stability control, rollover detection,
navigation, load leveling/suspension control, event recording, collision
avoidance; consumers, computer input devices, handheld computing devices,
game controllers, virtual reality gear, sports equipment, camcorders, robots;
industrial., navigation of autonomous (robotic) guided vehicles, motion
control of hydraulic equipment or robots, platform stabilization of heavy
machinery, human transporters, yaw rate control of wind-power plants;
aerospace/military; platform stabilization of avionics, stabilization of pointing
systems for antennas, unmanned air vehicles, or land vehicles, inertial
measurement units for inertial navigation, and many more. Different
15
application areas of MEMS gyroscopes based on different accuracy and range
requirements, are presented in Figure 2.5.
Anti-rollover mechanisms, GPS navigation and electronic stability
control can be achieved in the automotive sector, using gyroscopes. A lot of
automobile manufacturers and researchers are conducting research on new
applications (like active suspension, skid control, and …) of MEMS gyros in
cars. Moreover MEMS gyroscopes are being used by various consumer
electronics. To name a few, they are being used in camcorder stabilization,
game controllers, handheld computing devices, virtual reality gear and robots.
Furthermore, MEMS gyroscopes are also used in inertial navigation of
autonomous guided vehicles, platform stabilization of heavy machinery, yaw
rate control of wind power plants and inertial navigation for military
applications [12].
Figure 2.5: Applications of MEMS gyroscopes [13].
16
Different Types of MEMS Vibratory Gyroscopes
A number of vibratory gyroscopes have been demonstrated, including
tuning forks [14–17], vibrating beams [18], and vibrating shells [19]. Tuning
forks are a classical example of vibratory gyroscopes. The tuning fork, as
illustrated in Figure 2.6, consists of two tines that are connected to a junction
bar. In operation, the tines are differentially resonated to fixed amplitude, and
when rotated, Coriolis force causes a differential sinusoidal force to develop
on the individual tines, orthogonal to the main vibration. This force is detected
either as differential bending of the tuning fork tines or as a torsional vibration
of the tuning fork stem. The actuation mechanisms used for driving the
vibrating structure into resonance are primarily electrostatic, electromagnetic,
or piezoelectric. To sense the Coriolis-induced vibrations in the second mode,
capacitive, piezoresistive, or piezoelectric detection mechanisms can be used.
Optical detection is also feasible, but it is too expensive to implement [5].
Figure 2.6: Tuning fork gyroscope [20].
17
In a vibratory gyroscope, especially vibrating beam gyroscope which
consists of a metallic beam, the drive mode is a flexural vibration which has
been induced through a piezoelectric actuator, placed on the beam. A
secondary vibration (sense-mode), normal to the drive mode vibrations, is
induced in the beam due to the Coriolis force arises in the presence of the
rotation about the longitudinal axis of the beam. The secondary vibration can
be sensed by sensors placed on the beam as shown in Figure 2.7. From this
secondary vibration, the rate of rotation can be determined.
Figure 2.7: Vibrating beam gyroscope [1].
The main concept in a vibratory cylinder gyroscope is: the nodes on the
circumference of a vibrating cylinder (ring) do not stay fixed with respect to
the cylinder itself when it is rotated around its central axis, but they move by a
18
quantity proportional to the turn, as sketched in Figure 2.8. This is due to the
Coriolis coupling between the two vibration modes that change as cos2θ
(primary) and sen2θ (secondary) around the circumference. A certain number
of electrodes are located around the circumference with the aim to excite and
detect both the primary and secondary vibration modes [3].
Figure 2.8: Cylinder gyroscope operating principle [3].
Figure 2.9 (a) illustrates a vibrating ring gyroscope which comprises a
ring, eight semicircular support springs, and drive, sense and control
electrodes [19]. Symmetry considerations require at least eight springs to
result in a balanced device with two identical elliptically-shaped flexural
modes that have equal natural frequencies and are 45o apart from each other
[19].
The antinodes of the second flexural mode are located at the nodes of
the first flexural mode (see Figure 2.9 (b)). The ring is electrostatically
19
vibrated into the primary flexural mode with fixed amplitude. When device is
subjected to rotation around its normal axis, Coriolis force causes energy to be
transferred from the primary mode to the secondary flexural mode, which is
located 45o apart from the primary mode, causing amplitude to build up
proportionally in the latter mode; this build-up is capacitively monitored. The
amplitude of the second mode (sense mode) is proportional to the rotation rate
and can be measured to determine the rate of rotation [21].
(a)
(b)
Figure 2.9: Vibrating ring gyroscope; (a) Schematic of the gyroscope [21] and (b) Flexural
modes of a vibrating ring gyroscope [21].
20
Micromechanical gyros are usually designed as an electronically driven
resonator, often fabricated out of a single piece of quartz or silicon. Such
gyros operate in accordance with the dynamic theory that when an angular
rate is applied to a translating body, a Coriolis force is generated. When this
angular rate is applied to the axis of a resonating tuning fork, its tines
experience a Coriolis force, which then produces torsional forces about the
sensor’s axis. These forces, which are proportional to the applied angular rate,
cause displacements that can be measured capacitively in a silicon instrument
or piezoelectrically in a quartz instrument. The output is then demodulated,
amplified and digitized to form the device output [7].
In spite of having high quality factors, quartz vibratory gyroscopes do
not have compatibility of being processed with integrated circuit fabrication
technology. Based on this fact, they were replaced by silicon, when the batch
fabrication technology was introduced.
The Charles Draper Laboratory is one of the pioneers in manufacturing
the micromachined silicon rate gyroscopes. The first MEMS silicon gyro, they
built, had a double gimbals vibratory gyroscope supported by torsional
flexures, with the vibrating mechanical element made from p++ silicon [22].
After that, they built an improved silicon-on-glass tuning fork gyroscope [23].
It was fabricated through the dissolved wafer process [24]. In order to achieve
large amplitude of motion (10 μm), a set of interdigitated comb drives are
21
used to vibrate this gyroscope electrostatically in its plane [25]. The entire
structure will undergo an out-of-plane rocking motion, if it rotates about a
direction perpendicular to the drive mode. This motion can be capacitively
measured, to determine the rate of rotation. A Scanning Electron Microscope
(SEM) image of such device is illustrated in Figure 2.10.
Figure 2.10: SEM image of MEMS tuning fork gyroscope [26].
Different mechanisms are being used for sensing and actuation in
Micromachined gyros. Electromagnetic actuation has been utilized in some
tuning forks to achieve large amplitude of motion [27-29]). Reference [30]
discusses the use of piezoresistive detection in some of the MEMS
gyroscopes. Some researchers at University of California, Berkeley and
Samsung electronics, developed single and dual-axis polysilicon surface-
micromachined gyroscopes. The vibratory gyroscope developed by Samsung
is illustrated in Figure 2.11. It consists of a 7-μ m-thick polysilicon resonating
mass supported by four fishhook-shaped springs [31-35].
22
Figure 2.11: SEM image of comb driven surface micromachined vibratory gyroscope [31].
Figure 2.12 illustrates a surface-micromachined dual-axis gyroscope
which is based on rotational resonance of a polysilicon rotor disk [36]. The
sensor can sense rotation equally about these two axes, since the disk is
symmetric in two orthogonal axes. A bulk-micromachined, precision silicon
MEMS vibratory gyroscope for space applications was fabricated buy the Jet
Propulsion Laboratory (JPL), in collaboration with the University of
California, Los Angeles [37], [38]. The vibrating ring gyroscope which was
built by Researchers at General Motors and the University of Michigan,
consists of a ring, semicircular support springs, and drive, sense, and balance
electrodes, which are located around the structure [39].
Figure 2.12: Dual axis rate gyroscope [36].
23
An SEM image of a 1.7×1.7 mm2 PRG is illustrated in Figure 2.13.
Different important features of high-performance gyroscopes, like small ring-
to-electrode gap spacing for increasing the sense capacitance, large structural
height for increasing the radius and sense capacitance and reducing the
resonant frequency, and good structural material (polysilicon) with an
orientation-independent Young’s modulus are available in this device [5].
Figure 2.13: SEM image of ring gyroscope ([40], [41]).
Control Strategies of MEMS Gyroscopes
In MEMS vibrating mass gyroscopes, one of the modes of the mass is
actuated into a known oscillatory motion. Due to the Coriolis acceleration a
secondary vibration is induced in the other mode of the mass (sense-mode).
By measuring the response of the sense mode, the angular rate can be
determined.
24
The conventional mode of operation is classified into the open-loop
mode and the closed-loop mode. The main difference between the closed-loop
and open-loop mode of operation lies in that in the former the displacement of
the sense axis is controlled to zero, while in the latter it is measured. Most
MEMS gyroscopes are currently operated in the open-loop mode. The main
advantage of the open-loop mode of operation is that circuitry used for the
operation of gyroscope in this mode is simpler than in the other modes, since
there is no control action in the sense axis. Thus, this mode can be
implemented relatively easily and cheaply. However, under an open-loop
mode of operation, the gyroscope's angular rate scale factor is very sensitive
and not constant over any appreciable bandwidth, to fabrication defects and
environment variations. Therefore, the application areas for the open-loop
mode are limited to those which require low-cost and low-performance
gyroscopes. In contrast to the open-loop mode of operation, in the closed-loop
mode of operation, the sense amplitude of oscillation is continuously
monitored and driven to zero. As a consequence, the bandwidth and dynamic
range of the gyroscope can be greatly increased beyond what can be achieved
with the open-loop mode of operation [42]. However, under conventional
closed-loop mode of operation, it is difficult to ensure a constant noise
performance, in the face of environment variations such as temperature
changes, unless an on-line mode tuning scheme is included. Moreover, there
are practical difficulties in designing a feedback controller which closed-loop
25
system is stable and sufficiently robust, for gyroscopes with high Q (quality
factor) systems. Therefore, the application areas for conventional closed-loop
mode of operation are those which requires medium-cost and medium-
performance (large bandwidth but limited resolution) gyroscopes [43].
The control system of the MEMS gyroscope has to perform four main
tasks: (i) initiate drive axis vibrations at resonant frequency, (ii) maintain the
amplitude of primary vibration at a desired level, (iii) eliminate the cross
coupling errors in the gyroscope and finally (iv) determine the input angular
velocity [43].
Over the years, various controllers were developed for performing
these basic tasks for MEMS gyroscopes. To drive the input frequency of the
drive axis to resonance, phase locked loop technique is used by some
researchers [44]. In phase locked loop, the input frequency is adjusted until
the output of the drive axis is out of phase with the input by -90o, indicating
resonance [45].
In reference [46], an adaptive controller has been developed in which
the system parameters are adjusted using a feedback loop such that resonance
is achieved at a given input frequency. Different methods may be utilized to
adjust the amplitude of the primary vibrations, such as using an automatic
gain control loop [47] or using adaptive controller as described in [48].
26
Several different strategies were used for canceling the quadrature errors and
estimating the angular velocity as follows:
• For open-loop mode of operation: Feedforward scheme [49] and
Feedback control scheme [50];
• For closed-loop mode of operation, a Kalman filter based approach
[51] and force-balancing feedback scheme [52].
an observer-based adaptive controller that is self-calibrating,
compensates for fabrication errors and estimates the angular velocity is
offered in [43]. In reference [53] the Control scheme for a z -axis MEMS
vibrational gyroscope is developed using basic linear system techniques.
Summary
Different types of gyroscope were introduced and they were
categorized in different ways, based on their shapes, applications, and ….
Basic principles of conventional gyroscopes as well as the operational
principles of different types of gyroscope were investigated. Moreover,
several applications of MEMS gyroscopes were briefly presented. Finally, the
main tasks of control systems in gyroscopes and different control strategies of
MEMS gyroscopes were discussed.
27
Chapter 3
OVERVIEW OF SINGLE BEAM GYROSCOPE‡
Introduction
As shown in Figure 3.1, a vibrating beam gyroscope comprises a
cantilever beam with a tip mass, attached to a moving base. According to the
discussions of chapter 2, the gyroscope is based on the principle of Coriolis
acceleration. A lateral vibratory motion is induced in the beam using a
suitable actuation mechanism. In the presence of the angular rotation of the
beam along its longitudinal axis, secondary lateral vibrations are induced in
the beam in the direction orthogonal to the primary oscillations (Flexural-
Flexural). By measuring these secondary vibrations, rate of angular rotation
can be determined.
Figure 3.1: Schematic of a single beam gyroscope.
‡ This chapter is based on the work done in Ref. [1]. Since most of the concepts used in this chapter will be used in the next chapter, we will treat it in detail.
28
This chapter deals with a single beam gyroscope. First we derive the
equation of motion for such a system. Some sample simulations will be
presented, and finally we will investigate the cross-axis effect in this kind of
gyroscope.
A second type of vibrating beam gyroscope will be studied while a
cantilever beam with a tip mass is subjected to a combination of flexural-
torsional vibrations. Similar to the case of flexural-flexural vibrating beam
gyroscope, a flexural vibration is induced in the beam using piezoelectric
actuator (drive mode) and due to the Coriolis force, in presence of the rotation
about the z-axis, a secondary torsional vibration is induced in the beam (sense
mode). A schematic of such a system is illustrated in Figure 3.2.
Figure 3.2: Schematic of a flexural-torsional beam gyroscope [1].
29
The secondary torsional vibrations, which are proportional to the rate
of rotation of the beam, can be measured using an accelerometer, laser sensors
or piezoelectric sensor (as shown), in order to determine the rate of rotation.
The effect of the end mass is to improve the performance of the gyroscope by
increasing the gyroscopic effect.
The measurement of the rate of angular rotation will be accurate, if the
beam only rotates around the Z -axis. In practice, however, the base of the
gyroscope is subjected to secondary rotations as well (e.g. rotation about the
longitudinal axis), which can produce significant errors in the measurement of
the primary angular velocity. In fact, one of the major sources of error in the
vibrating beam gyroscope is the presence of these secondary base rotations
[54].
Many researchers have worked on the problem of coupled bending-
torsion vibrations of cantilever beams, because of its practical importance in
various applications. The theory of coupled flexural-torsional vibrations of
thin walled beams was first developed by Timoshenko and Young [55]. In
their research, they obtained the exact modal solutions for such systems. The
coupled free vibration frequencies of a cantilever beam, was calculated by
Dokumaci [56]. Bercin and Tanaka [57] included warping, shear deformation
and rotary inertia effects into the previous studies. Banerjee [58] developed a
dynamic stiffness matrix analysis method to obtain the natural frequencies and
30
mode shapes of the coupled Euler-Bernoulli beam. The effect of tip mass was
not considered in the mentioned works. Modeling of the tip mass, using the
Euler-Bernoulli beam theory is discussed in [59] and [60]; also Bhat and
Wagner [61] tried to find the frequency equations of a cantilever beam with
tip mass.
More recently, Kirk and Wiedemann [62] performed a study on the
free vibration of a flexible beam with rigid payloads at the tip. They used the
Euler-Benoulli theory but the effect of torsion was not considered.
Oguamanam [63] carried out research on a cantilever beam with a rigid tip
mass, whose center of gravity was not coincident with the attachment point.
