STRUCTURAL DYNAMICS

CHAPTER ONE 1
INTRODUCTION

1.1 FUNDAMENTAL OBJECTIVE OF STRUCTURAL DYNAMIC

ANALYSIS
Structural dynamics, is a type of structural analysis which covers the behavior of a structure
subjected to dynamic (actions having high acceleration) loading. The primary purpose of
structural dynamics is to analyze the stresses and deflections developed in any given type of
structure when it is subjected to an arbitrary dynamic loading. In one sense, this objective
may be considered to be an extension of standard methods of structural analysis, which
generally are concerned with static loading only, to permit consideration of dynamic loading
as well. In this context, the static loading condition may be looked upon merely as a special
form of dynamic loading. However, in the analysis of linear structure, it is convenient to
distinguish between the static and the dynamic components of the applied loading, to evaluate
the response to each type of loading separately, and then to superpose the two response
components to obtain their total effect. When treated thusly, the static and the dynamic
methods of analysis are fundamentally different in character.
The term dynamic may be defined simply as time-varying; thus a dynamic load is any
load of which its magnitude, direction, and/or position varies with time. Similarly, the
structural response to a dynamic load, i.e., the resulting stresses and deflections, is also time-
varying, or dynamic.
Two basically different approaches are available for evaluating structural response to
dynamic loads: deterministic and non-deterministic. The choice of method to be used in any
given case depends upon how the loading is defined. If the time variation of loading is fully
known, even though it may be highly oscillatory or irregular in character, it will be referred to
herein as prescribed dynamic loading; and the analysis of the response of any specified
structural system to a prescribed dynamic loading is defined as a deterministic analysis. On
the other hand, if the time variation is not completely known but can be defined in a statistical
sense, the loading is termed as random dynamic loading; and its corresponding analysis of
responses defined as a non-deterministic analysis.
In general, structural response to any dynamic loading is expressed basically in terms
of the displacements of the structure. Thus, a deterministic analysis leads directly to
displacement time-histories corresponding to the prescribed loading history; other related
response quantities, such as stresses, strains, internal forces, etc., are usually obtained as a
secondary phase of the analysis. On the other hand, a non-deterministic analysis provides
only statistical information about the displacements resulting from the statistically defined
loading; corresponding information on the related response quantities are then generated
using independent non-deterministic analysis procedures.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

1.2 TYPES OF PRESCRIBED OR DETERMINISTIC LOADING

2
The prescribed or deterministic loadings are classified into two categories, ‘periodic’ and
‘aperiodic or non-periodic’.

Periodic loading: In this type of loading, the loads exhibit the same time variation
successively for a large number of cycles. The simplest form of periodic loading is a
sinusoidal variation which is termed as ‘simple harmonic’. By means of Fourier series, any
periodic loading can be represented as the sum of a series of simple harmonic components.
Hence the analysis of response to any periodic loading follows the same general procedure.
Figure 1.1 shows the example of periodic loading.

Figure 1.1 Examples of periodic loading

Non- Periodic loading or aperiodic loading: In this type of loading, the loads do not exhibit
the same time variation successively, i.e. it may be either short duration or long duration
impulsive loadings. A blast or explosion is a typical source of impulsive loads: for short-
duration, special’ simplified forms of analysis may be employed, i.e. Duhamel’s integral may
be used. On the other hand, a general long-duration loading might result from an earthquake
and can be treated only by complete general dynamic analysis procedures. Figure 1.2 shows
the examples for non-periodic loading.

Figure 1.2 Example for non-periodic loading

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

1.3 DEGREES OF FREEDOM

3
The number of degrees of freedom that a vibrating system has will greatly affect how it
vibrates. In every structure, mass is continuously distributed throughout in a complex
manner. It is necessary to define the position of every mass point during vibration.
A simple definition of degrees of freedom is the number of coordinates necessary to
specify the position or geometry of mass point at any instant during its vibration. All real
structures possess infinite number of dynamic degrees of freedom. Hence, infinite number of
coordinates is necessary to specify the position of the structure completely at any instant of
time. Each degree of freedom is having corresponding natural frequency. Therefore, a
structure possesses as many natural frequencies as it has the degrees of freedom. For each
natural frequency, the structure has its own way of vibration. The vibrating shape is known as
characteristic shape or mode of vibration.
Consider a rigid block as shown in figure 1.3 that is free to move in 3-dimensional space,
which may move without rotation in each of the three directions X, Y and Z. These are called
the three degrees of translation.
The block may also rotate about its own axes; these are called the three degrees of
rotation.

