CHAPTER - 1

INTRODUCTION

1.1 General

Earthquakes by their furious nature, volume and vastness of devastation are by far
amongst the most dangerous and destructive of natures affects. Phenomenal damage in terms
of life and property occurs every time an earthquake activity is reported. The affects of
earthquakes are magnified in context of today’s modernisation with our cities turning into
concrete jungles. In this scenario, it is pertinent for design engineers to come up with more
scientific and innovative ways to impart more effective seismic resistance to structures.
The seismic threat is, effect wise, more severe today, since structures are reaching
farther into the sky than yester years. The threat assumes catastrophic dimension when
coupled with the fact that in today’s world, numerous such tall structures are closely co-
located and all most of them have a capacity to house a few thousand human beings, if not
more. Hence, the necessity to equip these structures in such a way that they can bear the
seismic effects with minimal damage has always been a topic of further research.
The design philosophy is essentially the same in all countries and differs only
superficially. In India, the basic design philosophy is that, structure damage must be minimal
during a design basis earthquake (DBE) and structural safety must be ensured for the
maximum considered earthquake (MCE). The basic parameters governing the reaction of a
structure to any external excitation are strength, ductility, damping and stiffness. These
parameters can be altered to check the effect on the response of the structure. Various
techniques developed over the years are based on these parameters.

1.2 Techniques Used in Aseismic Design

Earthquake resistant design has been the quest for engineers since a long time. The
desire to achieve a completely resistant design is still far flung but in recent years
considerable advancements have been made to ensure a fairly safe design against DBE. Many
parameters have been considered in various techniques to achieve an acceptable reduced
response of a structure. Research to achieve a seismic resistant design dates back to the early
20th century, which makes base isolation technique more than a century old, the best

1
understood as well as and the most widely accepted of all techniques. Other techniques, like
the dampers date back to the mid 20th century and later still the tuned mass dampers are still
newer than passive dampers.

1.2.1 Base Isolation

Firstly, the base isolation technique which is now considered as a mature and efficient
technology has its earliest reference as far back as 1909. The basic principle of functioning
involves decoupling of the superstructure from the base of the structure. This shifts the
fundamental time period of the structure to a larger value such that the base-isolated time
period is greater than the fixed base time period. Base-isolated fundamental frequency is
much lower than that of the fixed base structure. The isolator provides a means to deflect
energy through the system and reduces the demand rather than increasing the capacity.
The first dynamic mode of the base-isolated system involves deformation only in the
isolation system. The system further dissipates energy due to the damping inherent to it,
thereby limiting the energy transfer to the superstructure. As a result, the floor accelerations
and inter-storey drifts are controlled drastically. Broadly, the two main isolation systems are
the elastomeric bearings and the friction systems. There are various base isolation systems
that can be categorised under these headings, namely:
1. Laminated Rubber Bearing (LRB)
2. Lead Rubber Bearing, called the NZ system
3. Friction Pendulum System (FPS)
4. Pure Friction system (PF)
5. Resilient - Friction Base Isolator (R-FBI)
6. Electric de France system (EDF)

1.2.1.1 Laminated Rubber Bearing

The Laminated rubber bearings system comprises steel and rubber plates
consecutively placed one after another. This system is characterised by low horizontal
stiffness and high vertical stiffness and damping. The schematic diagram of the LRB system
is given in Fig. 1. The system is analysed as a linear system but it may possibly show
stiffening behaviour and a hysteresis loop at excessive deformations. For such a system, the

2
stiffness and damping parameters are selected ab-initio to achieve a desired value of isolation
time period (Tb) and damping ratio (ξb). The underlying relations are defined as:

M
T b=2 π
√ Kb (1.1)
cb
ξ b=
2 Mωb (1.2)

where M is the total mass of the building and

ωb is the isolation frequency and is equal to

2π
Tb .

The LRB system significantly reduces the acceleration response of the building but
then it has an inherent disadvantageous effect of enhancing the lateral displacements which
has other sequential requirements life provision of moat, which again comes with its
disadvantages in the modern day context of building design. This disadvantage of high
displacement which stems from low stiffness of the LRB can be reduced by addition of
energy dissipation mechanisms like Lead Rubber Bearings, also called the N-Z system.

1.2.1.2 Lead Rubber Bearing

The Lead rubber bearing is commonly referred to as the N-Z system owing to its
origin, development and wide use in New Zealand. It overcomes the disadvantage of the LRB
system. The N-Z system is similar to the LRB system but with an additional central lead core.
The lead core increases the initial stiffness of the system, thereby reducing its response due to
minor vibrations and further reducing its bearing displacement. Also, the yielding of the lead
core provides additional hysteretic damping. The N-Z system can be modeled in SAP 2000
using wen parameters. The N-Z system is shown schematically in Fig. 1. The N-Z system is

characterised by its isolation time period (Tb), its damping ratio (

ξ b ) and normalised yield
strength (Fy/W), where W is the total weight of the building and is equal to Mg.

1.2.1.3 Friction Pendulum System

The friction pendulum system is simply based on the mechanism of friction and the
energy dissipation caused due to the resistance to free motion of the bearings. It refers to

3
Coulomb’s friction. This resistance is due to the horizontal frictional force developed,
depending on the coefficient of friction ( μ ). Further the bearing is made to slide on an
articulated slider having a concave chrome surface. A typical friction system has two distinct
phases i.e. the stik phase and the slip phase. The system remains in stik phase till the external
energy does not overcome the frictional force, and then only the system gets activated.
During slip phase, the lateral resisting force developed in a FPS is due to two factors i.e. the
coefficient of friction of the sliding surface and the restoring force developed due to the rising
of the articulated sliding bearing on the spherical concave chrome surface. FPS is
schematically represented in Fig. 1.

1.2.1.4 Pure Friction System

A system based on just the mechanism of sliding on a surface with a specified value
of coefficient of friction ( μ ), wherein the horizontal frictional force at the interface of the
bearing and the surface offers resistance to free motion as well as dissipates energy, the
system is said to be a Pure friction system.

1.2.1.5 Resilient-Friction Base Isolator (R-FBI)

The R-FBI system consists of teflon plates concentrically placed on top of each other.
These plates are placed into a central rubber core. The system develops friction between the
plates to act as an isolation system.

1.2.1.6 Electric de France

The system gets its name from its source of development, “Electric de France” (EDF)
standardised for nuclear power plants in region of high seismicity. It is also basically a
friction system and has distinct stik-slip phases to develop its isolator characteristics.