This work was extended by Gokdag and Kopmaz [64] through analyzing the
coupled flexural-torsional free and forced vibrations of a beam with tip and in
span attachments. H. Salarieh and M. Ghorashi [65] performed the analyses
on the same system but having Timoshenko beam. Most of these works
analyze the cantilever beam when the base is stationary.
Esmaeili et al. in a series of publications studied the flexural-flexural
vibrations of a cantilever beam with tip mass and subjected to general support
motion [66- 68]. Bhadbhade et al [1] carried out a study on a piezoelectrically
actuated flexural/torsional vibrating beam gyroscope and investigated the
effects of secondary rotation (cross-axis effect) on the precision of the
gyroscope.
31
Governing Equations of Motion
Since we are to the study the coupled flexural-torsional vibrations of
the system, we will derive the two linear partial differential equations of
motion of the system. Several methods can be applied to derive the equations
of motion, such as Newton’s second law of motion, Lagrange’s equation and
extended Hamilton’s principle. In this work, we will use the extended
Hamilton’s Principle to derive the equations of motion.
Since the system is studied in the macro scaled, a few assumptions are
made accordingly. The beam is assumed to follow the Euler-Bernoulli theory
and i.e. the effects of warping and shear deformation are neglected. The beam
is considered to be a slender beam (with small thickness to length ratio). The
Poisson effect is also neglected.
In order to implement the extended Hamilton’s principle, we need to
know the total kinetic energy and potential energy of the system as well as the
total non-conservative work done on it. Beam kinematics should be studied to
define different motions in the systems and their relations, so that we can
define the total kinetic and potential energy of the system.
32
Beam Kinematics
Let’s consider the system shown in Figure 3.3. A rigid tip mass M of
finite dimensions (with length l) is attached to the end of a uniform and
straight metallic cantilever beam with length L and mass per unit length bρ .
We will define two different coordinate sytems: Inertial and rotating
coordinate. The inertial coordinate system is denoted by ( 1 2 3, ,A A A ). The
moving (rotating) coordinate system is denoted by (X,Y,Z) with orthogonal
unit vectors ( , ,X Y Za a a ). We also define a local curvilinear coordinate system
at point p, which is denoted by (x,y,z) and has orthogonal unit vectors
( , ,x y za a a ) but is not shown in the figure. As mentioned before, the primary
bending vibrations w(x,t) in the beam, using piezoelectric actuator attached on
the beam. The base is subjected to two angular rotations; (i) the primary
rotation 3Ω which is to be measured by the gyroscope and (ii) the
secondary rotation 1Ω which causes errors in the measurement of primary
rotation. In presence of the primary rotation 3Ω (about the Z -axis), due to the
Coriolis effect, the secondary torsional vibrations ( , )x tθ are induced in the
beam. After defining the basic requirements, now we can study different types
of motion in the system.
33
Figure 3.3: Cantilever beam kinematics [1].
Translational Motion
The base motion causes each point on the neutral axis to undergo an
elastic deformation and a rotation. Three Cartesian variables ui, i=1, 2, and 3
(where u1=u, u2=v, u3=w) measured in moving coordinate system (X,Y,Z), are
used in order to describe the translational motion of the beam (as shown in
Figure 3.3). As a result of this, the point P on the neutral axis of beam will be
moved to point P*. The position and velocity of point P* in reference frame
Ai can be defined as [54]:
* *
* *
*
*
( ) ( )( )
( ) ( ) ( )
pp p
p pbase p p
p ii j j X i i
r r u
d r d ur u
dt dtd r d u a a sa u a
dt dt
= +
= +Ω × +
= +Ω × +
(3.1)
34
where jΩ is the rotation of the base relative to the reference frame ( 1 2 3, ,A A A )
and s is the position of point P in the moving coordinate system. In this case,
we assume that, the beam has no axial and lateral vibrations, hence u=0 and
v=0. If we simplify the vector products used in Eq. (3.1), different velocities
of point P* can be expressed as:
*( )P
X Y Z
d rfa ga ha
dt= + + (3.2)
where
1 3
0fg w s
wht
== − Ω + Ω∂
=∂
(3.3)
Therefore, kinetic energy of the beam due to the translational motion is
given as
2 2 21 0
1 ( )2
L
bT f g h dxρ= + +∫ (3.4)
The same approach and procedure can be used, in order to calculate the
translational kinetic energy of the tip mass, since the mass is attached to the
end of the beam, with the center of mass collinear with the beam centroid.
35
According to Figure 3.4, position and velocity of the center of gravity
of the mass is
*
*
*
( )( ) ( ) ( )
Mass q mq
qMass mbase q mq
r r u r
d ud r d r r u rdt dt dt
= + +
= + +Ω × + +(3.5)
where q is the point where end mass attaches to the beam, q* is the deformed
position of point q and rm is the position vector of center of gravity of the end
mass from point q* , in the deformed position given as
1 1cos sin2 2m L X L Zr a aψ ψ= + (3.6)
where Lx L
wx
ψ=
∂=∂
.
Figure 3.4: End mass kinematics [1].
36
Consequently, we will have
( ) ( ) sin cos2 2
cos sin2 2
Mass i L Li L X L Z
j j i i X L X L Z
d r d u l la a adt dt t t
l la u a La a a
ψ ψψ ψ
ψ ψ
∂ ∂⎛ ⎞= + − +⎜ ⎟∂ ∂⎝ ⎠⎛ ⎞+Ω × + + +⎜ ⎟⎝ ⎠
(3.7)
By simplifying Eq. (3.7), we can get
( )MassM X M Y M Z
d r f a g a h adt
= + + (3.8)
where
3 1 3 1
sin2
cos sin2 2
cos2
( , ) .
LM L
M L L L
L LM L
L x L
lft
l lg L w
w lht t
w w x t
ψψ
ψ ψ
ψψ
=
∂= −
∂
= Ω −Ω +Ω −Ω
∂ ∂= +
∂ ∂=
(3.9)
Since we are working on a macro-scaled system, we can assume small
deflections and ignore the nonlinear terms. This yield
3 1 3 1
0
2 2
2
M
M L L
L LM
fl lg L w
w lht t
ψ
ψ
=
= Ω −Ω +Ω −Ω
∂ ∂= +
∂ ∂
(3.10)
37
Therefore, the translational kinetic energy of the end mass can be given
as:
( )2 2 21
12M M M MT M f g h= + + (3.11)
Rotational Motion
In order to describe the deformation of the system from its original
configuration, we use the Euler angle rotations. The rotating coordinate
system with orthogonal unit vectors ( , ,X Y Za a a ) is denoted by (X,Y,Z) (See
Figure 3.5). The X-axis coincides with the longitudinal/centroidal axis of the
beam before deformation. The local curvilinear coordinate system at arclength
s in the deformed position is denoted by (x,y,z), with orthogonal unit vectors
( , ,x y za a a ) [69].
Figure 3.5: Euler angle rotations [69].
38
Each cross section of the beam will have an elastic displacement of its
neutral axis and a rotation. We can find the displacement components
according to the equations given in the previous section. we use successive
Euler angle (counterclockwise) rotations with the angle of rotations denoted
by ψ(x,t) and θ(x,t), in order to describe the rotation of the neutral axis, from
the undeformed to the deformed position, as shown in Figure 3.5, where
( , )( , ) w x tx tx
ψ ∂=
∂.
(X,Y,Z) coordinate system will be taken to (x',y'=Y,z') by the first
rotation ψ, about Y. The second rotation θ about x' takes (x',y',z') to the final
orientation (x=x',y,z). The relationship of the three unit vector triads is given
in this form [69]
[ ] [ ]'
'
'
x x X
y y Y
Tz z Z
a a aa T a T T aa a a
θ θ ψ
⎧ ⎫ ⎧ ⎫ ⎧ ⎫⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎡ ⎤= =⎨ ⎬ ⎨ ⎬ ⎨ ⎬⎣ ⎦⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎩ ⎭ ⎩ ⎭ ⎩ ⎭
(3.12)
where
[ ]
[ ]
1 0 0 cos 0 sin0 cos sin , 0 1 00 sin cos sin 0 cos
cos 0 sinsin sin cos sin coscos sin sin cos cos
T T
T
θ ψ
ψ ψθ θθ θ ψ ψ
ψ ψθ ψ θ θ ψθ ψ θ θ ψ
−⎡ ⎤ ⎡ ⎤⎢ ⎥ ⎢ ⎥⎡ ⎤= =⎣ ⎦⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥−⎣ ⎦ ⎣ ⎦
−⎡ ⎤⎢ ⎥= ⎢ ⎥⎢ ⎥−⎣ ⎦
(3.13)
39
Thus, the angular velocity of the beam is given as
1 3X Z Y xa a a at tψ θω ∂ ∂
= Ω +Ω + +∂ ∂
(3.14)
Since the transformation matrices [Tθ], [Tψ] and [T] are orthogonal
matrices, they posses the property[ ] [ ]1 TQ Q− = . Using Eqs. (3.13) and (3.14)
and this property, we can obtain the absolute angular velocity ω(x,t) in this
form
1 3
1 3
1 3
cos sin
cos sin cos sin cos
sin cos sin cos cos
x
y
z
at
at
at
θω ψ ψ
ψ θ θ ψ θ ψ
ψ θ θ ψ θ ψ
∂⎛ ⎞= +Ω −Ω⎜ ⎟∂⎝ ⎠∂⎛ ⎞+ +Ω +Ω⎜ ⎟∂⎝ ⎠∂⎛ ⎞+ − +Ω +Ω⎜ ⎟∂⎝ ⎠
(3.15)
According to Eq. (3.15), the components of the absolute angular
velocity of the beam can be expressed as
1 3
1 3
1 3
cos sin
cos sin sin sin cos
sin cos sin cos cos
x
y
z
t
t
t
θω ψ ψ
ψω θ θ ψ θ ψ
ψω θ θ ψ θ ψ
∂⎛ ⎞= +Ω −Ω⎜ ⎟∂⎝ ⎠∂⎛ ⎞= +Ω +Ω⎜ ⎟∂⎝ ⎠∂⎛ ⎞= − +Ω +Ω⎜ ⎟∂⎝ ⎠
(3.16)
Using the assumption of small angles of bending and torsion, we get
40
( )
1 3
1 3
1 3
x
y
z
t
t
θω ψ
ψω θψ θ
ω ψ
∂⎛ ⎞= +Ω −Ω⎜ ⎟∂⎝ ⎠∂⎛ ⎞= +Ω +Ω⎜ ⎟∂⎝ ⎠
= Ω +Ω
(3.17)
The kinetic energy of the beam and end mass due to the rotational
motion is
( )2 2 22 0
12
L
xb x yb y zb zT I I I dxω ω ω= + +∫ (3.18)
( )2 2 22
12M xM x yM y zM zT I I Iω ω ω= + + (3.19)
where Ixb,Iyb,Izb and IxM,IyM,IzM are the mass moments of inertia of the beam
and end mass, about the X,Y and Z axes, respectively.
Piezoelectric Modeling
Piezoelectricity is the ability of some materials (notably crystals and certain
ceramics) to generate an electric potential in response to applied mechanical stress.
This may take the form of a separation of electric charge across the crystal lattice. If
the material is not short-circuited, the applied charge induces a voltage across the
material. The word is derived from the Greek piezein, which means to squeeze or
press. The piezoelectric effect is reversible in that materials exhibiting the direct
piezoelectric effect (the production of electricity when stress is applied) also exhibit
the converse piezoelectric effect (the production of stress and/or strain when an
41
electric field is applied). For example, lead zirconate titanate (PZT) crystals will
exhibit a maximum shape change of about 0.1% of the original dimension.Due to
their precise operation, piezoelectric actuators are becoming increasingly important
in micro-positioning technology. The direction of expansion with respect to the
direction of the electrical field depends on the constitutive equations of the
piezoelectric material [70].
Constitutive Equations
The fundamental relations for the piezoelectric materials are [70, 71]
E PS s T dE= + (3.20)
T PD dT Eε= + (3.21)
where d is the piezoelectric constant, S is the strain, EP is the electric field, T
is the mechanical stress, εT is the permittivity matrix under constant stress,
and sE refers to the compliance of material when the electric field is constant.
In Eq. (3.21), d relates the electric charge per unit area D (the dielectric
displacement) to the stress T under a zero electric field. We can also rewrite
Equs. (3-20) and (3-21) in the following form
E PT c S eE= − (3.22)
s PD eS Eε= + (3.23)
42
where cE = 1/sE is the Young’s modulus matrix under constant electric field,
e=d/sE is a constant matrix relating the charge per unit area to the strain, and
εS is the permittivity matrix under constant strain [70, 72].
Piezoelectric Patch Actuator
As shown in Figure 3.6, piezoelectric patch actuator comprises of a
thin piezoelectric film bonded on the structure. The geometrical arrangement
is such that d31 (the piezoelectric coefficient that relates the electrical field in
thickness direction to the strain in longitudinal direction) dominates the design
and the useful direction of expansion is normal to that of the electrical field.
Using standard engineering notations and one-dimensional deformation
assumption, Eq. (3.22) within the piezoelectric layer for patch actuator
reduces to [1, 70]
31( )
x P x PP
v tE E dt
σ ε= − (3.24)
where EP is the Young’s modulus of elasticity of the piezoelectric actuator
and v(t) is the applied voltage to the actuator.
43
Figure 3.6: Piezoelectric patch actuator [72].
Effect of Piezoelectric Actuator
In order to find the effect of the piezoelectric actuator, we first write
the equilibrium equation of a classical Euler-Bernoulli beam as follows
2 2
2 2
( , ) ( , )w x t M x tAt x
ρ ∂ ∂= −
∂ ∂ (3.25)
where M(x,t) is the cross-sectional bending moment acting at distance x from
the clamped end of the beam. Since the thickness of piezoelectric layer is very
much less than the thickness of the beam, we can assume that the neutral axis
of the beam does not change in the beam. Therefore, the bending moment is
expressed in this form [72]
2
312
( , ) 1( , ) ( ) ( )( ) ( )2 P b P
w x tM x t EI x bE d v t t t S xx
∂= + +
∂(3.26)
44
where
1 2( ) ( ) and ( ) ( ) ( )b PEI x EI S x EI S x H x l H x l= + = − − − (3.27)
In these equations, b and tb are the width and thickness of the beam,
respectively; tP, EP and d31 are the thicknesses, elastic modulus and
piezoelectric constant of the actuator, respectively; v(t) is the voltage applied
to the actuator, EIb and EIP are the flexural rigidities of the beam and actuator,
respectively. As shown in Figure 3.7, the piezoelectric layer is just bound on a
finite part of the beam, so H(x), the Heaviside function, is used to show that it
is located from l1 to l2. Introducing (3.26) into (3.25) yields
2 2 2 2
312 2 2 2
( , ) ( , ) 1( ) ( )( ) ( )2
Pb P
w x t w x tA EI x bE d v t t t S xt x t x
ρ⎛ ⎞∂ ∂ ∂ ∂ ⎛ ⎞+ = − +⎜ ⎟ ⎜ ⎟∂ ∂ ∂ ∂ ⎝ ⎠⎝ ⎠
(3.28)
Figure 3.7: Schematic of a piezoelectric actuator attached on the beam.