Figure 1.3 Rigid block

Thus, to uniquely define the position of the block in space, we need to define six coordinates,
i.e. three for translation and three for rotation.
Depending on the independent coordinates required to describe the motion, the vibratory
system is divided into the following categories.
I. Single Degree of Freedom system (SDOF system)
II. Multiple Degrees of Freedom system (MDOF system)
III. Continuous system

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

If a single coordinate is sufficient to define the position or geometry of the mass of the
system at any instant of time, it is known as a single or one degree of freedom of freedom 4
system. Examples for the SDOF system are given in figure 1.4.

Figure 1.4 Examples of SDOF system

If more than one independent coordinate is required to completely specify the position or
geometry of different masses of the system at any instant of time, it is called multiple
degrees of freedom system. Examples for MDOF system are given in figure 1.5.

Figure 1.5 Examples of MDOF system

If the mass of a system may be considered to be distributed over its entire length as shown in
figure 1.6, in which the mass is considered to have infinite degrees of freedom, it is referred
to as a continuous system. It is also known as a distributed system.

Figure 1.6 A cantilever beam (with an infinite number of degrees of freedom)

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

1.4 METHODS OF DISCRETIZATION

5
Lumped-mass procedure
An analysis of the dynamic system in fig. 1.7(b) is obviously made complicated by the fact
that the inertial forces result from structural time-varying displacements which in turn are
influenced by the magnitudes of inertial forces.

Figure 1.7 basic differences between static and dynamic loads: (a) static loading;
(b) dynamic loading.

This closed cycle of cause and effect can be attacked directly only by formulating the
problems in terms of differential equations. Furthermore, because the mass of the beam is
distributed continuously along its length, the displacements and accelerations must be defined
for each point along the axis if the inertial forces are to be completely defined. In this case,
the analysis must be formulated in terms of partial differential equations because the position
along the span as well as time must be taken as independent variables.
However, if one assumes the mass of the beam to be concentrated at discrete points as
shown in fig. 1.8, the analytical problem becomes greatly simplified because inertial forces
develop only at these mass points. In this case, it is necessary to define the displacements and
accelerations only at these discrete locations.

Figure 1.8 Lumped-mass idealization of a simple beam

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

The number of displacement components which must be considered in order to represent

the effects of all significant inertial forces of a structure may be termed as the number of 6
dynamic freedom of the structure. For example, if the three masses in the system of fig 1.8 are
fully concentrated and are constrained so that the corresponding mass points translate only in
a vertical direction, this would be called a three-degree-of-freedom (3 DOF) system. On the
other hand, if these masses are not fully concentrated so that they posses finite rotational
inertia, the rotational displacements of the three points will also have to be considered, in
which case the system has 6 DOF. If axial distortions of the beam are significant, translation
displacements parallel with the beam axis will also result giving the system 9 DOF. More
generally, if the structure can deform in three-dimensional space, each mass will have 6 DOF;
then the system will have 18 DOF. However, if the masses are fully concentrated so that no
rotational inertia is present, the three-dimensional system will then have 9 DOF. On the basis
of these considerations, it is clear that a system with continuously distributed mass, as in fig.
1.7(b), has an infinite number of degrees of freedom.