1.2.2 Tuned Mass Damper

The Tuned Mass Damper is probably amongst the oldest recorded structural vibration
control device that has been technically accepted and widely used today. The origin of the

4
modern day TMD can be linked to the vibration controlling device called the dynamic
vibration absorber developed by Frahm in 1909.
The modern day civil engineering structures are slender, need to be lighter and have a
low damping capacity. This structural form makes the structure vulnerable to nature’s fury in
terms of earthquake and wind effects. Hence, the importance of a matching modern day
damping device to keep pace with technology cannot be over emphasized. TMD has gained
importance for its use to limit excessive oscillation in structures in general and tall structures
specifically.
A TMD is a device consisting of a mass attached to a structure via a spring-dashpot
system, such that it oscillates at the same frequency as the main structure (resonance) but

with a 180° phase shift. The system dissipates energy as relative motion develops between
the mass and the structure. Passive TMD’s have been applied onto the modern day super tall
structures with desirable success. However, the world opinion is divided on the use of TMD’s
for use to control response of structures subjected to earthquake vibrations. These differing
schools of thought are based on results from different experiments which lead to
contradictions. Firstly, a TMD as a passive device requires the motion of the structure for it to
get activated. This is a good possibility for wind excitation but finds reserved thoughts on its
use for earthquake excitation, as these are usually be of shorter duration and much more
random. Hence, the TMD may not get activated at all. Secondly, assuming that the TMD
does get activated due to the earthquake, the duration of the earthquake may not be long
enough for the TMD to attain its fully functional form. It is an outflow of this understanding
that we know the TMD may actually enhance the response of the structure in the initial stage.
Thirdly, a single TMD turned to the fundamental mode of a building may not be effective for
earthquake induced response reduction, primarily due to the fact that the earthquake motion
possibly induces multi-mode response in tall structures. Fourthly, for a single TMD, the
damper parameters may not be optimum for the earthquake ground motion that the building is
subjected to.
Nevertheless, TMDs have been successfully installed in most of the super tall
buildings of today including the Petronas Towers, Al-Burj and Taipai-101, to name a few.

1.3 Need for the Present Study

5
 Both the techniques of base isolation and TMD have been independently used with
success in most modern day structures. Both these systems have their own inherent
limitations and disadvantages. The base isolation technique of achieving aseismic design can
be used for smaller structures say maximum upto a height of 20 m, thereafter its affect
diminishes. The base isolation system increases the displacement response of the structure
excessively, bordering at unacceptable limits. The TMD on the other hand, has its
effectiveness for total structures and is less effective if not adverse for low rise buildings.
TMD has its disadvantages as discussed in the preceding paragraphs.
Hence, it is aimed to achieve a system which has aspects of both the base-isolated
system as well as the TMD and also overcomes the short comings of both these systems.

1.3.1 Base Isolation with TMD

The aim of such a passive hybrid system which would result from the combination of
a tuned mass damper applied on a base-isolated structure would be to limit the excessive
displacement brought about by the isolation system without losing out on the reduction in the
acceleration response of the base-isolated structure. The hybrid passive system must possess
some basic dynamic properties. Firstly, the tuning of the damper of the hybrid passive
system, which is mandatory to achieve resonance response between the TMD and the base-
isolated structure. Secondly, the sensitivity and the interaction between the base-isolated
structure and the TMD, which defines the boundaries for activation of the TMD. Thirdly, the
resulting non-classical damping factor resulting from the three sub-systems i.e. the structure,
the TMD and the isolators. In the hybrid passive system, the TMD parameters like the mass
ratio and tuning ratio are adjusted specifically for the base-isolated structure to ensure
optimal performance. The location of the TMD is worth a discussion in itself. For this study
the TMD is assumed to be attached at the top of the structure.

1.4 Objectives of the Study

Keeping in view the background given above, it is justified to pursue the present
study. The primary objective of the study is to control excessively large displacements in the
base-isolated structure using a TMD. The other objectives of this study include:
1 To develop formulation for base-isolated structure equipped with tuned mass damper.

6
2 To numerically solve problem of the base-isolated structure equipped with tuned mass
damper when excited by earthquake.
3 To assess performance of low initial stiffness (flexible) base-isolated structure equipped
with tuned mass damper.
4 To assess the performance of the hybrid passive structure when the tuned mass damper is
placed on the base mass of the isolator.
5 To further carry out parametric study for
5.1 Variation in time period of isolation
5.2 Variation in the tuning frequency of the tuned mass damper

1.5 Organization of the Thesis (till Nov 2010)

The thesis has been divided into 4 chapters including the introduction chapter. Tables
and figures have been presented in a sequence at the end of the thesis;
Chapter 1 gives an overall idea of the aseismic design concepts. It also justifies the present
study and further emphasizes the objectives of work.
Chapter 2 provides the detailed literature review of the aseismic design concepts which form
part of this study.
Chapter 3 presents the numerical problem considered in the study and the results obtained
therein.
Chapter 4 summarises the study (till Nov 2010) and furnishes important conclusions that
have been drawn based on the study (till Nov 2010). Also, recommendations are made and
further scope is defined.
Finally, references have been provided at the end.

7
 CHAPTER - 2
LITERATURE REVIEW

2.1 General

Structures made in the early part of 19th century used primitive means and ideas and
also relied much on their size and shape to provide inherent stability. The intrinsic properties
of mass, damping and stiffness were present in these structures even though less known of, at
that time. The urge to go higher, lighter and slender keeping in view the requirement of the
modern world, have played negatively with mass, stiffness and damping parameters of the
structure but advanced technological awareness of material, construction techniques, software
etc have assured safe construction nevertheless. It is essential for all in the structural design
industry to keep abreast with these latest techniques for better solutions to tomorrow’s
problem.