If we compare the typical Euler-Bernoulli beam theory with Eq. (3.28)
we can conclude that the distributed actuator is equivalent to adding
45
concentrated moments MP at the boundaries of the actuator [73], where MP is
expressed as
31 01 ( ) ( ) ( ) ( ) ( )2P P b P PM bE d t t v t S x M v t S x= − + = (3.29)
Equations of Motion
As mentioned in previous chapters, the extended Hamilton’s principle
is used to derive the two linear partial differential equations (PDEs) governing
the flexural-torsional vibrations of the beam. The extended Hamilton’s
principle for a dynamic system is expressed as
2
1
0t
nctT V W dtδ δ δ− + =∫ (3.30)
where T is the total kinetic energy of the system, V is the total potential energy
and Wnc is the total non-conservative work done on the system. The total
kinetic energy of the beam can be defined as the summation of the kinetic
energy due to the translational motion and the rotational motion. Ignoring the
rotary inertia terms, the total kinetic energy of the beam can be obtained from
Eqs. (3.4) and (3.18) as follows
( )2 2 2 2
0
1 ( ) ( )2
L
b x xT x f g h I x dxρ ω⎡ ⎤= + + +⎣ ⎦∫ (3.31)
46
Combining Eq. (3.31) with Eqs. (3.11) and (3.19), we can write the
total kinetic energy of the system as
( )
( ) ( )
2 2 2 2
0
2 2 2 2 2 2
1 ( ) ( )21 12 2
L
x x
M M M xM x yM y zM z
T x f g h I x dx
M f g h I I I
ρ ω
ω ω ω
⎡ ⎤= + + +⎣ ⎦
+ + + + + +
∫ (3.32)
where
( )( )
1 2
( ) ( )
( ) ( )( ) ( ) ( )
b P
x xb xP
x S x
I x I S x IS x H x l H x l
ρ ρ ρ= +
= +
= − − −
(3.33)
and ρp and Ixp are the mass per unit length and mass moment of inertia of the
piezoelectric actuator, respectively.
The total potential energy of the system can be written as:
2 22
20
1 ( ) ( )2
L wV EI x GJ x dxx x
θ⎡ ⎤⎛ ⎞∂ ∂⎛ ⎞⎢ ⎥= +⎜ ⎟ ⎜ ⎟∂ ∂⎝ ⎠⎢ ⎥⎝ ⎠⎣ ⎦∫ (3.34)
where E and G are elastic and shear moduli of the beam, respectively. EI(x)
and GJ(x) are given as
( )( )
( ) ( )
( ) ( )b P
b P
EI x EI S x EI
GJ x GJ S x GJ
= +
= + (3.35)
47
where EIb and EIP are the flexural rigidities of the beam and actuator,
respectively, and GJb and GJP are the torsional rigidities of the beam and
actuator, respectively.
In the Hamiltonian approach, the piezoelectric actuator control moment
MP and the damping effects are collected in the following virtual work
expression [74]
2
20 0 0
12
L L LPnc B T
M wW wdx C wdx C dxx t t
θδ δ δ δθ∂ ∂ ∂= + +
∂ ∂ ∂∫ ∫ ∫ (3.36)
where CB and CT are the damping coefficients in bending and torsion,
respectively. The actuator control moment is given as
31 01 ( ) ( ) ( ) ( ) ( )2P P b P PM bE d t t v t S x M v t S x= − + = (3.37)
Finally, we should take the variations, from the defined equations, in
order to achieve the equations of motion and boundary conditions which are
expressed as Eqs. (3.38) through (3.43) (the detailed derivation of equations
of motion and boundary conditions is given in Appendix A). It should be
noted that, in these equations damping has been ignored in both bending and
torsion.
2 2 221 1 3 3 32 2
22 2
2 2 2
( ) ( )
( )
x
P
w wx w s I xt t x x
MwEI xx x x
θρ⎛ ⎞ ⎛ ⎞∂ ∂ ∂
− Ω + Ω Ω − Ω +Ω⎜ ⎟ ⎜ ⎟∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠⎛ ⎞⎛ ⎞ ∂∂ ∂
+ =⎜ ⎟⎜ ⎟⎜ ⎟∂ ∂ ∂⎝ ⎠⎝ ⎠
(3.38)
48
2 2
32( ) ( ) 0xwI x GJ x
t t x x xθ θ⎛ ⎞ ⎛ ⎞∂ ∂ ∂ ∂⎛ ⎞+Ω − =⎜ ⎟ ⎜ ⎟⎜ ⎟∂ ∂ ∂ ∂ ∂⎝ ⎠⎝ ⎠⎝ ⎠
(3.39)
0 00
0 , 0 , 0x x
x
wwx
θ= =
=
∂= = =
∂ (3.40)
1 3 3 3 1 1
2 3 3
2 2 3
( ) ( ) ( )2 2
( ) 02
xx L x L
x L x L
w l l wI x M L wt x x
w l w wM EI xt t x x
θ
= =
= =
∂ ∂ ∂⎛ ⎞ ⎛ ⎞+Ω −Ω Ω + −Ω + +Ω + Ω⎜ ⎟ ⎜ ⎟∂ ∂ ∂⎝ ⎠ ⎝ ⎠
⎛ ⎞ ⎛ ⎞∂ ∂ ∂− + + =⎜ ⎟ ⎜ ⎟∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠
(3.41)
2 3
1 1 3 32 2
3 2
1 3 1 3 12
2
1 3 1 2
2 2 2
( )
( ) 0
xMx Lx L
yM yM L
x L x L
zMx L x L
l l w l w wM It t x t x
w w wI It x t t t x x
w wI EI xx x
θ
θψ θ θ θ θ
==
= =
= =
⎛ ⎞⎛ ⎞∂ ∂ ∂ ∂⎛ ⎞−Ω − + + +Ω −Ω Ω⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟∂ ∂ ∂ ∂ ∂⎝ ⎠⎝ ⎠⎝ ⎠
⎛ ⎞ ⎛ ⎞∂ ∂ ∂ ∂ ∂− +Ω +Ω + +Ω +Ω Ω⎜ ⎟ ⎜ ⎟∂ ∂ ∂ ∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠
⎛ ⎞∂ ∂⎛ ⎞− Ω +Ω Ω − =⎜ ⎟⎜ ⎟∂ ∂⎝ ⎠ ⎝ ⎠
(3.42)
2 2 2
3 1 3 12
2
1 3 3 ( ) 0
xM yMx Lx L x L
yMx Lx L
w w w wI It t x t x x x
w wI GJ xt x x x
θ θ θ
θθ θ
== =
==
⎛ ⎞ ⎛ ⎞∂ ∂ ∂ ∂ ∂− −Ω + +Ω +Ω Ω⎜ ⎟ ⎜ ⎟∂ ∂ ∂ ∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠
⎛ ⎞∂ ∂ ∂⎛ ⎞− +Ω +Ω Ω − =⎜ ⎟ ⎜ ⎟∂ ∂ ∂ ∂⎝ ⎠⎝ ⎠
(3.43)
Ω3, which is the primary base rotation velocity, is supposed to be
measured in these equations. As seen from Eqs. (3.38) and (3.39), the system
governing equations are coupled through the base rotation velocity Ω3. In the
absence of the base rotation, the governing equations become decoupled. In
order to analyze the error caused in the output of the gyroscope due to the
presence of the secondary base rotation, Ω1 has been considered.
49
Time Response
In order to perform the time-domain analysis on the system, we use the
Assumed Mode Model (AMM) expansion to truncate the original partial
differential governing equations of motion to ordinary differential equations
[54, 75, 76]. In this method, the lateral displacement w(x,t) and torsional
displacement θ(x,t) are assumed to be linear functions of assumed modes and
generalized coordinates, in this form
1
1
( , ) ( ) ( )
( , ) ( ) ( )
n
j jj
n
j jj
w x t x p t
x t x q t
φ
θ ψ
=
=
=
=
∑
∑ (3.44)
where φj(x) and ψj(x) are the mode shapes of a cantilever beam (with
no base rotation and rigid mass) under bending and torsion, respectively; and
pj(t) and qj(t) are the generalized coordinates for bending and torsion,
respectively. The mode shapes used for this case are
( ) ( )( ) sin sinh cos coshn n n n nx x x x xφ β β α β β= − − − (3.45)( )2 1
( ) sin2
n xx
Lπ
ψ⎡ ⎤+
= ⎢ ⎥⎣ ⎦
(3.46)
where nβ is the nth natural frequency of a cantilever beam and
sin sinhcos cosh
n nn
n n
L LL L
β βαβ β
−=
+ (3.47)
50
Although the cantilever beam is subjected to base rotation and it has a
rigid mass attached to its end, we have used the mode shapes for a regular
cantilever beam (with no base rotation and rigid mass). The mode shapes
given by Eqs. (3.45) and (3.46) satisfy the geometrical boundary conditions
for the beam (Eq. (3.40)) but they do not satisfy the natural boundary
conditions (Eqs. (3.41-3.43)). Hence, these mode shapes can be called as
admissible functions [77].
For two reasons, using these admissible functions does not
significantly affect the results: (i) the base rotations do not considerably
change the natural frequencies [1]. Hence, the mode shapes for the beam
without any base rotations can be safely used, and (ii) the mode shapes for the
cantilever beam with a rigid mass attached to its end are very complicated.
The main purpose of this chapter is to study the cross-axis effect in the beam,
hence for simplifying the analysis we have not considered the end mass. This
is a valid simplification since ( )xφ and ψ(x) are the admissible functions.
Using these admissible functions does not alter the results as the generalized
coordinates pj(t) and qj(t) in Eq. (3.44) change accordingly to give correct
response for w(x,t) and θ(x,t) .
The system governing equations can now be obtained by substituting
Eq. (3.44) into Eqs. (3.38) and (3.39) and ignoring damping in the system as
follow [1]:
51
( )2 21 1 3 1 1 3 1 1 1 1 3 1( ) ( ) ( )M p t C q t K D E p t G F+ Ω + + Ω − Ω + Ω Ω = (3.48)
2 2 3 2( ) ( ) ( ) 0M q t C p t K q t+ Ω + = (3.49)
where
[ ]
1 10 0
1 10 0
1 10 0
1 0 2 1
2
( ) ( ) ( ) , ( ) ( ) ( ) ,
( ) ( ) ( ) , ( ) ( ) ( ) ,
( ) ( ) ( ) , ( )
( ) ( ) ( ) , , 1, 2,...,
(
L L
ij i j ij x i j
L L
ij x i j ij x i j
L L
ij i j ij
ij P i i
ij x
M x x x dx C I x x x dx
K EI x x x dx D I x x x dx
E x x x dx G x xdx
F M v t l l i j n
M I
ρ φ φ φ ψ
φ φ φ φ
ρ φ φ ρ
φ φ
′= =
′′ ′′ ′′= =
= =
′ ′= − =
=
∫ ∫∫ ∫∫ ∫
20 0
2 0
1 2 1 2
) ( ) ( ) , ( ) ( ) ( ) ,
( ) ( ) ( ) , , 1, 2,...,
, ,..., , , ,...,
L L
i j ij x i j
L
ij i j
T Tn n
x x x dx C I x x x dx
K GJ x x x dx i j n
p p p p q q q q
ψ ψ ψ φ
ψ ψ
′=
′ ′= =
= =
∫ ∫∫
(3.50)
Equations of motion given by Eqs. (3.48) and (3.49) can be solved by
MATLAB for two modes and system parameters given in Table 3.1. In this
work two modes have been considered, however using more numbers of the
modes will lead to more accurate results.
Table 3.1: Physical parameters of the system [1].
Parameter Notation Value Beam length (m) L 0.15 Beam thickness (m) tb 0.8×10-3 Beam width (m) b 1.5×10-2 Mass per unit length (kg/m) ρb 3960×bt Beam elastic modulus (Gpa) E 70 Beam shear modulus (Gpa) G 30 End mass length (m) l 0.01 End mass width (m) bM 0.02 End mass height (m) hM 0.02
52
Cross-Axis Effects
The vibrating beam gyroscope is basically used in order to measure the
rotational rate around one of the axes. In practice, however, there are always
some secondary rotations present in the system. These secondary base
rotations can produce significant errors in measurement of the gyroscope
output. These errors are sometimes referred as ‘cross-axis’ effects. In this
section, we discuss the effects of these secondary vibrations on the output of
the gyroscope.
Figure 3.8 illustrates the output of the gyroscope when it is subjected to
primary (Ω3) as well as secondary (Ω1) base rotations of constant angular
velocity. To analyze the ‘cross-axis’ effects, the magnitude of the secondary
base rotation (Ω1) is varied from 0 to 0.5 rad/s while keeping a constant
piezoelectric excitation voltage (V=300 volts) and primary base rotation
(Ω3=20 rad/s). Figure 3.8(a) shows the desired gyroscopic output which is to
be measured. Figures 3.8 (b), (c) and (d) show the gyroscopic output from the
system when the base has secondary rotations of very small magnitude (0.05 –
0.5 rad/s). It can be seen that the gyroscopic output is increased significantly
(almost 40 times more) even for such a small magnitudes of secondary
rotation. This increased output could be interpreted as a gyroscopic output due
to primary base rotation. Such interpretation can produce errors in the
measurement of base velocity.
53
Figure 4.8: System response (torsional deflection θ(L,t)) to secondary base rotation; (a) Ω1 = 0 rad/s , (b) Ω1 = 0.05 rad/s , (c) Ω1 = 0.1 rad/s , and (d) Ω1 = 0.5 rad/s [1].
This is the most important drawback of the single beam gyroscopes,
and is an important factor to take into account in the design of the vibrating
beam gyroscope and effective control strategies have to be developed to
54
eliminate this secondary output (‘cross axis’ effects). The rocking-mass
gyroscope, which will be explained in the next chapter, will eliminate this
error and give a very precise measurement.
Summary
A detailed mathematical modeling of a vibrating beam gyroscope
undergoing flexural-torsional vibrations was presented, in this chapter. The
extended Hamilton’s principle was used in order to derive the governing
equations and boundary conditions. Two base rotations were considered for
the beam: (i) primary base rotation about z-axis, Ω3, and secondary base
rotation about x-axis Ω1. Moreover, the adverse effect of secondary base
rotations (cross-axis effects) on the gyroscopic output signals was discussed.
We concluded that the main drawback of the single beam gyroscopes is the
significant error caused by cross-axis effects.
55
Chapter 4
MODELING OF THE ROCKING-MASS
GYROSCOPE
Introduction
In the previous chapters, we discussed a single beam gyroscope which
comprises a cantilever beam with a tip mass undergoing coupled flexural-
torsional vibrations. As mentioned before, the main drawback of this type of
gyroscope is the difficulty in measuring the secondary torsional vibrations
induced due to the Coriolis force. In this chapter we will work on a rocking-
mass gyroscope, which can be considered as an extension of the single beam
gyroscope, and can overcome the limitations of the single beam gyroscope.
The governing equations of motion as well as the boundary conditions for
such a system, which comprises a set of four cantilever beams and a rocking
mass attached to them in the middle, will be derived in full detail. A
schematic of a rocking-mass gyroscope is shown in Figure 4.1.
56
Figure 4.1: Schematic of a rocking-mass gyroscope.