Generalized Displacements
The lumped-mass idealization described above provides a simple means of limiting the
number of degrees of freedom that must be considered in conducting a dynamic analysis of
an arbitrary structural system. The lumping procedure is most effective in treating systems in
which a large proportion of the total mass actually is concentrated at a few discrete points.
Then the mass of the structure which supports these concentrations can be included in the
lumps, allowing the structure itself to be considered weightless.
However, in cases where mass of the system is quite uniformly distributed throughout,
an alternative approach to limiting the number of degrees of freedom may be preferable. This
procedure is based on the assumption that the deflected shape of the structure can be
expressed as the sum of a series of specified displacement patterns; these patterns then
become the displacement coordinates of the structure. A simple example of this approach is
the trigonometric-series representation of the deflection of a simple beam. In this case, the
deflection shape may be expressed as the sum of independent sine-wave contributions, as
shown in fig. 1.9, or in mathematical form,

∞
𝑛𝜋𝑥
v(x) = ∑ (𝑏𝑛 sin ) ………………………………….(1-1)
𝑛=1 𝐿

In general, any arbitrary shape compatible with the prescribed support conditions of the
simple beam can be represented by this infinite series of sine-wave components. The
amplitudes of the sine-wave shapes may be considered to be the displacement coordinates of
the system, and the infinite numbers of degrees of freedom of the actual beam are represented
by infinite number of terms included in the series. The advantage of this approach is that a
good approximation to the actual beam shape can be achieved by a truncated series of sine-
wave components; thus a 3 DOF approximation would contain only three terms in the series,
etc.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

Figure 1.9 Sine series representation of simple beam deflection.

This concept can be further generalized by recognizing that the sine-wave shapes used as the
assumed displacement patterns were an arbitrary choice in this example. In general, any
shapes ѱn(x) which are compatible with the prescribed geometric-support conditions and
which maintain the necessary continuity of internal displacements may be assumed. Thus a
generalized expression for the displacements of any one-dimensional structure might be
written
………………………………………….. (1.2)

For any assumed set of displacement functions ѱn(x), the resulting shape of the structure
depends upon the amplitude term Zn, which will be referred to as generalized coordinates.
The number of assumed shape patterns represents the number of degrees of freedom

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

considered in this form of idealization. In general, better accuracy can be achieved in a

dynamic analysis for a given number of degrees of freedom by using the shape-function 8
method of idealization rather than the lumped-mass approach. However, it also should be
recognized that greater computational effort is required for each degree of freedom when
such generalized coordinates are employed.

The Finite-Element Concept

A third method of expressing the displacements of any given structure in terms of a finite
number of discrete displacement coordinated, which combines certain features of both
lumped-mass and the generalized-coordinate procedures, has now become popular. This
approach, which is the basis of the finite-element method of analysis of structural continua,
provides a convenient and reliable idealization of the system and is particularly effective in
digital-computer analyses.
The finite-element type of idealization is applicable to structures of all types: framed
structures, which comprise assemblages of one-dimensional members (beams, columns, etc.);
plane-stress, plate- and shell-type structures, which are made up of two-dimensional
components; and general three-dimensional solids. For simplicity, only the one-dimensional
type of structural components will be considered in the present discussion, but the extension
of the concept to two- and three-dimensional structural elements is straightforward.

1.5 CONCEPT OF DAMPING

In a conservative system, the total energy at any instant is constant, i.e. there is a continuous
transformation between the potential energy and the kinetic energy. Whereas, in case of a
non-conservative system, damping forces are present, which dissipates (loss) energy from the
system. Therefore, to maintain the energy at a constant level, an external source must supply
energy to the system at a rate equal to the rate of energy loss or dissipation. But in case of
free vibration, it is not possible to supply energy to the system. Hence, vibration dies out
completely in due course of time.
The phenomenon in which the energy of the system is gradually reduced or the
amplitude of vibration goes on decreasing and finally the vibration of the system is
completely eliminated and the system is brought to rest is called damping. The rate of
decreasing amplitude depends upon the amount of damping. The advantage of damping is to
control the amplitude of vibration.
There are mainly five types of damping:
1. Viscous damping
2. Coulomb damping
3. Structural damping
4. Active damping or Negative damping
5. Passive damping

1.6 FORMULATION OF THE EQUATIONS OF MOTION

As mentioned earlier, the primary objective of a deterministic structural-dynamic analysis is
the evaluation of the displacement time-histories of a given structure subjected to a given

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

time-varying loading. In most cases, an approximate analysis involving only a limited

number of degrees of freedom will provide sufficient accuracy; thus, the problem can be 9
reduced to the determination of the time-histories of these selected displacement components.
The mathematical expressions defining the dynamic displacements are called the equations of
motion of the structure, and the solution of these equations of motion provides the required
displacement time-histories.
The formulation of the equations of motion of a dynamic system is possibly the most
important, and sometimes the most difficult, phase of the entire analysis procedure. In this
text, three different methods will be employed for the formulation of these equations, each
having advantages in the study of special classes of problems. The fundamental concepts
associated with each of these methods are described in the following paragraphs.