2.2 Base Isolation

The basic idea of base isolation relates to the work of J.A. Calantarients, a medical
doctor from Scarborough, England, from start of 20 th century up to 1906. The concept in the
new design was to uncouple a building from the damping effects of the earthquake ground
motion. “This invention relates to the building construction and has for its object to prevent
as much as possible the communication of shocks or vibrations of earthquakes to buildings
and so render them refer.” (U.S. Patent, August 31, 1909). The newly developed system
showed very promising future prospects which led to major research being conducted to get
practical designs of the base-isolated structure. Most of these systems were unacceptably
complicated but some had the correct concept to be tried and very few out of them have been
implemented and are called the base-isolated systems today. The purpose of these systems is
to reduce the accelerations in buildings to below the ground acceleration for which the
building must be flexible.
In a low to mid rise building, the necessary flexibility can only be achieved at the
foundation level by use of base isolation (Kelly, 1986). In the second and third decade of the
20th century, the concept of flexible first storey was proposed. This involved giving less
lateral stiffness to the columns of the first storey as compared to the columns above the first

8
storey. Hence, under earthquake loading, the deformations would be concentrated in these
first storey columns. But this system, to be effective in reducing accelerations at upper levels
had to have large deflections in the first storey and this led to guaranteed collapse (Kelly,
1986).
A point of contradiction for the patent of base isolation practically applied to bridges
and buildings is seen as part of history because a similar patent exists in name of Jules
Touaillon of San Francisco filed in US patent office in February 1870. The modern era of the
base isolation technique is said to have begun in New Zealand in mid seventies of the 20 th
century. Passive energy dissipaters were used in conjunction with elastomeric and sliding
isolators. Till early eighties of the 20 th century, it had been learnt that base isolation reduced
damage to the frame (ductile yielding) more than any other system developed till them
(Buckle and Mayes, 1990). In the mid of 20th century, a modified approach called the ‘soft
first storey’ approach was tried. It was proposed that the first storey columns yield during an
earthquake, producing an energy absorbing action and controlling the displacements.
However, to produce enough damping the displacement would still have to be several inches
and a yielded column had a greatly reduced buckling load such that column instability and
collapse would be inevitable (Kelly, 1986). The first possible use of rubber for earthquake
protection was in an elementary school in Skopje, Yugoslavia. This was a completely
unreinforced rubber block supporting the superstructure. This in a way is the first rubber
bearing.
Intensive study on influence of isolator characteristics on seismic response of
multistorey base-isolated structures has been investigated as a part of further development of
base-isolated systems. While comparing the response of the isolated structure for equivalent
linear and bi-linear force-deformation behaviour of the isolator is made. Further, effects of
the shape of isolator loop and superstructure flexibility on the seismic response of the base
isolated structure are also investigated. It has been clearly observed that equivalent linear
elastic-viscous damping model for a bi-linear hysteretic model of the isolator under predicts
the superstructure acceleration and over predicts the bearing displacement. Noteworthy is the
fact, that response of base-isolated structures is significantly influenced by the shape and
parameters of the bi-linear hysteresis loop of the isolator. If the base-isolated system has low
yield displacement, then more earthquake accelerations associated with high frequencies
would be transmitted to the superstructure. Flexibility of the superstructure does not influence
the bearing displacements much but certainly increases the superstructure acceleration
(Matsagar and Jangid, 2004). Extensive study on effects of eccentricity of superstructure and

9
the isolator revealed that effectiveness of base isolation reduces as eccentricity of
superstructure increases, though not much effect is seen on base displacements. The isolator
eccentricity increases the base displacement and decreases the superstructure displacement
perpendicular to the direction of eccentricity. The overall isolator effectiveness is also
reduced (Jangid and Datta, 1995).
The damping provided in various types of base isolators and its effect on the system
revealed that damping does reduce displacement but at the expense of increased floor
accelerations as well as interstorey drifts. This change is affected by increasing the response
in higher modes though it is known that higher modes in a base-isolated structure are
orthogonal to the base shear, hence reduction in base displacement and base shear does not
necessarily mean reduced floor accelerations. If the isolator damping and stiffness is
provided, so to say, for the MCE level, it means that for a more probable lower level
earthquake, the isolation will be too stiff and heavily damped such that the isolator may not
participate at all.
The dilemma of achieving lower displacements for a high input level earthquake as
well as good performance for low level earthquakes needs to be addressed. The need is for a
system which is very stiff initially, then becomes flexible as input level reaches DBE and
then again stiffens upto the MCE level. This can be done with relative ease with elastomeric
isolators and the FPS, as compared to the other systems (Kelly, 1990). A multistorey
structure base isolated with systems like the LRB, N-Z and FPS, and subjected to sinusoidal
ground acceleration has been considered. For the same input level of earthquake, the LRB
leads to lower peak accelerations transmitted to the structure and the N-Z system causes least
relative base displacements. The friction system generates high frequency content in the
acceleration transmitted to the structure (Tadjbakhsh, 1990).
The base isolation technique is very well understood and is being extended for
retrofitting of existing structures such as historical buildings, monuments, bridges, liquid
storage tanks etc. Some of the numerous advantages of such an option include considerable
reduction in structural response, reduction of the seismic forces by a factor of 0.3 to 0.8 in the
superstructure, thereby improving the overall economy and effectiveness. The original
aesthetic value of the structure remains unaltered, as also the retrofitting work does not
interfere with building usage in any way (Matsagar and Jangid, 2008).
The base isolation technique has been modified marginally to cater to the impact
scenario of buildings which leads to a pounding effect. The R-FBI system performs best of all
the base isolation systems i.e. it has the least loss of performance after impact, and the FPS

10
system has the worst performance out of all the isolation systems (Matsagar and Jangid,
2003). The overall advancement of the base-isolated system is still an ongoing process and
has its varied applications, which are still being researched.

2.3 Tuned Mass Damper

TMD’s are used principally for controlling wind vibrations in elastic structures. These
vibration absorbers dampen the motion by transferring the kinetic energy between various
modes of vibration (Buckle and Mayes, 1990). The concept of a tuned mass damper has its
origin from the attempt made by Frahm in 1911 to control rolling motion in ships. The
undamped mass-spring absorber was able to set the vibration amplitude of the main system to
zero for a single frequency. The performance of this system deteriorated suddenly for other
frequencies. This was the most primitive recorded use of a TMD. Ormondroyd and Den
Hartog then improved on Frahm’s design by their damped vibration absorber designed for
broad band attenuation. They also introduced the system of invariant points which has
evolved as the path for analytical optimal solution. Optimal damper parameters are those
which would control the response of the main system as well as its own motion.
Over the years, since Ormondroyd’s and Den Hartog’s work, different approaches
have been taken to find the optimal absorber parameters and their locations. Genetic
algorithms have been used by Hadi and Singh to achieve this. The importance of tuning of a
TMD and the effects of detuning were brought out by Rana and Soong. Eigen vector
normalisation was also used by them to find optimal absorber parameters. Lin et al. identified
the modal properties of a building through a random decrement method and Ibrahim
identified the modal properties through time domain method (Spencer and Nagarajaiah,
2003).
Since, the TMD can be optimally tuned to a single fundamental frequency of the
structure, the system can be most effective for a frequency ratio of one, further the mass ratio
will also be a major factor and the damping plays a significant role too. Ideally, if the
frequencies and amplitudes of the TMD and the structure match such that the reaction of the
TMD is equal and opposite to the action of the strucute subjected to the random vibration.
The efficiency in this regard was improved by using two TMD’s instead of one (Iwanami and
Seto, 1984). The disadvantage of a single TMD is its sensitivity of effectiveness to the error
in the natural frequency of the structure and the damping ratio of the TMD. Since, the
effectiveness of a TMD is dependent in a major way on the tuning and optimum damping; the