Principle of Operation
As shown in Figure 4.1, the rocking-mass gyroscope consists of four
beams attached to a rocking mass in the middle. In order to induce and sense
the vibrations in the beams, piezoelectric actuators are attached to beam 1 and
beam 3, and piezoelectric sensors are attached to beams 2 and 4. The primary
bending vibration is induced in the beams 1 and 3, by supplying a sinusoidal
voltage to the piezoelectric patches on them. Due to the bending of the beams
1 and 3, the rocking mass will rotate and produces a torsional vibration in
beams 2 and 4 as schematically shown in Figure 4.2. In presence of the base
angular rotation about the vertical axis, due to the Coriolis force a secondary
rocking motion of the mass is induced. As shown in Figure 4.3, bending is
induced in beams 2 and 4, as a result of the secondary rocking motion of the
mass. The amplitude of this bending vibration is proportional to the angular
57
velocity of the base. This secondary bending vibration, which can be
measured by the piezoelectric sensors placed on beams 2 and 4, gives the
angular velocity of the base.
Similar to the single beam gyroscope discussed in the previous chapter,
the rocking-mass gyroscope, uses the secondary induced vibrations to
determine the rate of rotation. It is usually difficult to measure the secondary
torsional vibrations for the single beam gyroscope, as their amplitude is
relatively small. This drawback is overcome by the rocking-mass gyroscope.
In the rocking-mass gyroscope, the torsional vibrations produced in two
beams are transferred to other two beams as bending vibrations which can be
easily sensed by placing sensors on the beams.
Figure 4.2: Primary rocking motion of the mass.
58
Figure 4.3: Secondary rocking motion of the mass.
Not many researchers have focused on the modeling and performance
evaluation of the ‘rocking-mass’ gyroscope. Tang and Gutierrez [78] dealt
with the fabrication and design of a rocking-mass gyroscope, but the operating
principle of the device was not discussed. Royle and Fox [79] presented an
analysis of the mechanics of an oscillatory rate gyroscope that is actuated and
sensed using thin piezoelectric actuators and sensors.
In this chapter, we derive the equations of motion and boundary
conditions governing the motion of a macro-scaled ‘rocking-mass’ gyroscope.
A method similar to the one developed for the single beam gyroscope in
chapter 3, is used here to derive the equations of motion, since the gyroscope
consists of four beams undergoing coupled flexural-torsional vibrations.
59
Governing Equations of Motion
The eight linear partial differential equations and their corresponding
boundary conditions governing the flexural-torsional motion of the four
beams of the gyroscope will be developed in this section, using the extended
Hamilton Principle. All beams are assumed to follow the Euler-Bernoulli
theory and accordingly the effects of warping and shear deformation are
neglected. The beams are considered to be slender. The poison effect is also
neglected. These assumptions usually hold for macro-scaled system not
micro-scaled ones.
We will follow the same methodology that discussed in chapter 3 to
develop the equations of motion and boundary conditions, since the rocking-
mass gyroscope similar to a single beam gyroscope. Figure 4.4 illustrates the
four coordinate systems that we will use. All the beams are identical with
mass per unit length ρb and thickness tb.
For this case, we consider a system with identical beams of equal
lengths denoted as Li (i=1,2,3,4). The length of the rocking-mass is denoted
by l. Bending and torsional deformations of the beams are denoted by wi
(i=1,2,3,4) and θi (i=1,2,3,4) respectively. In derivation of the two coupled
governing equations (one for bending and one for torsion) for each of the four
beams, we consider the rocking-mass to be attached to the first beam.
60
Figure 4.4: Top view of the rocking-mass gyroscope with the coordinate systems.
Similar to the single beam gyroscope, we need to find the total
translational kinetic energy, total rotational kinetic energy, total potential
energy and the non-conservative work done on the system, sequentially.
Translational Motion
The total translational kinetic energy of the four beams can be
determined by Extending Eq. (3.4) in the following form
2 2 2 2 2 21 1 1 1 1 2 2 2 20 0
2 2 2 2 2 23 3 3 3 4 4 4 40 0
1 1( ) ( )2 21 1( ) ( )2 2
L L
b b b
L L
b b
T f g h dx f g h dx
f g h dx f g h dx
ρ ρ
ρ ρ
= + + + + +
+ + + + + +
∫ ∫
∫ ∫(4.1)
where
61
1 2 3 4 4 2
1 1 1 1 3 2 2 3 3 3 1 3 3 4 4 3
31 2 41 2 2 2 3 4 4 2
0, 0, 0, ,, , , ,
, , , .
f f f f wg w s g s g w s g s
ww w wh h s h h st t t t
= = = = − Ω= − Ω + Ω = Ω = Ω + Ω = Ω
∂∂ ∂ ∂= = − Ω = = + Ω
∂ ∂ ∂ ∂
(4.2)
Also we can extend Eqs. (3.10) and (3.11), to define the translational
kinetic energy of the rocking mass, according to the following equation
( )2 2 21
12M M M MT M f g h= + + (4.3)
where
3 1 1 1 3 1 1
1 1
0,
,2 2
2
M
M L L
L LM
fl lg L w
w lht t
ψ
ψ
=
= Ω −Ω +Ω −Ω
∂ ∂= +
∂ ∂
(4.4)
and
1 1
1 1
11 1 1 1
1
( , ) , .L Lx Lx L
ww w x tx
ψ=
=
∂= =
∂ (4.5)
Rotational Motion
In order to describe the deformation of the beams and the rocking mass
from their original configuration, we use the Euler angle rotations. Again like
what we did for a single beam gyroscope, the two successive angles of
62
rotation for each beam are considered and denoted as ψi (i =1,2,3,4) and θi
(i=1,2,3,4). Similar to Eq. (3.17), if we assuming small angles of bending and
torsion and ignore nonlinear terms (which is fine for a macro-scaled system),
the angular velocity components of four beams and rocking mass are
expressed as
( )
( )
( )
1 11 1 3 1 1 1 1 1 3 1 1 1 1 3
2 22 3 2 2 2 3 2 2 2 2 3
3 33 1 3 3 3 1 3 3 3 3 3 1 3 3
44 3 4
, ,
, ,
, ,
x y z
x y z
x y z
x
t t
t t
t t
t
θ ψω ψ ω θψ θ ω ψ
θ ψω ψ ω θ ω θ
θ ψω ψ ω θψ θ ω ψ
θω ψ
∂ ∂⎛ ⎞ ⎛ ⎞= +Ω −Ω = +Ω +Ω = Ω +Ω⎜ ⎟ ⎜ ⎟∂ ∂⎝ ⎠ ⎝ ⎠∂ ∂⎛ ⎞ ⎛ ⎞= −Ω = +Ω +Ω = Ω +Ω⎜ ⎟ ⎜ ⎟∂ ∂⎝ ⎠ ⎝ ⎠∂ ∂⎛ ⎞ ⎛ ⎞= +Ω −Ω = +Ω +Ω = −Ω +Ω⎜ ⎟ ⎜ ⎟∂ ∂⎝ ⎠ ⎝ ⎠∂⎛= −Ω⎜ ∂⎝
( )44 2 3 4 4 2 4 3, ,y zt
ψω θ ω θ∂⎞ ⎛ ⎞= +Ω +Ω = Ω +Ω⎟ ⎜ ⎟∂⎠ ⎝ ⎠
(4.6)
where , 1, 2,3,4ii
i
w ix
ψ ∂= =∂
.
In this case again, we ignore the rotary inertia terms for the beams, i.e.,
Iybωy and Izbωz, therefore, the rotational kinetic energy of four beams can be
written as
1 2
3 4
2 22 1 1 2 20 0
2 23 3 4 40 0
1 12 21 12 2
L L
b xb x xb x
L L
xb x xb x
T I dx I dx
I dx I dx
ω ω
ω ω
⎛ ⎞ ⎛ ⎞= +⎜ ⎟ ⎜ ⎟⎝ ⎠ ⎝ ⎠⎛ ⎞ ⎛ ⎞+ +⎜ ⎟ ⎜ ⎟⎝ ⎠ ⎝ ⎠
∫ ∫
∫ ∫ (4.7)
And the rotational kinetic energy of the rocking mass is
63
( )2 2 22 1 1 1
12M xM x yM y zM zT I I Iω ω ω= + + (4.8)
where Ixb, Iyb, Izb and IxM, IyM, IzM are the mass moments of inertia of the beams
and rocking mass about the X, Y and Z axes, respectively.
Equations of Motion
The eight linear partial differential equations and boundary conditions
for the rocking-mass gyroscope are developed in this section. As mentioned in
Chapter 3, the extended Hamilton Principle for a dynamic system is expressed
as
2
1
0t
nctT V W dtδ δ δ− + =∫ (4.9)
where T is the total kinetic energy of the system, V is the total potential
energy and Wnc is the total non-conservative work done on the system.
Combining Eqs. (4.1), (4.3), (4.7) and (4.8), we can find the total kinetic
energy of the system as:
( )
1 2
3 4
2 2 2 2 2 2 2 21 1 1 1 1 1 1 2 2 2 2 2 2 20 0
2 2 2 2 2 2 2 23 3 3 3 3 3 3 4 4 4 4 4 4 40 0
2 2 2 2 21 1
1 1( ) ( ) ( ) ( )2 21 1( ) ( ) ( ) ( )2 21 12 2
L L
x x x x
L L
x x x x
M M M xM x yM y
T x f g h I x dx x f g h I x dx
x f g h I x dx x f g h I x dx
M f g h I I I
ρ ω ρ ω
ρ ω ρ ω
ω ω
⎡ ⎤ ⎡ ⎤= + + + + + + +⎣ ⎦ ⎣ ⎦
⎡ ⎤ ⎡ ⎤+ + + + + + + +⎣ ⎦ ⎣ ⎦
+ + + + + +
∫ ∫
∫ ∫
( )21zM zω
(4.10)
where
64
( )( )( )( )
1 2
3 4
1 2
3 4
( ) ( ) , ( )
( ) ( ) , ( )
( ) ( ) , ( )
( ) ( ) , ( )
b p b
b p b
x xb xp x xb
x xb xp x xb
x S x x
x S x x
I x I S x I I x I
I x I S x I I x I
ρ ρ ρ ρ ρ
ρ ρ ρ ρ ρ
= + =
= + =
= + =
= + =
(4.11)
In Eq. (4.11) ρp and Ixp are the mass per unit length and mass moment
of inertia of the piezoelectric actuator, respectively, and S(x) is used to define
the finite length of the piezoelectric and is given as
1 2( ) ( ) ( )S x H x l H x l= − − − (4.12)
where l1 and l2 are the starting and end position, respectively, of piezoelectric
actuator on beam 1.
There is no vertical motion for the rocking mass, therefore, the total
potential energy of the system consists of only the four beams which can be
stated as:
1 2
3
2 2 2 22 21 1 2 2
1 1 1 2 2 22 20 01 1 2 2
2 22 23 3 4
3 3 3 4203 3 4
1 1( ) ( ) ( ) ( )2 2
1 1( ) ( ) ( )2 2
L L
L
w wV EI x GJ x dx EI x GJ x dxx x x x
w wEI x GJ x dx EI xx x x
θ θ
θ
⎧ ⎫ ⎧ ⎫⎛ ⎞ ⎛ ⎞ ⎛ ⎞ ⎛ ⎞∂ ∂ ∂ ∂⎪ ⎪ ⎪ ⎪= + + +⎨ ⎬ ⎨ ⎬⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠ ⎝ ⎠ ⎝ ⎠⎪ ⎪ ⎪ ⎪⎩ ⎭ ⎩ ⎭⎧ ⎫⎛ ⎞ ⎛ ⎞∂ ∂ ∂⎪ ⎪+ + +⎨ ⎬⎜ ⎟ ⎜ ⎟∂ ∂ ∂⎝ ⎠ ⎝ ⎠⎪ ⎪⎩ ⎭
∫ ∫
∫4
2 2
44 420
4
( )L
GJ x dxxθ⎧ ⎫⎛ ⎞ ⎛ ⎞∂⎪ ⎪+⎨ ⎬⎜ ⎟ ⎜ ⎟∂⎝ ⎠ ⎝ ⎠⎪ ⎪⎩ ⎭
∫
(4.13)
Where EIb and EIp are the flexural rigidities of the beam and actuator, and GJb
and GJp are the torsional rigidities of the beam and actuator respectively. And
also
65
( )( )( )( )
1 2
3 4
1 2
3 4
( ) ( ) , ( )
( ) ( ) , ( )
( ) ( ) , ( )
( ) ( ) , ( )
b p b
b p b
x xb xp x xb
x xb xp x xb
x S x x
x S x x
I x I S x I I x I
I x I S x I I x I
ρ ρ ρ ρ ρ
ρ ρ ρ ρ ρ
= + =
= + =
= + =
= + =
(4.14)
The piezoelectricity and the effect of piezoelectric, as well as its
controlling moment were comprehensively discussed in chapter 3. Based on
the obtained equations, the piezoelectric actuator control moments Mp of
beams 1 and 3 (ignoring damping) are collected in the following virtual work
expression
1 32 2
1 1 3 32 20 01 3
L Lp pnc
M MW w dx w dx
x xδ δ δ
∂ ∂= +
∂ ∂∫ ∫ (4.15)
where
( )31 01 ( ) ( ) ( ) ( )2p p b p pM bE d t t v t S x M v t S x= − + = (4.16)
In these equations the parameters are defined similar to those defined
in chapter 3, but for beams 1 and 3. In other words, b is the width of the
beams 1 and 3; Ep and d31 are the elastic modulus and piezoelectric constant
of the actuators placed on the two beams, respectively; v(t) is the voltage
applied to the actuators and H(x) is the Heaviside function.