Direct Equilibration Using D’Alembert’s Principle

The equations of motion of any dynamic system represent expressions of Newton’s second
law of motion, which states that the rate of change of momentum of any mass particle m is
equal to the force acting on it. This relationship can be expressed mathematically by the
differential equation
𝑑 𝑑𝑣
p(t) = (𝑚 ) .......……………………………………..(1.3)
𝑑𝑡 𝑑𝑡
Where, p(t) is the applied force vector and v(t) is the position vector of particle mass m. For
most problems in structural dynamics it may be assumed that mass does not vary with time,
in which case Eq. (1.3) may be written
𝑑2 𝑣
p(t) = 𝑚 𝑑𝑡 2 = 𝑚𝑣̈ (𝑡)............................................................(1.3a)
Where, the dots represent differentiation with respect to time. Equation (1.3a), indicating that
force is equal to the product of mass and acceleration, may also be written in the form
p(t) ─ 𝑚𝑣̈ (𝑡) = 0............................................................(1.3b)
in which case, the second term 𝑚𝑣̈ (𝑡) is called the inertial force resisting the acceleration of
the mass.
The concept that a mass develops an inertial force proportional to its acceleration and
opposing it is known as D’Alembert’s principle. It is a very convenient device in problems of
structural dynamics because it permits the equations of motion to be expressed as equations
of dynamic equilibrium. The force p(t) may be considered to include many types of force
acting on the mass: elastic constraints which oppose displacements, viscous forces which
resist velocities, and independently defined external loads. Thus if an inertial force which
resists acceleration is introduced, the equation of motion is merely an expression of
equilibration of all forces acting on the mass. In many simple problems, the most direct and
convenient way of formulating the equations of motion is by means of such direct
equilibrations.

Principle of Virtual Displacements

However, if the structural system is reasonably complex involving a number of
interconnected mass points or bodies of finite size, the direct equilibration of all the forces
acting in the system may be difficult. Frequently, the various forces involved may readily be

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

expressed in terms of the displacement degrees of freedom, but their equilibrium

relationships may be obscure. In this case, the principle of virtual displacements can be used 10
to formulate the equations of motion as a substitute for the direct equilibrium relationships.
The principle of virtual displacements may be expressed as follows. If a system
which is in equilibrium under the action of a set of externally applied forces is subjected to a
virtual displacement, i.e., a displacement pattern compatible with the system’s constraints, the
total work done by the set of forces will be zero. With this principle, it is clear that the
vanishing of the work done during a virtual displacement is equivalent to a statement of
equilibrium. Thus, the response equations of a dynamic system can be established by first
identifying all the forces acting on the masses of the system, including inertial forces defined
in accordance with D’Alembert’s principle. Then, the equations of motion are obtained by
separately introducing a virtual displacement pattern corresponding to each degree of
freedom and equating the work done to zero. A major advantage of this approach is that the
virtual-work contributions are scalar quantities and can be added algebraically, whereas the
forces acting on the structure are vectorial and can only be superposed vectorially.