11
next step was to increase the number of dampers. Hence, the concept of multiple tuned mass
dampers (MTMD’s) with varying dynamic characteristics has been evaluated and desirable
results achieved. The optimum damping ratio of MTMD system decreases with increase in
number of MTMD’s and increases with increase of mass ratio. As for the optimum tuning
frequency, it increases with the increase in number of MTMD’s and decreases with the
increase of the mass ratio (Jangid, 1999).
MTMD’s are being extensively used in all super tall constructions today with Taipei-
101 being the latest amongst such constructions. An added element of active control
synergised with the TMD is also incorporated thereby making the complete system to be such
that it is not just reacting to the changing structural situation but predicting and forecasting
the structural response and thereby pre-empting the TMD action.

2.4 Hybrid Passive System

The LRB based base isolation system has low initial stiffness and as a result such a
system has large lateral displacements. Such a system is modified with a central rubber or
lead core to provide some level of initial stiffness to the system. It is this parameter of initial
stiffness which requires to be addressed in the passive hybrid system such that the
displacements are not too much and yet the system gives good structural response reduction.
The affect of applying a TMD to such a system with the aim of controlling the displacements
or the lateral deformation have brought out desirable initial indications. It is also a fact that
since a TMD requires a few seconds to get fully operational, the TMD actually has little
effect on the structural response during this period. In fact, the response during this phase can
get amplified marginally. However, subsequent response would tend to be lower. The overall
response of the hybrid passive structure can also be reduced with increase in damping of the
first mode (base isolator mode). However, damping of the isolator cannot be increased
beyond a certain limit to reduce the amplitude of vibration. The structural response control
increases for higher damping upto a certain limit and thereafter the effect gets reversed. It can
thus be stated that for higher damping of structure, it is possible to get additional damping
effect from TMD, for a structure with optimum damping value of base isolation.
A major point of contention is the adequacy of the TMD in limiting the structural
response caused due to earthquake vibrations. There are a few reasons for contradictory
results from studies on TMD’s. Firstly, a single TMD can be tuned to a single frequency
which is usually the fundamental frequency of the structure and so the TMD may not be

12
significantly effective in reducing the earthquake response which may be a multiple mode
response. Secondly, the inaccuracy in tuning and providing the incorrect mass ratio and/or
damping ratio can lead to incorrect results. Thirdly, a TMD reacts to the response of the
structure, and so for a short duration and/or short intensity earthquakes, the TMD may not
activate at all. There exists a critical value of damping for the damper which, if not provided,
will enhance the structural response rather than controlling it. Another important aspect is
that of the input frequency, if the natural frequency of the structure is higher than the
earthquake frequency, the structural response would be enhanced rather than being curbed
(Tsai, 1995).
It can be said with impunity that the displacement demand of a base-isolated structure
can be reduced by adding a TMD to it. The amount of reduction in displacement will depend
on the mass, frequency and damping characteristics (Taniguchi, 2008). TMD has to be
applied to a base isolation system with low initial stiffness to do away with the requirement
of providing inherent stiffness to the isolator in terms of a lead core or likewise. A TMD
designed integrally with the building proves to be more beneficial in terms of economy. Even
though the TMD may experience large displacements, these motions will be within
acceptable tolerance limits due to their low frequency (Taniguchi, 2008).

13
 CHAPTER - 3
MATHEMATICAL MODELLING AND ANALYSIS

3.1 General

The feasibility of using the TMD on the top floor level of a base-isolated structure to
reduce the structural response in terms of top floor accelerations, top floor displacements and
storey shears was carried out in a phased manner. Firstly, a five storey structure was
modelled as a fixed base and then the same structure was base-isolated with laminated rubber
bearings (“Dynamics of Structures”, Chopra, 2008). The modelling was carried out in SAP
2000 software. Also a twelve storey fixed base structure was then modelled with a TMD on
its top floor (Chung and Loh, 2005). This validation verified the correctness of the modelling.
Secondly, a hybrid structure was modelled and an initial numerical study was carried out to
find the range of height of structure for which the hybrid passive structure was the most
efficient aseismic design method by modelling all four types of structures ranging from five
storey height to thirty five stories. Thirdly, some key parameters were identified for further
investigation on the four types of structures, like the time period of isolation, damping of base
isolator and damping of TMD. As a part of the study, five earthquake time histories have
been considered to get the structural response and further arrive at the conclusions. The
earthquakes considered for the study are given in succeeding paragraphs, the details of which
are given in Table 1. The mathematical model of the four types of structures mentioned above
is shown in Fig. 2.

3.2 Governing Equations of Motion

The governing equation of motion for the fixed base structure, the base-isolated
structure, the structure with TMD and the hybrid passive structure have been described. The
modelling involved, the solution procedure and matrices for mass, stiffness and damping
have been detailed below. Lumped mass modelling is done for the superstructure and the
isolator. The effect of rotation in the structure, isolator and TMD is not taken into
consideration. Some assumptions have been made for the structural systems under
consideration:

14
 1. The superstructure is assumed to remain within the elastic limit during the
earthquake excitation.
2. The floors are assumed to be rigid in their own plane and the mass is lumped at
the two ends of each floor level.
3. The columns and beams are inextensible and weightless providing lateral
stiffness.
4. The system is subjected to a horizontal component of the earthquake ground
motion in one direction.
5. The effect of soil structure interaction is neglected.
All mathematical modelling and numerical formulation has been shown on a five
storey structure but the same can be extended for a N storey structure.