66
Finally, we need to take the variations of these equations in order to
derive the governing equations of motion and corresponding boundary
conditions for the rocking-mass gyroscope. Thus
2 2 221 1 1
1 1 1 1 1 3 1 3 32 21 1
2 221
12 2 21 1 1
( ) ( )
( )
x
P
w wx w s I xt t x x
w MEI xx x x
θρ⎛ ⎞⎛ ⎞∂ ∂ ∂
− Ω + Ω Ω − Ω +Ω⎜ ⎟⎜ ⎟∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠⎛ ⎞⎛ ⎞∂ ∂∂
+ =⎜ ⎟⎜ ⎟⎜ ⎟∂ ∂ ∂⎝ ⎠⎝ ⎠
(4.17)
2 21 1 1
1 3 121 1 1
( ) ( ) 0xwI x GJ x
t t x x xθ θ⎛ ⎞⎛ ⎞ ⎛ ⎞∂ ∂ ∂∂
+Ω − =⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟∂ ∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠⎝ ⎠ (4.18)
2 2 2 2222 2 2 2
2 2 2 2 3 3 22 2 2 22 2 2 2
( ) ( ) ( ) 0xw w wx w I x EI xt t x x x x
θρ⎛ ⎞⎛ ⎞ ⎛ ⎞⎛ ⎞∂ ∂ ∂ ∂∂
− Ω − Ω +Ω + =⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎜ ⎟∂ ∂ ∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠ ⎝ ⎠⎝ ⎠ (4.19)
2 22 2 2
2 3 222 2 2
( ) ( ) 0xwI x GJ x
t t x x xθ θ⎛ ⎞⎛ ⎞ ⎛ ⎞∂ ∂ ∂∂
+Ω − =⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟∂ ∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠⎝ ⎠ (4.20)
2 2 223 3 3
3 3 1 3 1 3 3 3 32 23 3
2 223
32 2 23 3 3
( ) ( )
( )
x
P
w wx w s I xt t x x
w MEI xx x x
θρ⎛ ⎞⎛ ⎞∂ ∂ ∂
− Ω + Ω Ω − Ω +Ω⎜ ⎟⎜ ⎟∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠⎛ ⎞⎛ ⎞∂ ∂∂
+ =⎜ ⎟⎜ ⎟⎜ ⎟∂ ∂ ∂⎝ ⎠⎝ ⎠
(4.21)
2 23 3 3
3 3 323 3 3
( ) ( ) 0xwI x GJ x
t t x x xθ θ⎛ ⎞⎛ ⎞ ⎛ ⎞∂ ∂ ∂∂
+Ω − =⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟∂ ∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠⎝ ⎠ (4.22)
2 2 2 2224 4 4 4
4 4 2 4 3 3 42 2 2 24 4 4 4
( ) ( ) ( ) 0xw w wx w I x EI xt t x x x x
θρ⎛ ⎞⎛ ⎞ ⎛ ⎞⎛ ⎞∂ ∂ ∂ ∂∂
− Ω − Ω +Ω + =⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎜ ⎟∂ ∂ ∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠ ⎝ ⎠⎝ ⎠ (4.23)
2 24 4 4
4 3 424 4 4
( ) ( ) 0xwI x GJ x
t t x x xθ θ⎛ ⎞⎛ ⎞ ⎛ ⎞∂ ∂ ∂∂
+Ω − =⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟∂ ∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠⎝ ⎠ (4.24)
and the boundary conditions:
67
1 1
1
11 10 0
1 0
0 , 0 , 0 ,x x
x
wwx
θ= =
=
∂= = =
∂ (4.25)
2 2
2
22 20 0
2 0
0 , 0 , 0,x x
x
wwx
θ= =
=
∂= = =
∂ (4.26)
3 3
3
33 30 0
3 0
0 , 0 , 0 ,x x
x
wwx
θ= =
=
∂= = =
∂ (4.27)
4 4
4
44 40 0
4 0
0 , 0 , 0,x x
x
wwx
θ= =
=
∂= = =
∂ (4.28)
1 1 1 1
1 1 1 1
1 1 11 3 1 3 1 3 1 1 1
1 1
2 3 31 1 1
12 2 31 1
( )2 2
( ) 02
x
x L x L
x L x L
w wl lI x M L wt x x
w w wlM EI xt t x x
θ
= =
= =
⎛ ⎞⎛ ⎞ ⎛ ⎞∂ ∂ ∂⎛ ⎞Ω +Ω −Ω + Ω −Ω + +Ω +⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟⎜ ⎟∂ ∂ ∂⎝ ⎠⎝ ⎠ ⎝ ⎠⎝ ⎠
⎛ ⎞ ⎛ ⎞∂ ∂ ∂− + + =⎜ ⎟ ⎜ ⎟∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠
(4.29)
( )1 1 1 1
1 11 1
1 1
2 31 1 1 1
1 3 1 32 21 1
3 21 11 1 1 1
1 3 1 1 1 1 3 121 1 1
11 1 3
1
2 2 2 xM
x L x L
yM yM L
x Lx L
zM
x L
w w wl l lM It t x t x
w w wI It x t t t x x
wI Ex
θ
θψ θ θ θ θ
= =
==
=
⎛ ⎞⎛ ⎞ ⎛ ⎞∂ ∂ ∂ ∂⎜ ⎟−Ω − + + Ω +Ω −Ω⎜ ⎟ ⎜ ⎟⎜ ∂ ∂ ∂ ⎟ ∂ ∂⎝ ⎠ ⎝ ⎠⎝ ⎠
⎛ ⎞∂ ⎛ ⎞∂ ∂ ∂ ∂− +Ω +Ω + Ω +Ω +Ω⎜ ⎟ ⎜ ⎟∂ ∂ ∂ ∂ ∂ ∂ ∂⎝ ⎠⎝ ⎠
⎛ ⎞∂− Ω Ω +Ω −⎜ ⎟∂⎝ ⎠
1 1
21
1 21
( ) 0x L
wI xx
=
⎛ ⎞∂=⎜ ⎟∂⎝ ⎠
(4.30)
1 11 1 1 1
1 1 1 1
2 2 21 1 1 1 1
3 1 1 3 1 121 1 1 1
21 1 1
3 1 1 3 1 11 1 1
( ) 0
zM yMx Lx L x L
yM
x L x L
w w w wI It t x t x x x
w w wI GJ xt x x x
θ θ θ
θ θ
== =
= =
⎛ ⎞ ⎛ ⎞∂ ∂ ∂ ∂ ∂− −Ω + +Ω +Ω Ω⎜ ⎟ ⎜ ⎟∂ ∂ ∂ ∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠
⎛ ⎞ ⎛ ⎞∂ ∂ ∂− Ω +Ω +Ω − =⎜ ⎟ ⎜ ⎟∂ ∂ ∂ ∂⎝ ⎠ ⎝ ⎠
(4.31)
If we look at the boundary conditions (4-29) through (4-31), we can
see that they are all written in terms of the first beam, however in reality the
mass is attached to all the four beams, not jut the first beam. The first
derivative and torsional displacement of beams 2, 3 and 4, are related to those
68
of the first beam. These relationships can be defined according to the
following 9 continuity equations.
( ) ( ) ( )2 1 1 2 1 2 1, , ,2L L L L L L Llw wδ δ δ ψ δ ψ δθ δθ δ ψ= + = = (4.32)
( ) ( ) ( )3 1 1 3 1 3 1, , ,L L L L L L Lw w lδ δ δ ψ δ ψ δ ψ δθ δθ= + = − = − (4.33)
( ) ( ) ( )4 1 1 4 1 4 1, , .2L L L L L L Llw wδ δ δ ψ δ ψ δθ δθ δ ψ= + = − = − (4.34)
Where
1 1 2 2
3 3 4 4
1 1 1 2 2 2
3 3 3 4 4 4
( , ) , ( , ) ,
( , ) , ( , ) .L Lx L x L
L Lx L x L
w w x t w w x t
w w x t w w x t= =
= =
= =
= = (4.35)
1 1 2 2
3 3 4 4
1 1 1 2 2 2
3 3 3 4 4 4
( , ) , ( , ) ,
( , ) , ( , ) .L Lx L x L
L Lx L x L
x t x t
x t x t
θ θ θ θ
θ θ θ θ= =
= =
= =
= = (4.36)
1 1 2 2 4 43 3
31 2 41 2 3 4
1 2 3 4
, , , .L L L Lx L x L x Lx L
ww w wx x x x
ψ ψ ψ ψ= = ==
∂∂ ∂ ∂= = = =∂ ∂ ∂ ∂
(4.37)
and δ( ) represents variation of corresponding terms.
Governing equations of motion, i.e. Eqs. (4.17-4.24), shows that the
two equations of each beam are coupled with each other through the base
rotation velocity Ω3 and each of the beams experiences a coupled flexural-
torsional vibration due to the gyroscopic terms such as 2
3i
xi
It xθ∂
Ω∂ ∂
and
69
2
3i
xi
wIt x∂
Ω∂ ∂
present in the equations of motion. Consequently, in absence of the
base rotation velocity, the two equations of each beam will be decoupled. This
is not the only coupling in the system. The second type of coupling is
expressed through the continuity equations (Eqs. (4.32–4.34)). The continuity
equations represent the relationship between the flexural and torsional
vibrations of beams 2, 3 and 4 in terms of the corresponding vibrations of
beam 1. As discussed before, bending of beams 1 and 3, induces a rocking
motion in the mass. In presence of the base rotation, this induced motion
produces a flexural vibrations in beams 2 and 4 and torsional vibrations in
beam 1 and 3. Since all the beams are connected to each other through the
rocking mass, the flexural and torsional deflections of each beam are also
dependent on each other.
To validate the mathematical modelling, if the beams 2, 3 and 4 are
neglected in the obtained equations, we will get to the governing equations of
a single beam gyroscope, which shows the accuracy of the attained equations.
It is noteworthy to mention, since in contrast with the single beam
gyroscopes, the piezoelectric sensors are placed on top of the beams 2 and 4,
not the sides, the secondary base rotation does not affect the measurement of
the primary base rotation, because the PZT sensors sense the bending of the
70
beams 2 and 4, not torsion. Therefore, the rocking-mass gyroscope is not
subject to the cross-axis effect.
Summary
In this chapter, we first introduced the rocking-mass gyroscope, which
consists of four beams with a finite mass attached to them in the middle, to
overcome the main drawback of the single beam gyroscopes. The principle of
operation of this kind of gyroscope was expressed and then compared with
that of the single beam gyroscope. Finally the eight linear partial differential
equations of motion as well as the boundary conditions were derived for the
71
Chapter 5
ROCKING-MASS GYROSCOPE ANALYSIS AND
NUMERICAL SIMULATIONS
In chapter 4, we developed the mathematical model of the system. We
also discussed different types of coupling available in the system, through the
governing equations of motion and the continuity equations. In this chapter
we mainly focus on the development of the frequency equations and
simulations of the gyroscopic system.
No work has been done on frequency analysis of the rocking-mass
gyroscope, in the past, however some work has been done on developing the
frequency equations of the cantilever beam undergoing coupled flexural-
torsional vibrations. Meirovitch [80-92] established some fundamentals on the
analysis of gyroscopic systems. Oguamanam [63] and Gokdag and Kopmaz
[64] developed the frequency equations of a cantilever beam with bending-
torsion vibrations. However, the effects of base rotation were ignored in their
work. Esmaeili et al. have worked on the problem of a cantilever beam
gyroscope with coupled flexural-flexural vibrations [66-68]. Frequency
equations of the system were developed as well as the gyroscopic effects
induced in the beam due to base rotations were demonstrated by simulating
the system governing equations. Vikrant et al [1] worked on a
72
piezoelectrically actuated flexural/torsional single beam gyroscope, and
performed some analysis on the system, in frequency and time domain.
This chapter essentially deals with the development of the frequency
equation and simulation of the equations of motion of the rocking-mass
gyroscope. The frequency equation is developed and the fundamental natural
frequency of the system and the corresponding mode shape will be found.
Finally a time-domain analysis will be performed to evaluate the effectiveness
of this kind of gyroscope.
Development of the Frequency Equation
An exact method is utilized to develop the frequency equation.
Assuming harmonic motion with frequency ω, the solutions of the equations
of motions can be represented in the following form
( , ) ( )1,2,3,4
( , ) ( )
i tn n
i tn n
w x t P x en
x t Q x e
ω
ωθ⎧ =
=⎨=⎩
(5.1)
where Pn(x) and Qn(x) are the amplitudes of the sinusoidally varying bending
and torsional displacements, respectively.
Substituting Eq. (5.1) into Eqs. (4.17–4.34) yields the following set of
equations of motion
73
0)()()()( 11111231113111
21 =−′′Ω+′Ω+ xPEIxPIxQiIxP IV
xx ωωρ (5.2)
0)()()( 1111131112
1 =″+′Ω− xQGJxPiIxQI xx ωω (5.3)
0)()()()()( 22222232223222
22
22 =−′′Ω+′Ω+Ω+ xPEIxPIxQiIxP IV
xx ωωρ (5.4)
0)()()( 2222232222
2 =″+′Ω− xQGJxPiIxQI xx ωω (5.5)
0)()()()( 33333233333333
23 =−′′Ω+′Ω+ xPEIxPIxQiIxP IV
xx ωωρ (5.6)
0)()()( 3333333332
3 =″+′Ω− xQGJxPiIxQI xx ωω (5.7)
0)()()()()( 44444234443444
22
24 =−′′Ω+′Ω+Ω+ xPEIxPIxQiIxP IV
xx ωωρ (5.8)
0)()()( 4444434442
4 =″+′Ω− xQGJxPiIxQI xx ωω (5.9)
and boundary conditions
,0)0(,0)0(,0)0( 111 ==′= QPP (5.10)
,0)0(,0)0(,0)0( 222 ==′= QPP (5.11)
,0)0(,0)0(,0)0( 333 ==′= QPP (5.12)
,0)0(,0)0(,0)0( 444 ==′= QPP (5.13)
2 2 21 1 1 3 1 1 1 1 1 1 3 1 1
1( ) ( ) ( ) ( ) 02 x xM P L Ml I P L EI P L iI Q Lω ω ω⎛ ⎞ ′ ′′′+ − Ω + + Ω =⎜ ⎟
⎝ ⎠ (5.14)
( )( )( )
2 2 2 2 21 1 3 1 1 1 1 1
3 1 1
2 ( ) 4 4 ( ) 4 ( )
4 ( ) 0
xM yM
xM yM
Ml P L Ml I I P L EI P L
i I I Q L
ω ω ω
ω
′ ′′+ − Ω + −
⎡ ⎤+ Ω − =⎣ ⎦
(5.15)
( ) ( )( )[ ] 0)()()( 1131111123
2 =′−Ω+′−Ω− LPIIiLQGJLQII yMxMyMxM ωω (5.16)
74
2 2 1 1 1 1( ) ( ) ( )2lP L P L P L′= + (5.17)
)()( 1122 LQLP =′ (5.18)
)()( 1122 LPLQ ′= (5.19)
3 3 1 1 1 1( ) ( ) ( )P L P L lP L′= + (5.20)
)()( 1133 LPLP ′−=′ (5.21)
)()( 1133 LQLQ −= (5.22)
4 4 1 1 1 1( ) ( ) ( )2lP L P L P L′= + (5.23)
)()( 1144 LQLP −=′ (5.24)
)()( 1144 LPLQ ′−= (5.25)
where 1−=i and ( )´ represents the derivative with respect to xi.
To solve Eqs. (5.2 – 5.9) an exact approach is utilized. We assume the
solutions of Eqs. (5.2) and (5.3) are in the following forms
1)( 11sxAexP = (5.26)
1)( 11sxeBxQ = (5.27)
where, “s” is an expression in terms of the system parameters and ω, and can
be found through Eq. (5-32).