Variational Approach
Another means of avoiding the problems of establishing the vectorial equations of
equilibrium is to make use of scalar quantities in a variational form known as Hamilton’s
principle. Inertial and elastic forces are not explicitly involved in this principle; instead,
variations of kinetic and potential energy terms are utilized. This formulation has the
advantage of dealing only with purely scalar energy quantities, whereas the forces and
displacements used to represent corresponding effects in the virtual-work procedure are all
vectorial in character, even though the work terms themselves are scalars. It is of interest to
note that Hamilton’s principle can also be applied to statics problems. In this case, it reduces
to the well-known principle of minimum potential energy so widely used in static analyses.
It has been shown that the equation of motion of a dynamic system can be formulated
by any one of three distinct procedures. The most straightforward approach is to establish
directly the dynamic equilibrium of all forces acting in the system, taking account of inertial
effects by means of D’Alembert’s principle. In more complex systems, however, especially
those involving mass and elasticity distributed over finite regions, a direct vectorial
equilibration may be difficult, and work or energy formulations which involve only scalar
quantities may be more convenient. The most direct of these procedures is based on the
principle of virtual displacements, in which the forces acting on the system are evaluated
explicitly but the equations of motion are derived by consideration of the work done during
appropriate virtual displacements. On the other hand, the alternative energy formulation,
which is based on Hamilton’s principle, makes no direct use of the inertial or conservative
forces acting in the system; the effects of these forces are represented instead by variations of
the kinetic and potential energies of the system. It must be recognized that all three
procedures are completely equivalent and lead to identical equations of motion. The method
to be used in any given case is largely a matter of convenience and personal preference; the
choice generally will depend on the nature of the dynamic system under consideration.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

CHAPTER TWO 11
SINGLE DEGREE OF FREEDOM SYSTEM

2.1 VIBRATION ANALYSIS

The vibration analysis of a real system consists of the following steps:
➢ Mathematical modeling of a real system: It is done to determine the nature of the
system, its features and aspects and the components involved in the real system.
➢ Formulation of governing equations: After developing the mathematical model, the
differential equations that govern the behavior of the vibrating system are obtained by
using the law of mechanics and the principles of dynamics like Newton’s second law
of motion, D’Alembert’s principle and the principle of conservation of energy.
➢ Solution of the governing equation of motion: the governing equations of motion of
a vibrating system are solved to find the response of the vibration of an SDOF system
lead to one ordinary differential equation of motion in the form of second order linear
differential equation of motion. There are many techniques available for solving the
equation of motion, such as ordinary differential equation, matrix methods, finite
element methods and numerical methods.
➢ Interpretation of results: the solution of the equation of motion for the real physical
system generally gives the displacement, velocity and acceleration of the various
masses or inertia of the system. Interpretation of the results is an important and final
step in the various analysis procedures. It includes drawings (which give the general
interferences from results), development of design curves and recommendations if
any.

2.2 MATHEMATICAL MODELLING OF AN SDOF SYSTEM

Let us consider a simple portal frame as shown in fig 2.1. While developing a
mathematical model, some assumptions are made to simplify the analysis. They are:

Figure 2.1 Portal frame

• The total mass of a portal frame is assumed to act at the slab level, since the
masses of columns are very less when compared to that of slab; i.e., masses of
columns are ignored.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

• The beam/slab is assumed as infinitely rigid, so that the stiffness of the structure is
provided only by column, i.e., flexibility of slab/beam is ignored. 12
• Since the beam is usually built monolithically within the columns, the beam
column joint can be assumed to be rigid as without any rotations at joint.

By these assumptions, the possibility of lateral deformation or displacement is due to only

rigid beam/slab. The model resulting from all above mentioned assumptions is called as
shear building model. This shear building idealization although unrealistic is necessary for
mathematical formulation of vibration problems.

Figure 2.2 Response of SDOF system

The portal frame of fig. 2.1 under the influence of a lateral load F(t) can be represented
mathematically as the response of SDOF system as shown in fig. 2.2. This typical discrete
spring-mass system is equivalent to the response of a portal frame. The parameters of the
mathematical model are related to its prototype (actual frame) as:
i. Mass m representing the total mass of the beam and the slab of the frame and
inertial characteristics of the structure, energy is stored by mass in the form of
kinetic energy.
ii. Stiffness of spring k represents the combined stiffness of two columns for lateral
deformation that is the elastic restoring force and it stores the potential energy
(internal strain energy) due to columns.
iii. Dashpot having damping coefficient C represents the energy dissipation, i.e., the
frictional characteristics and energy losses of the frame and
iv. An excitation force F(t) representing the external lateral force applied on the
portal frame.
From above information, it is seen that the following four elements are the most
important to determine the dynamic behavior.
a. The inertia force (= 𝑚𝑥̈ )
b. The restoring force or spring force (= kx)
c. The damping force (𝑐𝑥̇ )
d. The exciting force (= F(t))
Considering equilibrium of all the forces in X direction, the governing equation of motion
for the SDOF system is
𝑚𝑥̈ + 𝑐𝑥̇ + 𝑘𝑥 = 𝐹(𝑡)..................................................... (2.1)