3.2.1 Fixed Base Structure

The general equation of motion for a fixed base structure subjected to a horizontal
ground motion is described. The displacement of the ground is denoted by ug , the total (or

absolute) displacement of the mass mj by utj , and the relative displacement between this
mass and the ground by uj . Hence, the acceleration, velocity and displacements are actually
due to the ground motion because of the earthquake as well as the superstructure. At each
instant of time these displacements are related by

t
u j (t )=u j (t )+u g (t ) (3.1)

The resultant governing equation for a fixed base structure is given by

.. . ..
[ M s ]{u (t )}+[C s ]{u ( t )}+[ K s ]{u( t )}=−[ M ]{u g ( t )} (3.2)

Here
M s is the mass of the superstructure, C s is the damping of the

superstructure and
K s is its stiffness. ü , u̇ and u are the acceleration, velocity and
displacement, respectively. The mass matrix can be given as

15
 m5 0 0 0 0

M=
The stiffness matrix can be given as
[ 0
0
0
0
m4
0
0
0
0

0
0
m3 0
0
0
0 m2 0
0 m1
] (3.3)

k5 −k 5 0 0 0

K=
[ −k 5
0
0
0
k 5+ k 4
−k 4
0
0
−k 4
k 4+ k 3
−k 3
0
0
−k 3
k 3+ k 2
−k 2
0
0
−k 2
k 2 +k 1
] (3.4)

The damping matrix of the super structure, C is not known explicitly. It is constructed
by assuming the modal damping ratio for superstructure, which is kept constant.

3.2.2 Base Isolated Structure

The general equation of motion for the superstructure-isolator model can be expressed
as

[ M ]{üt (t )}+[ C ]{u̇ (t )}+[ K ]{u(t )}=0 (3.5)

t
Here u={u j } is the column vector of relative structural displacements with respect
t t t
u
to the isolator and u ={u j } is the column vector of the total structural displacements. j
is the relative floor displacement with respect to the isolator at the jth floor, mj is the floor
mass at the jth floor, kj is the stiffness of the jth floor, ub is the displacement of the isolator and
mb is the mass of the isolator

utj=u j +u b +u g (3.6)

16
 Where, ug is the displacement of the ground due to the earthquake. ub is the
displacement of the isolator. Hence, it can be said that the acceleration, velocity and
displacements are due to the ground motion because of the earthquake, the isolator motion
and the superstructure interaction. Now for the five storey base isolated building the
governing equations of motion are given by

[ M ]{üt (t )}+[ C ]{u̇ t (t )}+[ K ]{ut (t )}=−[ M̄ ]{ü g } (3.7)

Where,

m5 0 0 0 0 m5

[ ]
0 m4 0 0 0 m4
0 0 m3 0 0 m3
M=
0 0 0 m 2 0 m2
0 0 0 0 m1 m1
0 0 0 0 0 mb
(3.8)

k5 −k 5 0 0 0 0

[ ]
−k 5 k5+ k4 −k 4 0 0 0
0 −k 4 k 4+ k 3 −k 3 0 0
K=
0 0 −k 3 k3+ k2 −k 2 0
0 0 0 −k 2 k 2 +k 1 −k 1
0 0 0 0 −k 1 k1+ kb
(3.9)

t
2 m5

{}
2m 4
2 m3
M̄ =
2 m2
2 m1
mb
(3.10)

17
 t
2 u5

{}
2u 4
2 u3
ut =
2 u2
2 u1
ub
(3.11)
The damping matrix of the superstructure, C is not known explicitly. It is constructed by
assuming the modal damping ratio for superstructure, which is kept constant.

3.2.3 Structure with Tuned Mass Damper

Consider a structure fixed at its base and having a TMD of mass md, damping cd and
stiffness kd attached at the top floor. The mass, damping and stiffness of the superstructure are
same as described in section 3.2.1. The equation of motion for the model thus depicted can be
given as

[ M ]{ü(t )}+[ C ]{u̇(t )}+[ K ]{u (t )}=−[ M̄ ]{üg } (3.12)

Where, ut is a vector of horizontal displacements relative to the ground, with

components ud and us for the TMD and the superstructure, respectively. The mass matrix for
the system can thus be written as

md 0 0 0 0 0

[ ]
0 m5 0 0 0 0
0 0 m4 0 0 0
M=
0 0 0 m3 0 0
0 0 0 0 m2 0
0 0 0 0 0 m1
(3.13)

The stiffness matrix for the structure with TMD at its top can be written as

18
 kd −k d 0 0 0 0

[ ]
−k d kt + k5 −k 5 0 0 0
0 −k 5 k 5+ k 4 −k 4 0 0
K=
0 0 −k 4 k 4+ k3 −k 3 0
0 0 0 −k 3 k 3+ k 2 −k 2
0 0 0 0 −k 2 k 2 +k 1
(3.13)

The damping matrix can be given as

cd −c d 0 0 0 0

[ ]
−c d c t +c 5 −c 5 0 0 0
0 −c 5 c 5 +c 4 −c 4 0 0
C=
0 0 −c 4 c 4 +c 3 −c 3 0
0 0 0 −c 3 c 3 +c 2 −c 2
0 0 0 0 −c 2 c 2 +c 1
(3.14)

The damping constants in equation 3.14 are arbitrary; hence the damping matrix is
not, in general, proportion to the mass and stiffness matrix. Some other essential parameters
are

kd
ω2d =
md (3.15)

ks
ω2s =
ms (3.16)

ωd
Ω=
ωs (3.17)

md
μ=
ms (3.18)

19
 Equation 3.15 provides the circular frequency of the upper part of the structure, when
considered independently. Equation 3.16 gives the circular frequency of the lower part of the
structure, when considered independently. Equation 3.17 represents the frequency ratio, also
called the tuning ratio, between the independently considered upper and lower portions of the
structure. Equation 3.18 gives the mass ratio of the TMD.