Substituting (5.26) and (5.27) into (5.2) and (5.3), yields
041
223131
21 =−Ω+Ω+ AsEIAsIsBiIA xx ωωρ (5.28)
02131
21 =+Ω− BsGJsAiIBI xx ωω (5.29)
75
From (5.28) and (5.29) we have
41
2231
21
31
sEIsIsiI
BA
x
x
−Ω+Ω−
=ωρ
ω (5.30)
siIsGJI
BA
x
x
31
21
21
Ω+
=ω
ω (5.31)
Comparing (5.30) and (5.31), we can conclude
( )( ) ( )6,...,2,1,
0231
21
21
41
2231
21
=⇒=Ω++−Ω+
nssiIsGJIsEIsI
n
xxx ωωωρ(5.32)
Therefore, the solutions of (5.2) and (5.3) can be written in the
following form
∑=
=6
111
1)(n
xsn
neAxP (5.33)
∑=
=6
111
1)(n
xsnn
neAxQ α (5.34)
where
6,...,2,1,21
21
31 =+Ω
= nsGJI
siI
nx
nxn ω
ωα (5.35)
We can assume the solutions of Eqs. (5.4) and (5.5) are in the
following forms
76
2)( 22sxAexP = (5.36)
2)( 22sxeBxQ = (5.37)
Substituting (5.36) and (5.37) into (5.4) and (5.5), yields
0)( 42
223232
22
22 =−Ω+Ω+Ω+ AsEIAsIsBiIA xx ωωρ (5.38)
02232
22 =+Ω− BsGJsAiIBI xx ωω (5.39)
From (5.38) and (5.39) we have
42
2232
22
22
32
)( sEIsIsiI
BA
x
x
−Ω+Ω+Ω−
=ωρ
ω (5.40)
siIsGJI
BA
x
x
32
22
22
Ω+
=ω
ω (5.41)
Comparing (5.40) and (5.41), we can conclude
[ ][ ] ( )12,...,8,7,
0)( 232
22
22
42
2232
22
22
=⇒=Ω++−Ω+Ω+
nssiIsGJIsEIsI
n
xxx ωωωρ (5.42)
Therefore, the solutions of (5.4) and (5.5) can be written in the
following form
∑=
=12
722
2)(n
xsn
neAxP (5.43)
∑=
=12
722
2)(n
xsnn
neAxQ α (5.44)
77
where
12,...,8,7,22
22
32 =+Ω
= nsGJI
siI
x
xn ω
ωα (5.45)
The solutions of Eqs. (5.6) and (5.7) can be assumed in the following
forms
33 3( ) sxP x Ae= (5.46)
3)( 33sxeBxQ = (5.47)
Substituting (5.46) and (5.47) into (5.6) and (5.7), yields
043
223333
23 =−Ω+Ω+ AsEIAsIsBiIA xx ωωρ (5.48)
02333
23 =+Ω− BsGJsAiIBI xx ωω (5.49)
From (5.48) and (5.49) we have
43
2233
23
33
sEIsIsiI
BA
x
x
−Ω+Ω−
=ωρ
ω (5.50)
siIsGJI
BA
x
x
33
23
23
Ω+
=ω
ω (5.51)
Comparing (5.50) and (5.51), we can conclude
( )( ) ( )18,...,14,13,
0233
23
23
43
2233
23
=⇒=Ω++−Ω+
nssiIsGJIsEIsI
n
xxx ωωωρ(5.52)
78
Therefore, the solutions of (5.6) and (5.7) can be written in the
following form
∑=
=18
1333
3)(n
xsn
neAxP (5.53)
∑=
=18
1333
3)(n
xsnn
neAxQ α (5.54)
where
18,...,14,13,23
23
33 =+Ω
= nsGJI
siI
nx
nxn ω
ωα (5.55)
Finally, we take the solutions of Eqs. (5.8) and (5.9) in the following
forms
4)( 44sxAexP = (5.56)
4)( 44sxeBxQ = (5.57)
Substituting (5.56) and (5.57) into (5.8) and (5.9), yields
0)( 44
223434
22
24 =−Ω+Ω+Ω+ AsEIAsIsBiIA xx ωωρ (5.58)
02434
24 =+Ω− BsGJsAiIBI xx ωω (5.59)
From (5.58) and (5.59) we have
79
44
2234
22
24
34
)( sEIsIsiI
BA
x
x
−Ω+Ω+Ω−
=ωρ
ω (5.60)
siIsGJI
BA
x
x
34
24
24
Ω+
=ω
ω (5.61)
Comparing (5.60) and (5.61), we can conclude
[ ][ ] ( )24,...,20,19,
0)( 234
24
24
44
2234
22
24
=⇒=Ω++−Ω+Ω+
nssiIsGJIsEIsI
n
xxx ωωωρ (5.62)
Therefore, the solutions of (5.8) and (5.9) can be written in the
following form
∑=
=24
1944
4)(n
xsn
neAxP (5.63)
∑=
=24
1944
4)(n
xsnn
neAxQ α (5.64)
where
24,...,20,19,24
24
34 =+Ω
= nsGJI
siI
x
xn ω
ωα (5.65)
In summary the solutions of Eqs. (5.2)-(5.9) can be written as
∑∑==
==6
111
6
111
11 )(,)(n
xsnn
n
xsn
nn eAxQeAxP α (5.66)
80
∑∑==
==12
722
12
722
22 )(,)(n
xsnn
n
xsn
nn eAxQeAxP α (5.67)
∑∑==
==18
1333
18
1333
33 )(,)(n
xsnn
n
xsn
nn eAxQeAxP α (5.68)
∑∑==
==24
1944
24
1944
44 )(,)(n
xsnn
n
xsn
nn eAxQeAxP α (5.69)
Since each of these si’s is a very long expression in terms of the other
parameters, it is best to present the Maple code in Appendix C to obtain the
roots.
Substituting Eqs. (5.66)-(5.69) into Eqs. (5.10)-(5.25), we will have the
following system of equations. (Details of this system of equations can be
found in Appendix B.)
[ ] 01242424 =× ×× AC (5.70)
Therefore, the frequency equation is
[ ]( ) ( ) 0,,,det 232424 =ΩΩ=× ωGeometryfC (5.71)
The complete Maple code to get the characteristic equation of the
system is available in Appendix C. It is noteworthy that the usage of
traditional commands to calculate the determinant of the matrix “C” will not
work here; since “C” is a 24×24 matrix and each of its elements is a very long
expression in terms of the system parameters. So in the code provided, we
81
used the command “LUdecomp” to change the matrix to a lower-triangular
matrix and therefore, the product of the elements, located on the diagonal of
the new matrix will give the determinant.
Since the final determinant is a very long and complicated problem,
solving that equation and finding its roots needs very high computational cost
(The final expression is 5000 pages long).
Table 5.1: Physical parameters of the system. Parameter Notation Value Beam length (m)
1L 0.15 Beam length (m)
2L 0.15 Beam length (m)
3L 0.15 Beam length (m)
4L 0.15 Beam thickness (m)
1bt 0.8×10-3 Beam thickness (m)
2bt 0.8×10-3 Beam thickness (m)
3bt 0.8×10-3 Beam thickness (m)
4bt 0.8×10-3 Beam width (m)
1b 1.5×10-2 Beam width (m)
2b 1.5×10-2 Beam width (m)
3b 1.5×10-2 Beam width (m)
4b 1.5×10-2 Mass per unit length (kg/m)
1ρ 3960
Mass per unit length (kg/m) 2ρ 3960
Mass per unit length (kg/m) 3ρ 3960
Mass per unit length (kg/m) 4ρ 3960
Beam elastic modulus (Gpa) E 70 Beam shear modulus (Gpa) G 30 End mass length (m) l 0.01 End mass width (m) bM 0.02 End mass height (m) hM 0.02 Base rotation (rad/s) 3Ω 10 Secondary rotation (rad/s) 2Ω 5
82
For a system with specifications according to Table 5.1, the frequency
equation is solved and the fundamental frequency is found to be 131 rad/sec.
Finding the higher natural frequencies, which are of less importance in
compression to the fundamental one, needs a very strong computer with high
computational power.
Substituting s1 through s24 into Eqs. (5.66)-(5-69) for the fundamental
frequency, we will find the first mode shape of the system which corresponds
to its fundamental natural frequency. The first mode is illustrated in Figure
5.1. It can be seen that the first mode deals with the bounce of the rocking
mass.
Figure 5.1: First mode shape of the system.
83
Validation of the Utilized Method for Solving EVP
In order to validate the applied method for solving the Eigenvalue
problem (EVP), an analysis will be performed on a Timoshenko beam to
solve its EVP, since the governing equations of a Timoshenko beam are very
similar to the ones available in the rocking-mass gyroscope.
Frequency Analysis of a Timoshenko Beam
The Timoshenko beam is an extension of the Euler-Bernoulli beam in
which the effect of shear deformation and rotary inertia are included in the
governing equations. The equations governing the motion of a uniform
Timoshenko beam are [93]:
2 2
2 2
w wGA Ax x t
ψκ ρ⎛ ⎞∂ ∂ ∂
− =⎜ ⎟∂ ∂ ∂⎝ ⎠ (5-72)
2 2
2 2
wEI GA Ix x tψ ψκ ψ ρ∂ ∂ ∂⎛ ⎞+ − =⎜ ⎟∂ ∂ ∂⎝ ⎠
(5-73)
Nondimensional variables are introduced according to
* * * *, , ,w x tw x tL L T
ψ ψ= = = = (5-74)
Substitution of Equ. (5-74) into Equs. (5-71) and (5-72) and dropping
the *s sign from nondimensional variables leads to
84
2 2
2 2
w wx x t
ψ∂ ∂ ∂− =
∂ ∂ ∂ (5-75)
2 2
1 22 2
wx x tψ ψη ψ η∂ ∂ ∂
+ − =∂ ∂ ∂
(5-76)
where T has been chosen for convenience as
T LGρκ
= (5-77)
and
1 22 2,EI IGL A AL
η ηκ
= = (5-78)
A set of solutions can be considered for Equations (5-75) and (5-76) in
the following form
( , ) ( ). i tw x t P x e ω= (5-79)
( , ) ( ). i tx t Q x e ωψ = (5-80)
Substitution of Equs. (5-79) and (5-80) into Equs. (5-75) and (5-76)
leads to
2 0P Q Pω′′ ′− − = (5-81)
21 2(1 ) 0P Q Qη η ω′ ′′+ − + = (5-82)
85
To solve EVP, a set of solutions can be considered for Equations
(5-81) and (5-82) in the following form
sxP Ae= (5-83)
sxQ Be= (5-84)
Substitution of Equs. (5-83) and (5-84) into Equs. (5-81) and (5-82),
yields
2 2 0As Bs Aω− + = (5-85)
2 21 2 0As B s B Bη η ω+ − + = (5-86)
Therefore,
( )4 2 2 4 21 1 2 2 0
, 1, 2,3, 4n
s s
s n
η η η ω η ω ω⎡ ⎤+ + − =⎣ ⎦⇒ =
(5-87)
Finally, the exact solutions of the Equs. (5-81) and (5-82) can be
expressed as
4
1
ns xn
nP A e
=
= ∑ (5-88)
4
1
ns xn n
n
Q A eα=
=∑ (5-89)
where
86
2 2n
nn
ssωα +
= (5-90)
For a fixed-free beam (cantilever beam) the four boundary conditions
are
(0, ) 0 , (0, ) 0 , (1, ) (1, ) 0 , (1, ) 0ww t t t t tx x
ψψ ψ∂ ∂= = − = =
∂ ∂ (5-91)
Introducing Equs. (5-88) and (5-98) into Equs. (5-91), will give the
frequency equation, as follow
( ) ( ) ( ) ( )
[ ]
31 2 4
31 2 4
1
1 2 3 4 2
1 1 2 2 3 3 4 4 3
1 1 2 2 3 3 4 4 4
1 1 1 1
0ss s s
ss s s
C
AA
s e s e s e s e As e s e s e s e A
α α α αα α α αα α α α
⎡ ⎤ ⎡ ⎤⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ =⎢ ⎥ ⎢ ⎥− − − −⎢ ⎥ ⎢ ⎥
⎣ ⎦⎣ ⎦
(5-92)
Hence, the frequency equation (characteristic equation) is
[ ]( )det 0C = (5-93)
Solving Equ. (5-93), we will find the natural frequencies of the system.
It is noteworthy to mention that in the latest work done before this
research [93] the solutions of the Equs. (5-81) and (5-82) were chosen as
1 2 3 4cosh( ) sinh( ) cos( ) sin( )P A x A x A x A xμ μ υ υ= + + + (5-94)
87
2 2 2 2
1 2
2 2 2 2
3 4
sinh( ) cosh( )
sin( ) cos( )
Q A x A x
A x A x
ω μ ω μμ μμ μ
ω υ ω υυ υυ υ
⎛ ⎞ ⎛ ⎞+ += +⎜ ⎟ ⎜ ⎟
⎝ ⎠ ⎝ ⎠⎛ ⎞ ⎛ ⎞− −
+ +⎜ ⎟ ⎜ ⎟⎝ ⎠ ⎝ ⎠
(5-95)
where
( ) ( )1/2
22 4 21 2 1 2 1
1
1 42
μ η η ω η η ω η ωη
⎡ ⎤= − + + − +⎢ ⎥⎣ ⎦ (5-96)
( ) ( )1/2
22 4 21 2 1 2 1
1
1 42
υ η η ω η η ω η ωη
⎡ ⎤= + + − +⎢ ⎥⎣ ⎦ (5-97)
It should be noted, although the solutions given by Equs. (5-94) and
(5-95) are the most precise ones offered by that time, but substitution of these
equations into Equ. (5-81), yields
4 0Aε = (5-98)
where ε is a small number which is not equal to zero. However if we
substitute Equs. (5-88) and (5-89) into (5-81), we will get
0 0= (5-99)
This shows the precision of the applied method in this work.
88
The first three natural frequencies of the fixed-free Timoshenko beam
are presented in Table 5.2 for the utilized method and Reference [93] when
31 1.6 10η −= × and 4
2 6.25 10η −= × .
Table 5.2: The first three natural frequencies of the fixed-free Timoshenko beam
nω Mode number (n) Ref [93] Proposed Method
Percentage of error
1 0.1383 0.139920099 1.17 2 0.8417 0.8512000000 1.13 3 2.2586 0.2826000000 1.06
It can be seen that, the application of the offered method, yields to a
very exact calculation.
The first three mode shapes of a cantilever Timoshenko beam, found
through the exact method, is illustrated in Figure V-1.
(a) (b) (c)
Figure 5.2: Mode shapes of a cantilever Timoshenko beam (a) First mode (b) Second mode (c) Third mode
89
Time Response of the System
The time response of the system can be found, using the mode
superposition principle [94]. For this, the bending and torsional motion of
each beam will be assumed as:
1 1 1 1 1 1 1 1 1 11 1
( , ) ( ) ( ) , ( , ) ( ) ( )n n n nn n
w x t P x t x t Q x tψ θ ϕ∞ ∞
= =
= =∑ ∑ (5.100)
2 2 2 2 2 2 2 2 2 21 1
( , ) ( ) ( ) , ( , ) ( ) ( )n n n nn n
w x t P x t x t Q x tψ θ ϕ∞ ∞
= =
= =∑ ∑ (5.101)
3 3 3 3 3 3 3 3 3 31 1
( , ) ( ) ( ) , ( , ) ( ) ( )n n n nn n
w x t P x t x t Q x tψ θ ϕ∞ ∞
= =
= =∑ ∑ (5.102)
4 4 4 4 4 4 4 4 4 41 1
( , ) ( ) ( ) , ( , ) ( ) ( )n n n nn n
w x t P x t x t Q x tψ θ ϕ∞ ∞
= =
= =∑ ∑ (5.103)
Where win (i=1,2,3,4) is the nth normal mode of each beam.
Substituting Eqs. (5.100)-(5.103) into equations of motion of the system, i.e.
Eqs. (4.17)-(4.24) and integrating over the intervals [0,Li], we will have a set
of eight ordinary differential equations in terms of ψi and φi (i=1,2,3,4) which
will be solved by MATLAB© to find the time response of the system.