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

2.3 UNDAMPED FREE VIBRATION ANALYSIS OF A SDOF

13
SYSTEM
In an undamped system, the frictional forces or damping is also neglected as shown in fig.
2.3. Under these conditions, the vibration is influenced only by the initial conditions such
as the given displacement and at time t=0 when the motion of the system is initiated. This
undamped SDOF system is known as undamped oscillator.

Figure 2.3 Free vibrations of undamped SDOF system

Before writing the equation of motion, the most important practice to follow in any
dynamic analysis is to draw a free body diagram of the system. Free body diagram of a
mass can be obtained by isolating the mass and representing all externally applied forces,
reactive forces and the inertia forces.

Figure 2.4 Free body diagram

Figure 2.4 represents the FBD of the mass of the oscillator, displaced in the X direction
that is inertia force Ft and acted upon by spring force Fs. since the spring used in this
system is assumed to behave linearly, the deformation is directly proportional to the
force, i.e.
Fs ∝ 𝑥
The constant of proportionality between the force and displacement of a linear spring is
known as spring constant or stiffness of spring k. Therefore,
𝐹𝑠 = 𝑘𝑥
According to Newton’s second law, 𝐹 = 𝑚𝑎 = 𝑚𝑥̈
Hence inertia force 𝐹𝑡 = 𝑚𝑥̈
The equation of motion of such an undamped free vibration system can be written as,
𝐹𝑡 + 𝐹𝑠 = 0
𝑚𝑥̈ + 𝑘𝑥 = 0 .............................................. (2.2)

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

2.3.1 DERIVATION OF EQUATION OF MOTION

14
a) Newton’s second law of motion
Consider a spring mass system of fig. 2.5 which is assumed to move only along the
vertical direction. It has only one degree of freedom, because its motion is described by a
single coordinate x. a massless spring of constant stiffness or spring factor k is shown in
fig. 2.5. Stiffness is defined as “the load required to produce unit deformation”.
Mathematically, it can be represented as
𝑤
𝑘 = ∆ ............................................................... (2.3)
Where, Δ is the static deflection of the spring. After attaching a load w to the spring as in
fig. 2.5, the spring elongates or displaces from its equilibrium position vertically
downwards. This position is called as “equilibrium position”.
From eq. (2.3), 𝑤 = 𝑘∆ ........................................................ (2.4)
From the equilibrium position, the load w is pulled down a little by some force and then
the pulling force is removed. The load w will continue to execute vibrations up and down
which is called free vibration.
∴ Restoring force in x direction = 𝑤 − 𝑘(∆ + 𝑥)
= 𝑤 − 𝑘∆ − 𝑘𝑥
We know that, from eq. (2.4), 𝑤 = 𝑘∆
= 𝑘∆ − 𝑘∆ − 𝑘𝑥
= −𝑘𝑥
According to Newton’s law, 𝑚𝑥̈ = −𝑘𝑥
⟹ 𝑚𝑥̈ + 𝑘𝑥 = 0
𝑘
⟹ 𝑥̈ + 𝑚 𝑥 = 0 (dividing by m)....................... (2.5)

Compared with 𝑥̈ + 𝜔𝑛2 𝑥 = 0, we get

𝑘
𝜔𝑛2 = 𝑚

𝑘𝑔. 𝑚/𝑠 2
𝑘 𝑁/𝑚 √ 𝑚 1 1
𝜔𝑛 = √ = √ = =√ 2=
𝑚 𝑘𝑔 𝑘𝑔 𝑠 𝑠

The constant of proportionality 𝜔𝑛 is known as the natural frequency or angular velocity. It

is measured in radian/second.

b) Energy method
According to the law of conservation of energy,
Total energy = Constant
i.e. KE + PE = Constant

where, KE is kinetic energy and PE is potential energy.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

The time rate of change of total energy will be zero.