3.2.4 Hybrid Passive Structure

The hybrid passive structure has a TMD attached to the roof level of a base-isolated
structure. Assume all notations same as mentioned in the preceding paragraphs. The
governing equation of motion for such a structure can be written as

[ M h ]{üt (t )}+[C h ]{u̇ t (t )}+[ K h ]{ut (t )}=−[ M̄ h ]{ü g } (3.19)

Where,

md 0 0 0 0 0 md

[ ]
0 m5 0 0 0 0 m5
0 0 m4 0 0 0 m4
M h= 0 0 0 m3 0 0 m3
0 0 0 0 m2 0 m2
0 0 0 0 0 m1 m1
0 0 0 0 0 0 mb
(3.20)

kd −k d 0 0 0 0 0

[ ]
−k d kd + k5 −k 5 0 0 0 0
0 −k 5 k 5 +k 4 −k 4 0 0 0
K h= 0 0 −k 4 k 4 +k 3 −k 3 0 0
0 0 0 −k 3 k 3 +k 2 −k 2 0
0 0 0 0 −k 2 k 2 +k 1 −k 1
0 0 0 0 0 −k 1 k1+ kb
(3.21)

20
 2 md

{}
2m5
2 m4
M̄ h = 2m3
2m2
2m1
mb
(3.22)

t
2 ud

{}
2u 5
2 u4
t
x = 2u 3
2u 2
2u 1
ub
(3.23)

The damping matrix of the superstructure, C is not known explicitly. It is constructed

by assuming the modal damping ratio for superstructure, which is kept constant.

3.3 Modelling and Validation

A sample problem has been validated each for the fixed base structure, base-isolated
structure and a similar structure with a TMD at its top floor. The structure was a five storey
structure. For the fixed base, the superstructure was defined by its mass, damping and
stiffness as M, C and K, respectively. Mass of the structure was idealised as lumped at the

floor levels. M is a diagonal matrix with diagonal element

m jj =m j , the mass lumped at the

jth floor. Total mass of the building is

M=∑ m j . The natural periods and modes of

vibration of the fixed base system are denoted by

T nf and φnf , respectively, where

n=1,2,3...... N . For the next part of the validation the five storey building is provided with

21
base isolation, the base mass of
mb is supported on a LRB isolation system. Two essential
parameters that characterise the isolation system are its time period and damping ratio and are
given in equation 1.1 and 1.2. For both, the fixed base and the base-isolated structures, the
mass and stiffness properties were assumed uniform over the height of the structure. The
results for the modal analysis of the fixed base structure (“Dynamics of Structures, Chopra,
2008) and the time histories and response of the base-isolated structure (Matsagar and Jangid,
2003) have been achieved. The comparison with published results is as shown in Table 2 and
Fig. 3.
The twelve storey structure with the TMD was validated (Chung and Loh, 2003). It
was assumed that the structure has one degree of freedom at each floor level and constant
damping of 5% in all modes. The mass ratio was taken as 1.98%, with the total mass of the
structure as 3024.6 tons. The results are as shown in Fig. 4.

3.4 Numerical Study

The basic structure for the study was taken as a five storey structure to start with. The
results of the study on the five storey structure in terms of the time history plot are as shown
in Fig. 5. However, the height of the structure being an essential parameter, further study
involved structures upto 35 storey height. Some other important parameters that were
investigated were the stiffness of the base isolation, damping of the base isolation and the
TMD.

3.4.1 Effect of Number of Storeys

The aim of this parametric study was to arrive at the most suitable height of structure
for which the hybrid passive structure would prove more beneficial than any other means of
aseismic design. The structures that were modelled were 8 storey, 12 storey, 14 storey, 16
storey, 18 storey, 20 storey, 22 storey, 25 storey, 30 storey and 35 storey high. Some of the
results in terms of time histories are shown in Fig. 5 to Fig. 11 for the 5, 8, 12, 16, 20, 25 and
35 storey high structures, respectively. Also the response of these structures in terms of the
peak acceleration of top floor and top floor displacements is as shown in Table 3. The
response in terms of top floor displacements and peak acceleration of top floor is shown in
Fig. 12 and Fig.13.

22
 The base isolation provided reduces the response of the structure very efficiently for
lower stories. The structures with TMD start giving better response for heights greater than
25 stories. In general, the response was similar for most earthquakes other than Kobe, 1995.
The best response reduction was for the Imperial Valley (El Centro) earthquake, 1940 and the
least response results were that of Loma Prieta, 1989. On the basis of the results, it is seen
that the hybrid passive structure performs best for storey range of 10 to 18. The results have
been diagrammatically shown in Fig. 14.
The Fast Fourier Transform (FFT) has been shown in Fig. 15 to Fig. 18. It is well
understood from the FFT that an earthquake has multiple frequencies of vibration. It is seen
that the frequency content significantly increases for the TMD structure which means that
larger number of frequencies are participating in the response reduction where as in base
isolated structure largely smaller no of frequencies are participating. This is bad in the sense
that smaller /secondary structures get affected with these smaller frequencies as they start
resonating with theses frequencies. Hence instead of main structure, the secondary structures
get affected. The addition of the base isolation reduces the response by a good margin for the
low height structures and this performance reduces as the storey height increases. It is clearly
observed that base isolated structure has lower frequency content. The structure with TMD
has lower frequency content than the fixed base structure but higher than the base-isolated
structure. The hybrid passive structure has frequency content lower than the structure with
TMD for lower heights and higher than structure with TMD for higher structures. The TMD
also reduces the response of the structure but not as much as the base isolation but this
performance increases as the height of the structure increases. The base isolation has
immediate effect on the structural response while the TMD has a delayed response; in fact in
some cases the TMD initially even increases the response marginally.
The range of effectiveness of the three aseismic systems is directly dependent on the
fundamental time period of the structure it is attached to. It has been observed that the hybrid
passive structure performs most efficiently for structures with fundamental time period
between 1.1 seconds to 2.1 seconds. For further part of the study, the structure was taken as a
regular twelve storey structure and the Loma Prieta earthquake for the time history.

3.4.2 Effect of Isolation Time Period

The effect of time period of the base isolation provided was studied to investigate its
effect on the structural response. The time period of isolation was changed by altering the
stiffness of the link element. The time period changed from the initial 2.5 seconds to 2.73

23
seconds, 2.9 seconds and 3.1 seconds. The study was restricted to 12 storey structure and an
additional Kern County time history was also applied. The results are as shown in Fig. 19 to
Fig. 25. The isolation time period has an effect on the structural response. As the isolation
time period is increased, the structural response reduces. The response reduction is also
dependent on the fundamental time period of the fixed base structure. If the fixed base
structure has considerably high flexibility, then addition of the base isolation will affect a
little change on the structural response and also each successive increment in the isolation
time period will have a comparatively lesser effect. The situation is totally different for a stiff
structure. Since, the fundamental time period of a stiff structure would be low, addition of
base isolation will considerably increase its time period, thereby reducing its structural
response. Increment of base isolation time period, like mentioned earlier, will also have a
higher response reduction for each increment step. The storey shears for the 5, 12 and 20
storey hybrid passive structures were checked to investigate the response. The results are as
shown in Fig. 22.
It is understood that the storey shears reduce as the structure height increases. The
storey shears are high for low height structures and hence the reduction in response of the
structure is much more than a taller structure which has a lesser storey shear, all other
parameters kept constant. The storey shear was reduced maximum in the base-isolated
structure for lower heights of structure while more efficiently for greater heights of structure
by the TMD. The peak base shears were also plotted and are shown in Fig. 23. Higher storey
shears in the fixed base structure means higher capacity demand of the structural members.
The results show that fixed base has high energy demand and base-isolated structure has very
low capacity demand. The TMD reduces capacity demand and it gets efficient as the height
of structure increases. The hybrid passive structure has a lower capacity demand for the range
between 10 to 18 storeys height. The hybrid passive structure performed better for the
fundamental time period range of the structure ranging from 1.1 seconds to 2 seconds.