Since defining the exact frequency equation and the exact mode shape
is hard to do, usually assumed-modes (comparison or admissible functions
are used instead of the normal modes in determining the time response and
therefore in that case it is need to consider enough number of modes [77] to
90
find the exact response. But in the present work, since we have found the
exact equation for the mode shape of the system, using the fundamental mode
shape will give us a very good and acceptable time response, however using
more numbers of the modes will lead to more accurate results.
Figure 5.3 (a) shows the bending of the beam 1 (drive direction), for
different cases of Ω3=0, Ω3=25 rad/s and Ω3=50 rad/s. The bending of the
second beam (sense direction) for these three cases are shown in Figures 5.3
(b), 5.3 (c) and 5.3 (d) respectively. This figure shows that, as the magnitude
of the base rotation increases, due to the corresponding increase in the
gyroscopic coupling, vibrations of the second beam also increase
proportionally. Hence, it can be concluded that the amplitudes of the
vibrations of the second and fourth beams are directly proportional to the
magnitude of the base rotation. This is an important conclusion as it shows the
effectiveness of this type of gyroscope as a device for measuring base angular
velocity.
91
(a)
(b)
92
(c)
(d)
Figure 5.2: System response (a) bending deflection w1 and bending deflection w2 to different base rotation rates: (b) Ω3=0, (c) Ω3=25 rad/s and (d) Ω3=50 rad/s
93
Summary
In this chapter, the exact frequency equation of the system was
developed in full detail, using an exact approach. The fundamental frequency
of the system and the corresponding mode shape was found as well.
Furthermore, by simulating the system, through mode superposition, the
presence of the gyroscopic coupling present in the system was validated. It
was shown that the gyroscopic output from the system is directly proportional
to the base rotation rate. So by sensing the output we would be able to
measure the rate of the base rotation.
94
Chapter 6
CONCLUSIONS AND FUTURE WORK
The primary goal of this research was to develop and analyze a new
type of vibrating gyroscope, called rocking-mass gyroscope. The operating
principle of the gyroscope, which consists of a set of four cantilever beams
with a rigid mass attached to them in the middle while subjected to base
rotations, was presented. First by considering a single beam gyroscope and the
cross-axis effect in them, the main drawback of this type of gyroscopes was
pointed out. For the rocking-mass gyroscope, a thorough analysis was carried
out in order to obtain the governing equations of motion of the system in their
most general form. The analysis was further extended to obtain the frequency
equation. The system response was also obtained for different conditions. The
effectiveness of the rocking-mass gyroscope was analyzed by simulating the
equations using mode superposition method. The results demonstrated that
this type of gyroscope can be used for sensing the rotational motions
accurately. An experimental setup can be used to validate the accuracy of the
simulations.
95
Recommendations for Future Work
A thorough design optimization can be performed on the system, using
powerful computing facilities, to find the optimum parameters such that the
best system performance is obtained. It is also a good idea to build an
experimental setup to study the system in practice, and validate the theoretical
studies in another way too.
The ultimate goal of this research could be to develop a new type of
MEMS vibrating gyroscope. As described in this thesis, MEMS gyroscopes
have tremendous potential for being used in many applications. To this end,
the simulations have to be extended to a micro-scaled gyroscope. In the
present work, the focus was mainly on a macro-scaled gyroscope. Extending
these simulations to a micro-scaled gyroscope is an important step in
determining the effectiveness of a similar MEMS gyroscope.
96
APPENDICES
97
APPENDIX A Detailed Derivation of Equations of Motion for a
Single Beam Gyroscope
The extended Hamilton’s Principle is given as
(A.1)
Using the expressions for kinetic energy, potential energy and virtual
work (Eqs. (3.32), (3.34) and (3.36)), and ignoring damping, we can express
the different components of Eq. (A-1) as
Kinetic energy
(A.2)
Substituting values of f, g, h, ωx, ωy and ωz from Eqs. (3.3) and (3.16),
we can simplify the above expression as follows
(A.2)
98
(A.3)
Integrating by parts, we get
(A.4)
Simplifying and combining similar terms we get
(A.5)
99
Total kinetic energy of the end mass is given as
(A.6)
Taking the variation of the above expression yields,
(A.7)
Substituting values of f, g, h, ωx, ωy and ωz from Eqs. (3.10) and (3.16),
we can simplify the above expression as follows
(A.8)
Potential energy
(A.9)
100
Simplifying and combining similar terms we get,
(A.10)
Virtual work is represented as
(A.11)
where [69]
(A.12)
Using Eqs. (A-1 to A-11) and taking into account the fact that
, , Lw wδ δθ δ and Lwxδ∂
∂could have any arbitrary values; the coefficients of
these terms in Hamilton’s equation must vanish. Hence, after substituting
values of f, g, h, ωx, ωy and ωz and ignoring rotary inertia for the beam, the
equations of motion and boundary conditions can be obtained as given in Eqs.
(3.38-3.43) [91].
101
APPENDIX B Frequency Equation
102
1 2 3 4 5 6
1 2 3 4 5 6
7 8 9 10 11 12
7 8 9 10 11 12
1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 0 00 0
s s s s s s
s s s s s s
α α α α α α
α α α α α α
13 14 15 16 17 18
13 14 15 16 17 18
19 20 21 22 23 24
19 20 21 22 23 24
131 132 133 134 135 136
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
s s s s s s
s s s s s s
C C C C C C
α α α α α α
α α α α α α
141 142 143 144 145 146
151 152 153 154 155 156
161 162 163 164 165 166 167 168 169 1610 1611 1612
171 172 173 174 175 176 177 178 179 1710 171
0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0
C C C C C CC C C C C CC C C C C C D D D D D DC C C C C C D D D D D 1 1712
181 182 183 184 185 186 187 188 189 1810 1811 1812
191 192 193 194 195 196 1913 1914 1915 1916 1917 1918
201 202 203 204 205 206 2013 2014 2015 2016 20
0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0
DC C C C C C D D D D D DC C C C C C E E E E E EC C C C C C E E E E E 17 2018
211 212 213 214 215 216 2113 2114 2115 2116 2117 2118
221 222 223 224 225 226 2219 2220 2221 2222 2223 2224
231 232 233 234 235 236 2319 2320 2321 232
0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0
EC C C C C C E E E E E EC C C C C C F F F F F FC C C C C C F F F F
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
1
2 2323 2324
241 242 243 244 245 246 2419 2420 2421 2422 2423 24240 0 0 0 0 0 0 0 0 0 0 0
AAAAAAAAAAAAAAAAAAA
F FC C C C C C F F F F F F
⎡ ⎤⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎢ ⎥⎣ ⎦
9
20
21
22
23
24
13
000000000000000000000000
n
AAAAA
C Mω
⎡ ⎤ ⎡ ⎤⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥
=⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎣ ⎦⎣ ⎦
= [ ] [ ] [ ]1 1 1 1
1 1 1
2 2 2 31 3 1 1 3 16 19 22
2 2 2 2 2 214 3 1 3 17 20 23
(0.5. ) ; 1 0.5 ; 1 ; 1 0.5
2 ( 4 4 ) 4 4 ( ) ; ; ;
n n n n
n n n
s L s L s L s Lx n n x n n n n n n n
s L s L s Ln xM yM n n n xM yM n n n n
Ml I s EI s iI e C ls e C ls e C ls e
C Ml Ml I I s EI s i I I e C e C s e C
ω ω α
ω ω ω α ω α
⎡ ⎤+ − Ω + + Ω = + = + = +⎣ ⎦⎡ ⎤= + − Ω + − + Ω − = =⎣ ⎦
1
1 1 1 1
2 2 2
3 3
2 215 3 1 3 18 21 24
16 17 18
19 20 21
1,2,...,6
( ) ( ) ; ; ;
; ; , 7,8,...,12
; ;
n
n n n n
n n n
n n
s Ln n
s L s L s L s Ln xM yM n n n xM yM n n n n n n n
s L s L s Ln n n n n
s L s Ln n n
e n
C I I GJ s i I I s e C s e C e C s e
D e D s e D e n
E e E s e E
α
ω α α ω α
α
⎫⎪⎪= =⎬⎪
⎡ ⎤= − Ω − + Ω − = = = ⎪⎣ ⎦ ⎭= − = − = − =
= − = 3
4 4 422 23 24
, 13,14,...,18
; ; , 19,20,..., 24
n
n n n
s Ln n
s L s L s Ln n n n n
e n
F e F s e F e n
α
α
= =
= − = = =
103
1 2 3 4 5 6
1 2 3 4 5 6
7 8 9 10 11 12
7 8 9 10 11 12
1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 0 00 0
s s s s s s
s s s s s s
α α α α α α
α α α α α α
13 14 15 16 17 18
13 14 15 16 17 18
19 20 21 22 23 24
19 20 21 22 23 24
131 132 133 134 135 136
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
s s s s s s
s s s s s s
C C C C C C
α α α α α α
α α α α α α
141 142 143 144 145 146
151 152 153 154 155 156
161 162 163 164 165 166 167 168 169 1610 1611 1612
171 172 173 174 175 176 177 178 179 1710 171
0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0
C C C C C CC C C C C CC C C C C C D D D D D DC C C C C C D D D D D 1 1712
181 182 183 184 185 186 187 188 189 1810 1811 1812
191 192 193 194 195 196 1913 1914 1915 1916 1917 1918
201 202 203 204 205 206 2013 2014 2015 2016 20
0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0
DC C C C C C D D D D D DC C C C C C E E E E E EC C C C C C E E E E E 17 2018
211 212 213 214 215 216 2113 2114 2115 2116 2117 2118
221 222 223 224 225 226 2219 2220 2221 2222 2223 2224
231 232 233 234 235 236 2319 2320 2321 232
0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0
EC C C C C C E E E E E EC C C C C C F F F F F FC C C C C C F F F F 2 2323 2324
241 242 243 244 245 246 2419 2420 2421 2422 2423 2424
0
0 0 0 0 0 0 0 0 0 0 0 0F F
C C C C C C F F F F F F
=
104
APPENDIX C Maple Code to Develop Frequency Equation
105
106
107
108
109
110
REFERENCES
[1] Bhadbhade V., Jalili N. and Mahmoodi N, “A novel piezoelectrically
actuated flexural/torsional vibrating beam gyroscope”, Journal of Sound and
Vibration, 311, pp. 1305-1324, 2008.
[2] Post E. J., “Sagnac effect,” Review of Modern Physics, 39, pp. 475-493,
1967.
[3] Armenise M. N., Ciminelli C., De Leonardis F., Diana R., Passaro F.,
Peluso F., “Gyroscope technologies for space applications,” 4th Round Table
on Micro/Nano Technologies for Space, ESTEC, Noordwijk, Netherlands,
May 2003.
[4] Reunert M., Yoshida B., “Fibre-optic gyroscopes: a new sensor for
robotics and autonomous vehicles,” Sensor Review, 16(1), pp. 32-34, 1996.
[5] Yazdi N., Ayazi F., Najafi K., “Micromachined inertial sensors,”
Proceedings of the IEEE, 86(8), pp. 1640-1659, 1998.
[6] Acar C., “Robust Micromachined Vibratory Gyroscopes”, PhD Thesis,
University of California, Irvine, 2004.
[7] Barbour N., Schmidt G., “Inertial sensor technology trends,” IEEE
Sensors Journal, 1(4), pp. 332-339, 2001.
111
[8] V. Apostolyuk “Theory and design of micromechanical vibratory
gyroscopes”, MEMS/NEMS Handbook, Springer, Vol.1, pp.173-195, 2006.
[9] H. Lefevre, The Fiber-Optic Gyroscope. Norwood, MA: Artech House,
1993.
[10] Lefevre H., “The fiber optic gyroscope,” Norwood, MA, Artech House,
1993.
[11] Lawrence A., “Modern inertial technology: Navigation, Guidance and
Control,” New York: Springer-Verlag, 1993.
[12] Nasiri S., “A critical review of MEMS gyroscope technology and
commercialization status,” A Report for InvenSense, Santa Clara, CA
[13] http://www.sensorsportal.com/HTML/DIGEST/may_06/MEMS_Gyroscope_market.htm
[14] J. Bernstein, S. Cho, A. T. King, A. Kourepenis, P. Maciel, and M.
Weinberg, “A micromachined comb-drive tuning fork rate gyroscope,” in
Proc. IEEE Micro Electro Mechanical Systems Workshop (MEMS’93), Fort
Lauderdale, FL, Feb. 1993, pp. 143–148.
[15] M. Hashimoto, C. Cabuz, K. Minami, and M. Esashi, “Silicon resonant
angular rate sensor using electromagnetic excitation and capacitive detection,”
J. Micromech. Microeng., pp. 219–225, 1995.
112
[16] M. Lutz, W. Golderer, J. Gerstenmeier, J. Marek, B. Maihofer, S.
Mahler, H. Munzel, and U. Bischof, “A precision yaw rate sensor in silicon
micromachining,” in Tech. Dig. 9th Int. Conf. Solid-State Sensors and
Actuators (Transducers’97), Chicago, IL, June 1997, pp. 847–850.
[17] R. Voss, K. Bauer, W. Ficker, T. Gleissner, W. Kupke, M. Rose, S.
Sassen, J. Schalk, H. Seidel, and E. Stenzel, “Silicon angular rate sensor for
automotive applications with piezoelectric drive and piezoresistive read-out,”
in Tech. Dig. 9th Int. Conf. Solid-State Sensors and Actuators
(Transducers’97), Chicago, IL, June 1997, pp. 879–882.
[18] K. Maenaka and T. Shiozawa, “A study of silicon angular rate sensors
using anisotropic etching technology,” Sensors Actuators A, vol. 43, pp. 72–
77, 1994.
[19] M. W. Putty and K. Najafi, “A micromachined vibrating ring gyroscope,”
in Tech. Dig. Solid-State Sensor and Actuator Workshop, Hilton Head Island,
SC, June 1994, pp. 213–220.
[20] Painter C.C., “Micromachined vibratory gyroscopes with imperfections,”
Ph.D. Dissertation, University of California, Irvine, 2005.
[21] Ayazi F., Najafi K., “A HARPSS polysilicon vibrating ring gyroscope,”
Journal of Microelectromechanical Systems, 10(2), pp. 169-179, 2001.
113
[22] Greiff P., Boxenhorn B., King T., Niles L., “Silicon monolithic
micromechanical gyroscope,” 6th International Conference on Solid-State
Sensors and Actuators (Transducers’ 91), San Francisco, CA, June 1991, pp.
966-968.
[23] Bernstein J., Cho S., King A. T., Kourepenis A., Maciel P., Weinberg M.,
“ A micromachined comb-drive tuning fork rate gyroscope,” Proceedings of
IEEE Micro Electro Mechanical Systems Workshop (MEMS’93), Fort
Lauderdale, FL, February 1993, pp. 143-148.
[24] Gianchandani Y., Najafi K., “A bulk silicon dissolved wafer process for
Microelectromechanical systems,” Journal of Microelectromechanical
Systems, pp. 77-85, June1992.
[25] Tang W. C., Lin M. G., Howe R. T., “Electrostatically balanced comb
drive for controlled levitation,” Technical Digest, Solid-State Sensors and
Actuators Workshop, Hilton Head, SC, June 1990, pp. 23-27.