15
𝑑
i.e., (𝐾𝐸 + 𝑃𝐸) = 0
𝑑𝑡

kinetic and potential energies for systems shown in fig. 2.5 are given by
1 1
𝐾𝐸 = 2 𝑚𝑣 2 = 2 𝑚𝑥̇ 2

1
𝑃𝐸 = 2 𝑘𝑥 2

𝑑 1 1
So, ( 𝑚𝑥̇ 2+ 2 𝑘𝑥 2 ) = 0
𝑑𝑡 2

1 1
Hence, 𝑚 × 2𝑥̇ × 𝑥̈ + 2 × 𝑘 × 2𝑥 × 𝑥̇ = 0
2

𝑚𝑥̇ 𝑥̈ + 𝑘𝑥𝑥̇ = 0

Dividing by 𝑥̇ , 𝑚𝑥̈ + 𝑘𝑥 = 0

This is the same equation as obtained by Newton’s method.

2.4 DAMPED FREE VIBRATION ANALYSIS

In previous topic, we studied the equation of motion of a SDOF system where the
vibration persists without any external damping force under the idealized conditions. The
system which vibrates indefinitely with constant amplitude at its natural frequency.
However in reality such a vibration without decrease in amplitude is never realized.
Frictional forces (or) damping forces are always present in any physical system while
undergoing motion. The presence of damping forces or frictional forces, form a
mechanism through which the mechanical energy of the system: kinetic energy or
potential energy is transformed to other forms of energy such as heat energy. The
mechanism of this energy transformation is called dissipation of energy. This is quite
complex in nature.

2.4.1 TYPES (OR) NATURE OF DAMPING

I. Viscous damping
When the system is made to vibrate in a surrounding viscous medium that is under the
control of highly viscous fluid, the damping is called viscous damping. Viscosity is
the property of a fluid by virtue of which it offers resistance to the motion of one layer
over the adjacent layer. This type of damping is achieved by means of a device called
hydraulic dashpot.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

16

Figure 2.5 Dashpot

The main components of a viscous damper or dashpot are cylinder, piston and
viscous fluid as shown in fig. 2.5.
Since this type of damping is involved with fluids, the concept of fluid mechanics is
important to determine the damping force. Let us consider that the two plates are
separated a fluid film of thickness t as shown in fig. 2.5.
The upper plate is allowed to move parallel to the fixed plate with a velocity 𝑥̇ .
The force F required for maintaining this velocity 𝑥̇ of the plate is given by
𝜇𝐴
𝐹= 𝑥̇ ......................................................... (2.6)
𝑡
Where, µ is the coefficient of absolute viscosity of the fluid.

Figure 2.6 Mechanism

Equation (2.6) can also be written as damping force, 𝐹 = 𝑐𝑥̇
Hence,
𝜇𝐴
𝑐 = ⁄𝑡
Where, c is called as damping coefficient and it has units of force over velocity (N-
s/m).

Equation of motion for Viscous Damping

Asst. Prof. Amrendra Swarnakar
 STRUCTURAL DYNAMICS

Let us consider a mass m attached from one end of the linear spring having stiffness of k
and also connected by means of a dashpot or damper that provides viscous damping with 17
a damping coefficient c as shown in fig. 2.7. The force provided by the dashpot is equal to
𝑐𝑥̇ .

Figure viscous damped oscillator free body diagram

From the FBD, the governing differential equation of motion can be written as

𝑚𝑥̈ + 𝑐𝑥̇ + 𝑘𝑥 = 0 ......................................................... (2.7)

Assuming the solution may be in the form of

𝑥 = 𝑒 𝜆𝑡

Where, λ is a constant to be determined. This exponential function leads to algebraic

equation instead of differential equation.