3.4.3 Effect of Damping

The structure under investigation was taken to be a concrete structure with an
idealised damping of 5%. All the three aseismic design structures were provided a damping
of 10% in the first mode and 5% constant damping in all other modes. To study the effect of
damping on the structural response, the damping was changed to 15% in the isolator mode for
the base isolated structure, 10% for the first mode in the structure with the TMD and 15% and
10% for the first and second modes, respectively for the hybrid passive structure. The results

24
are as shown in Fig 20. The increase in damping of the isolator improves the structural
response of the base isolated structure and marginally in the hybrid passive structure also.
The increase in damping of the structure with TMD has no definite affect on the structure. A
combined effect of the change of isolation time period and the damping was also
investigated; it revealed that the base isolated structure and the hybrid passive structure
gained maximum reduction of structural response by increase of isolation time period as well
as the damping of the isolator. The result of the combined isolator time period and damping
parameter are shown in Fig. 21. Every time a change was made in the time period of the
structure, the TMD had to be tuned again to the fundamental frequency of the structure that it
was attached to. The tuning of the TMD was done with the fundamental frequency of the
fixed base structure and with the fundamental frequency of the base-isolated structure for the
structure equipped with TMD and the hybrid passive structure, respectively.

25
 CHAPTER 4
CONCLUSIONS

4.1 Summary

The seismic threat is, effect wise, more severe today, since structures are reaching
farther into the sky than yester years. The threat assumes catastrophic dimension when
coupled with the fact that in today’s world, numerous such tall structures are closely co-
located and all most of them have a capacity to house a few thousand human beings, if not
more. Hence, the necessity to equip these structures in such a way that they can bear the
seismic effects with minimal damage has always been a topic of further research. The
feasibility of using the TMD on the top floor level of a base-isolated structure to reduce the
structural response in terms of top floor accelerations, top floor displacements and storey
shears was carried out in a phased manner.
Firstly, a five storey structure was modelled as a fixed base and then the same
structure was base-isolated with laminated rubber bearings and a twelve storey fixed base
structure was then modelled with a TMD on its top floor.
Secondly, a hybrid structure was modelled and an initial numerical study was carried
out to find the range of height of structure for which the hybrid passive structure was the
most efficient aseismic design method by modelling all four types of structures ranging from
five storey height to thirty five stories.
Thirdly, some key parameters were identified for further investigation on the four
types of structures, like the time period of isolation, damping of base isolator and damping of
TMD. As a part of the study, five earthquake time histories have been considered to get the
structural response and further arrive at the conclusions.

4.2 Conclusions

The design philosophy is essentially the same in all countries and differs only superficially.
In India, the basic design philosophy is that, structure damage must be minimal during a
design basis earthquake and structural safety must be ensured for the maximum considered
earthquake. The basic parameters governing the reaction of a structure to any external

26
excitation are strength, ductility, damping and stiffness. These parameters can be altered to
check the effect on the response of the structure. Various techniques developed over the years
are based on these parameters.
The investigation carried out in a sequential manner brought out a number of
conclusions based on the results obtained, both in the form of figures and tables. Some
conclusions that can drawn from the study are
1. It can be said that the base isolation reduces the response of the structure very
efficiently for lower storeys. The structures with TMD start giving better response for
heights greater than 25 storeys. The best response reduction was for the Imperial
Valley (El Centro) earthquake, 1940 and the least response results were that of Loma
Prieta, 1989. On the basis of the results, it is seen that the hybrid passive structure
performs best for storey range of 10 to 18.
2. It is well understood from the FFT, that an earthquake excitation has multiple
frequencies of vibration. The addition of the base isolation reduces the response by a
good margin for the low height structures and this performance reduces as the storey
height increases. It is clearly observed that base isolated structure has lower frequency
content. The structure with TMD has lower frequency content than the fixed base
structure but higher than the base-isolated structure. The hybrid passive structure has
frequency content lower than the structure with TMD for lower heights and higher
than structure with TMD for higher structures. The TMD also reduces the response of
the structure but not as much as the base isolation and this performance increases as
the height of the structure increases.
3. The base isolation has immediate effect on the structural response while the TMD has
a delayed response; in fact in some cases the TMD initially even increases the
response marginally.
4. The range of effectiveness of the three aseismic systems is directly dependent on the
fundamental time period of the structure it is attached to. It has been observed that the
hybrid passive structure performs most efficiently for structures with fundamental
time period between 1.1 seconds to 2.1 seconds.
5. The isolation time period has an effect on the structural response. As the isolation
time period is increased, the structural response reduces. The response reduction is
also dependent on the fundamental time period of the fixed base structure. If the fixed
base structure has considerably high flexibility, then addition of the base isolation will
affect a little change on the structural response and also each successive increment in

27
 the isolation time period will have a comparatively lesser effect. The situation is
totally different for a stiff structure. Since, the fundamental time period of a stiff
structure would be low, addition of base isolation will considerably increase its time
period, thereby reducing its structural response.
6. The storey shears for the 5, 12 and 20 storey hybrid passive structures were checked
to investigate the response. It is understood that the storey shears reduce as the
structure height increases. The storey shears are high for low height structures and
hence the reduction in response of the structure is much more than a taller structure
which has a lesser storey shear, all other parameters kept constant. The storey shear
was reduced maximum in the base-isolated structure for lower heights of structure
while more efficiently for greater heights of structure by the TMD. The hybrid passive
structure performed well for a fundamental time period range of 1.1 seconds to 2
seconds of the structure.
7. The increase in damping of the isolator improves the structural response of the base
isolated structure and marginally in the hybrid passive structure also. The increase in
damping of the structure with TMD has no definite effect on the structure. A
combined effect of the change of isolation time period and the damping was also
investigated; it revealed that the base isolated structure and the hybrid passive
structure gained maximum reduction of structural response by increase of isolation
time period as well as the damping of the isolator.