[26] Weinberg M., Bernstein J., Cho S., King A. T., Kourepenis A., Ward P.,
Sohn J “A micromachined comb-drive tuning fork gyroscope for commercial
applications,” Proceedings of Sensor Expo, Cleveland, OH, pp. 187-193,
1994.
114
[27] Hashimoto M., Cabuz C., Minami K., Esashi M., “Silicon resonant
angular rate sensor using electromagnetic excitation and capacitive detection,”
Journal of Micromechanics and Microengineering, 5, pp. 219-225, 1995.
[28] Lutz M., Golderer W., Gerstenmeier J., Marek J., Maihofer S., Mahler S.,
Munzel H., Bischof U., “ A precision yaw rate sensor in silicon
micromachining,” Technical Digest, 9th International Conference on Solid-
State Sensors and Actuators (Transducers’ 97), Chicago, IL, June 1997, pp.
847-850.
[29] Paoletti F., Gretillat M.A., de Rooij N. F., “A silicon micromachined
vibrating gyroscope with piezoresistive detection and electromagnetic
excitation,” Proceedings of IEEE Micro Electro Mechanical Systems
Workshop (MEMS’96), San Diego, CA, 1996, pp. 162-167.
[30] Voss R., Bauer K., Ficker W., Gleissner T., Kupke W., Rose M., Sassen
S., Schalk J., Stenzel E., “Silicon angular rate sensor for automotive
applications with piezoelectric drive and piezoresistive read-out,” Technical
Digest, 9th International Conference on Solid-State Sensors and Actuators
(Transducers’ 97), Chicago, IL, June 1997, pp. 879-882.
[31] Park K. Y., Lee C. W., Oh Y. S., Cho Y. H., “Laterally oscillated and
forcebalanced micro vibratory rate gyroscope supported by fish hook shape
115
springs,” Proc. IEEE Micro Electro Mechanical Systems Workshop
(MEMS’97), Japan, pp. 494–499, 1997.
[32] Clark W. A., Howe R. T., Horowitz R., “Surface micromachined z -axis
vibratory rate gyroscope,” Tech. Dig. Solid-State Sensor and Actuator
Workshop, Hilton Head Island, SC, June 1996, pp. 283–287.
[33] Juneau T., Pisano A. P., “Micromachined dual input axis angular rate
sensor,” Tech. Dig. Solid-State Sensor & ActuatorWorkshop, Hilton Head
Island, SC, June 1996, pp. 299–302.
[34] Oh Y., Lee B., Baek S., Kim H., Kim J., Kang S., Song C., “A
surfacemicromachined tunable vibratory gyroscope,” Proc. IEEE Micro
Electro Mechanical Systems Workshop (MEMS’97), Japan, pp. 272–277,
1997.
[35] An S., Oh Y. S., Lee B. L., Park K. Y., Kang S. J., Choi S. O., Go Y. I.,
Song C. M., “Dual-axis microgyroscope with closed-loop detection,” Proc.
IEEE Micro Electro Mechanical Systems Workshop (MEMS’98), Heidelberg,
Germany, February. 1998, pp. 328–333.
[36] Juneau T., Pisano A. P., Smith J. H., “Dual axis operation of a
micromachined rate gyroscope,” Tech. Dig. 9th Int. Conf. Solid-State Sensors
and Actuators (Transducers’97), Chicago, IL, June 1997, pp. 883–886.
116
[37] Tang T. K., Gutierrez R. C., Wilcox J. Z., Stell C., Vorperian V., Calvet
R., Li W. J., Charkaborty I., Bartman R., Kaiser W. J., “Silicon bulk
micromachined vibratory gyroscope,” Tech. Dig. Solid-State Sensor and
Actuator Workshop, Hilton Head Island, SC, June 1996, pp. 288–293.
[38] Tang T. K., Gutierrez R. C., Stell C. B., Vorperian V., Arakaki G. A.,
Rice J. T., Li W. J., Chakraborty I., Shcheglov K., Wilcox J. Z., Kaiser W. J.,
“A packaged silicon MEMS vibratory gyroscope for microspacecraft,” Proc.
IEEE Micro Electro Mechanical Systems Workshop (MEMS’97), Japan, pp.
500–505, 1997.
[39] Putty M. W., Najafi K., “A micromachined vibrating ring gyroscope,”
Tech. Dig. Solid-State Sensor and Actuator Workshop, Hilton Head Island,
SC, June 1994, pp. 213–220.
[40] Ayazi F., Najafi K., “Design and fabrication of a high performance
polysilicon vibrating ring gyroscope,” Proc. IEEE Micro Electro Mechanical
Systems Workshop (MEMS’98), Heidelberg, Germany, February 1998, pp.
621–626.
[41] Putty M., “A micromachined vibrating ring gyroscope,” Ph.D.
Dissertation, University of Michigan, March 1995.
117
[42] Jiang, X., J. Seeger, M. Kraft and B.E. Boser, “A monolithic surface
micromachined Z-axis gyroscope with digital output”, 2000 Symposium on
VLSI Circuits, Honolulu, HI, USA, pp.16-19, June 2000.
[43] Park S., Horowitz R., “Adaptive control for z -axis MEMS gyroscope,”
Proceddeings of the American Control Conference, Arlington, VA, 2001, pp.
1223-1228.
[44] Sun X., Horowitz R., Komvopoulos K., “Stability and resolution analysis
of a phase-locked loop natural frequency tracking system for MEMS fatigue
testing,” Journal of Dynamic Systems, Measurement and Control, 124(4), pp.
599–605, 2002.
[45] Leland R. P., “Lyapunov based adaptive control of a MEMS gyroscope,”
Proceedings of the American Control Conference, Anchorage, AK, 2002, pp.
3765-3770.
[46] Leland R. P., “Adaptive mode tuning of vibrational gyroscopes,” IEEE
Transactions on Control Systems Technology, 11(2), pp. 242-247, 2003.
[47] M’Closkey R. T., Vakakis A., Gutierrez R., “Mode localization induced
by a nonlinear control loop,” Nonlinear Dynamics., 25(1–3), pp. 221–236,
2001.
118
[48] Leland R. P., Lipkin Y., Highsmith A., “Adaptive oscillator control for a
vibrational gyroscope,” Proceedings of the American Control Conference,
Denver, 2003, CO, pp. 3347–3352.
[49] Clark W. A., “Micromachined Vibratory Rate Gyroscopes,” Doctoral
Thesis, University of California at Berkeley, CA, 1997.
[50] Ward P., “Electronics for Coriolis Force and Other Sensors,” IEEE U.S.
Patent. 5 672 949, 1997.
[51] Ljung P. B., “Micromachined Gyroscope with Integrated Electronics,”
Doctoral Dissertation, University of California at Berkeley, CA, 1997.
[52] Chang S., Chia M., Castillo-Borelley P., Higdon W., Jiang Q., Johnson
J., Obedier L., Putty M., Shi Q., Sparks D., Zarabadi S., “An electroformed
CMOS integrated angular rate sensor,” Sensors and Actuators A, 66, pp. 138–
143, 1998.
[53] Oboe R., Antonello R., Lasalandra E., Durante G. S., Prandi L., “Control
of z–axis MEMS vibrational gyroscope,” IEEE/ASME Transactions on
Mechatonics, 10 (4), pp. 364-370, 2005.
[54] Esmaeilli, M., Jalili, N., Durali, M., “Dynamic modeling and
performance evaluation of a vibrating microgyroscope under general support
motion”, Journal of Sound and Vibration, Vol. 301, pp. 146-164, 2007.
119
[55] Timoshenko S., Young D. H., “Vibration problems in engineering,” New
York: Van Nostrand, 1955.
[56] Dokumaci E., “An exact solution for coupled bending and torsion
vibrations of uniform beams having single cross- sectional symmetry,”
Journal of Sound and Vibration, 119, pp. 443-449, 1987.
[57] Bercin A. N., Tanaka M., “Coupled flexural-torsional vibrations of
Timoshenko beams,” Journal of Sound and Vibration, 207(1), pp. 47-59,
1997.
[58] Banerjee J. R., “Explicit frequency equation and mode shapes of a
cantilever beam coupled in bending and torsion,” Journal of Sound and
Vibration, 224(2), pp. 267-281, 1999.
[59] Goel R. P., “Vibrations of a beam carrying a concentrated mass,” Journal
of Applied Mechanics, 40, pp. 421-422, 1974.
[60] Laura P. A. A., Pombo J. L., Susemihl E. A., “A note on the vibrations of
a clamped-free beam with a mass at the free end,” Journal of Sound and
Vibration, 37, pp. 161-168, 1974.
[61] Bhat R., Wagner H., “Natural frequencies of a uniform cantilever with a
tip mass slender in the axial direction,” Journal of Sound and Vibration, 45,
pp. 304-307, 1976.
120
[62] Kirk C. L., Weidemann S. M., “Natural frequencies and mode shapes of
a free-free beam with large end masses,” Journal of Sound and Vibration,
254(5), pp. 939-949, 2002.
[63] Oguamanam D. C. D., “Free vibration of beams with finite mass rigid tip
load and flexural-torsional coupling,, International Journal of Mechanical
Sciences, 45, pp. 963-979, 2003.
[64] Gokdag H., Kopmaz O., “Coupled bending and torsional vibrations of a
beam with tip and in-span attachments,” Journal of Sound and Vibration, 287,
pp. 591-610, 2005.
[65] Salarie, H., Ghorashi, M., “Free vibration of Timoshenko beam with
finite mass rigid tip load and flexural–torsional coupling”, International
Journal of Mechanical Sciences, Vol. 48, pp. 763–779, 2006.
[66] Esmaeilli M., Durali M., Jalili N., “Modeling and vibration analysis of
vibrating beam microgyroscopes under longitudinal rotation of the support,”
Proceedings of 2005 ASME International Mechanical Engineering
Conference & Exposition, Orlando, Fl, 2005.
[67] Esmaeilli M., Durali M., Jalili N., “Closed form solutions of frequency
equations for a cantilever flexible beam under general base excitation”,
Proceedings of ASME International Design Engineering Technical
121
Conferences and Information in Engineering Conferences-DETC2005: 20th
Biennial Conference on Mechanical Vibration and Noise, Long beach, CA,
pp. 145-151, 2005.
[68] Esmaeilli M., Durali M., Jalili N., “Ring microgyroscope modeling and
performance evaluation,” Journal of Vibration and Control, 12(5), pp. 537-
553, 2006.
[69] Malatkar P., “Nonlinear vibrations of cantilever beams and plates”, Ph.D.
Dissertation, Virginia Polytechnic Institute and State University, 2003.
[70] Rüdiger G. Ballas, “Piezoelectric Multilayer Beam Bending Actuators:
Static and Dynamic Behavior and Aspects of Sensor Integration”, Springer,
2007.
[71] Ikeda T., “Fundamental of Piezoelectricity,” Oxford Scientific, Oxford,
New York, 1990.
[72] Dadfarnia M., “Lyapunov-based piezoelectric control of hybrid flexible
structures”, M.S. Thesis, Clemson University, 2003.
[73] Andre′ Preumont, “Vibration Control of Active Structures: An
Introduction,” Kluwer Academic Publishers, 1997.
122
[74] Dadfarnia M., Jalili N., Xian B., Dawson D. M., “Lyapunov-based
piezoelectric control of flexible cartesian robot manipulators,” Proceedings of
the American Control Conference, 6, 2003, pp. 5227-5232.
[75] Dadfarnia M., Jalili N., Xian B., Dawson D. M., “A Lyapunov-based
piezoelectric controller for flexible cartesian robot manipulators”, Journal of
Dynamic Systems, Measurement and Control, 126(2), pp. 347-358, 2004.
[76] Dadfarnia M., Jalili N., Liu Z., Dawson D. M., “An observer-based
piezoelectric control of flexible Cartesian robot arms: Theory and
experiment”, Control Engineering Practice, 12(8), pp. 1041-1053, 2004.
[77] Meirovitch L, “Principles and techniques of vibration”, Prentice-Hall,
1997.
[78] Tang T. K., Gutierrez R.C., “A packaged silicon MEMS vibratory
gyroscope for microspacecraft,” IEEE Microelectromechanical Systems
Conference, Nagoya, 1997.
[79] Royle C. M., Fox C. H. J., “The mechanics of an oscillatory rate
gyroscope with piezoelectric film actuation and sensing,” Proceedings of the
Institution of Mechanical Engineers, part C: Journal of Mechanical
Engineering Science, 215 (10), pp. 1211-1221, 2001.
123
[80] Meirovitch, M., Baruh H., “Optimal Control of Damped Flexible
Gyroscopic Systems”, Journal of Guidance and Control, Vol 2, No. 2, pp.
157-163, pp. 88-90, 1981.
[81] Canavin, J. R., and Meirovitch, L, “Eigensolution for Large Flexible
Spacecraft with Singular Gyroscopic Matrices”, Journal of Guidance and
Control, Vol 2, No. 2, 1981.
[82] Meirovitch L., and Ryland G., “A Perturbation Technique for Gyroscopic
Systems with Small internal and External Damping”, Journal of Sound and
Vibration, 100 (3), pp. 393-408, 1985.
[83] Meirovitch L., “A stationarity Principle for the Eigenvalue Problem for
Rotating Structures”, AIAA Journal, Vol 14, No. 10, pp. 1387-1394, 1976.
[84] Meirovitch L., “A Modal Analysis for the Response of Linear
Gyroscopic Systems”, Transaction of ASME Journal of Applied Mechanics,
Vol. 42, No. 2, pp. 446-450, 1975.
[85] Oz, H., Meirovitch, L., “Optimal Modal-Space Control of Flexible
Gyroscope Systems, Journal of Guidance and Control, Vol 3, pp. 218-226 No.
3, 1980.
[86] Meirovitch, L., Oz, H., “ Modal-Space Control of Distributed Gyroscopic
Systems”, Journal of Guidance and Control, Vol 3, No. 3, pp. 140-150,1980.
124
[87] Meirovitch, L., Ryland, G., “Response of Slightly Damped Systems”,
Journal of Sound and Vibration, Vol. 67 (1), pp. 1-19, 1979.
[88] Meirovitch, L., “On the Reduction of the Eigenvalue Problem for
Spinning Axisymmetric Structures”, AIAA Journal, Vol. 14, No. 2, pp. 150-
157, 1976.
[89] Meirovitch, L., “A New Method of Solution of the Eigenvalue Problem
for Gyroscopic Systems”, AIAA Journal, Vol. 12, No. 10, pp. 1337-1342,
1974.
[90] Meirovitch, L., “A Separation Principle for Gyroscopic Conservative
Systems”, Transaction of ASME Journal of Vibration and Acoustics, Vol. 119
(1), pp. 110-119, 1997.
[91] Meirovitch, L., Kwak, M. K., “Convergence of the Classical Rayleigh-
Ritz Method and the Finite Element Method”, AIAA Journal, Vol. 28, No. 8,
pp. 1509-1516, 1990.
[92] Meirovitch, L., “Fundamentals of Vibrations”, McGraw-Hill, 2002.
[93] S. Graham Kelly, “Advanced vibration analysis”, CRC/Taylor & Francis,
2007.
[94] Rao, S. S., “Mechanical Vibrations”, Pearson Prentice Hall, 2004.