4.3 Scope for Future Work

The tuned mass damper performs well on the top floor level. The results have given
clear conclusions in some cases and some results do not give direct conclusions. Some
important work has been completed and some areas require further work. It can be said about
the total work on the topic that objectives 1 to 3 have been achieved and the some work that
needs to be done includes
1. To assess the performance of the hybrid passive structure when the tuned mass damper is
placed on the base mass of the isolator.
2. To further carry out parametric study for
Variation in time period of isolation.
Variation in the tuning frequency of the tuned mass damper.

28
 REFERENCES

1. Ormondroyd, J. and Den Hartog, J.P., “The theory of dynamic vibration absorber”,
Trans.,ASME, 1928.
2. Crandall, S.H. and Marks, W.D.,”Random vibration in mechanical systems”,
Academic Press, New York, 1963.
3. Buckle, I.G. and Mayes, R.L., “Seismic isolation: history, application, and
performance—a world view”, Earthquake Spectra, Vol. 6, 1990.
3. Pan, T.C. and Kelly, J.M., “Aseismic base isolation: review and bibliography”, Soil
Dynamics and Earthquake Engineering, Vol. 5, 1983, 1984.
4. Younis, C.J and Tadjbakhsh, I.G., “Response of a Sliding Rigid Structure to Base
Excitation”, ASCE Journal of Engineering Mechanics, 110, 1984.
5. Kelly, J.M., “Aseismic Base Isolation”, The Shock and Vibration Digest, 1986.
6. Su, Chao, “Response of frictional base isolation systems to horizontal-vertical random
earthquake excitations”, Probabilistic Engineering Mechanics, Vol. 3, 1988.
7. Kelly, J.M., “Application of fractional derivatives to seismic analysis of base-isolated
models”, Earthquake Engineering & Structural Dynamics, Vol. 19, 1990.
8. Jangid, R.S. and Datta, T.K.,”Seismic behaviour of base-isolated buildings: a state-of-
the-art review”, 1995.
9. Naeim, F. and Kelly, J.M., “Design of Seismic Isolated Structures”, John Wiley, New
York, 1999.
10. Kelly, J.M. and Jangid, R.S., “Base isolation for near-fault motions”, Earthquake
Engineering & Structural Journal, 2000.
11. Soong, T.T. and Spencer, B.F., “Supplemental energy dissipation: state-of-the-art and
state-of-the-practice”, Engineering Structures, 2002.
12. Kulkarni, J.A. and Jangid, R.S., “Effects of superstructure flexibility on the response
of base-isolated structures”, Shock and Vibration Journal, 2002, 2003.
13. Cheung, Loh and Lin, “State of the art of structural control”, Journal of Structural
Engineering, 2003.
14. Matsagar, V.A. and Jangid, R.S.,” Seismic response of base-isolated structures during
impact with adjacent structures”, Engineering Structures, 2003.
15. Matsagar, V.A. and Jangid, R.S.,”Influence of isolator characteristics on the response
of base-isolated structures”, Engineering Structures, 2004.

29
16. Matsagar, V.A. and Jangid, R.S.,”Base isolation for Seismic Retrofitting of
Structures”, Practice periodical on structural design and construction, Vol 13, 2008.
17. Iwanami, K. and Seto, K., “An optimum design method for the dual dynamic damper
and its effectiveness”, Bulletin of the JSME, 1984.
18. Igusa, T. and Xu, K., “Dynamic characteristics of multiple substructures with closely
spaced frequencies”, Earthquake Engr. and Structural Dyn., Vol. 21, 1992, 1994.
19. Yamaguchi, H. and Harnpornchai,N., “Fundamental Characteristics of Multiple
Tuned Mass Dampers for Suppressing Harmonically Forced Oscillations”,
Earthquake Engr. and Structural Dyn., Vol. 22, 1993.
20. Ankireddi, S. and Yang, H.T.Y., “Simple ATMD control methodology for tall
buildings subject to wind loads”, Journal of Structural Engineering, 1996.
21. Soong, T.T. and Dargush, G.F., “Passive Energy Dissipation Systems in Structural
Engineering”, John Wiley & Sons, New York., 1997.
22. Rana, R. and Soong, T.T., “Parametric study and simplified design of tuned mass
dampers”, Engineering Structures, 1998.
23. Jangid, R., ‘‘Optimum Multiple Tuned-Mass-Dampers for Based-Excited Undamped
Systems’’, Earthquake Eng. Struct. Dyn., pp. 1041–1049, 1999.
24. Spencer, B.F. Jr, Nagarajaiah, S., “State of the art of structural control”, Journal of
Structural Engineering, 2003.
25. Sladek, J.R. and Klingner, R.E., “Effect of tuned-mass dampers on seismic response”,
Journal of Structural Engineering, ASCE, 1983.
26. Kelly, J.M., “Aseismic Base Isolation”, The Shock and Vibration Digest, 1986.
27. Hsiang-Chuan T., “The effect of tuned mass dampers on the seismic response of base
isolated structures”, Int. J. Solid Structures , Vol 32, No. 8/9, 1995.
28. Loh C.H. and Chao C.H., “Effectiveness of active tuned mass damper and seismic
isolation on vibration control of multi-storey building”, Journal of Sound and
Vibration 193(4), 1996.
29. Soong, T.T. and Spencer, B.F., “Supplemental energy dissipation: state-of-the-art and
state-of-the-practice”, Engineering Structures, 2002.
30. Taniguchi, “Effect of tuned mass damper on displacement demand of base-isolated
structures” , Engineering Structures 30, 2008.

30
 mn mn mn
mn k k k
k xn xn
xn xn n
n
n
mn-1
n
mn-1 mn-1
mn-1 Kn- Kn- Kn-
Kn- xn-1 xn-1 xn-1
xn-1 1 1
1
1
M
m2 m2 m2
s
m2 k k k
k x2 x2
C x2 x2
2 m1
2 m1 2 m1
2 m1 k k k
s k x x
x1 x mb1 1 mb 1
1 mb 1 1
x ẍ
1
K
ẍ x ẍ
b
xbẍ b
s g g
g
g

31