𝑥̇ = 𝜆𝑒 𝜆𝑡
or,
𝑥̈ = 𝜆2 𝑒 𝜆𝑡

Substituting the values of x, 𝑥̇ and 𝑥̈ in eq. (2.7), we get

𝑚𝜆2 𝑒 𝜆𝑡 + 𝑐𝜆𝑒 𝜆𝑡 + 𝑘𝑒 𝜆𝑡 = 0

⟹ (𝑚𝜆2 + 𝑐𝜆 + 𝑘)𝑒 𝜆𝑡 = 0

The non trivial solution is

𝑚𝜆2 + 𝑐𝜆 + 𝑘=0
𝑐 𝑘
⟹ 𝜆2 + 𝑚 𝜆 + 𝑚 = 0 .................................................... (2.8)

The above equation is known as the characteristic equation which has two roots.

−𝑐 𝑐 2 𝑘
±√( ) −4( )
𝑚 𝑚 𝑚
𝜆1,2 = ..................................... (2.9)
2

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

−𝑐±√𝑐2 −4𝑘𝑚
= 18
2𝑚

From equation (2.9), the discriminant or radical term is,

𝑐2 4𝑘
−
𝑚2 𝑚

The roots to the characteristic equation fall into one of these cases:

• Discriminant > 0 then the system is termed as overdamped system.

• Discriminant < 0 then the system is termed as underdamped system.
• Discriminant = 0 then the system is termed as critically damped system.

Equation of motion of critically damped system

When the discriminant value becomes zero, the special value of damping present in the
system is called critical damping.

𝑐2 4𝑘
− =0
𝑚2 𝑚
𝑘
⟹ 𝑐𝑐2 = 4 𝑚 𝑚2

⟹ 𝑐𝑐 = 2√𝑘𝑚 = 2𝑚𝜔𝑛 ........................................................ (3.0)

Where, 𝑐𝑐 is critical damping coefficient.

Thus, eq. (3.0) gives the relation between critical damping coefficient and the natural
frequency.

The ratio of the actual damping to the critical damping coefficient is called as damping
ratio. It is denoted by a symbol ρ and is a dimensionless quantity. It can be written as
𝑐
𝜌=
𝑐𝑐

Damping ratio is one of the most important parameter in damping.

For a critical damping, the discriminate must be equal to zero that means c tends to 𝑐𝑐 .
Thus
𝑐
=1
𝑐𝑐

That is ρ = 1 for a critically damped system. In which roots of the characteristic equation
are equal and they are obtained from eq. (2.9). let α, β are the roots of the equation
𝑐
𝛼, 𝛽 = −
2𝑚

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

Whenever the roots are equal, the general solution is

19
𝑥(𝑡) = 𝑒 𝛼𝑡 [𝐴 + 𝐵𝑡]
−𝑐
⟹ 𝑥(𝑡) = 𝑒 2𝑚𝑡 [𝐴 + 𝐵𝑡] ......................................... (3.1)
𝑐
In which can be rearranged as,
2𝑚

𝑐 𝑐 𝑐𝑐 𝑐 2√𝑘𝑚
= × = ×
2𝑚 2𝑚 𝑐𝑐 2𝑚 𝑐𝑐

𝑐 2√𝑘𝑚
= × = 𝜌𝜔𝑛
𝑐𝑐 2𝑚

𝑐
⸫ substituting in eq. (3.1), = 𝜌𝜔𝑛
2𝑚

𝑥(𝑡) = 𝑒 −𝜌𝜔𝑛𝑡 [𝐴 + 𝐵𝑡]

For a critical damping system, ρ = 1. Thus

𝑥(𝑡) = 𝑒 −𝜔𝑛𝑡 [𝐴 + 𝐵𝑡] ............................... (3.2)

This is the general solution for a critically damped system.

Critical damping may be defined as the minimum amount of damping for which the
system will not vibrate when disturbed initially, but it will return to the equilibrium
position. This will result in non-periodic motion (i.e., simple decay). The displacement
decays to a negligible level after one natural period T.

Figure 2.7 The displacement plot of a critically damped system with positive initial
displacement and velocity

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

Equation of motion for overdamped system (𝝆 > 1)

20
In an overdamped system, the damping coefficient is higher than the amount of critical
damping that is 𝑐 > 𝑐𝑐 . Hence, α and β are real and distinct. They are,

−𝑐 𝑐2 𝑘
𝛼, 𝛽 = ±√ 2 −
2𝑚 4𝑚 𝑚

Multiplying and dividing by 𝜔𝑛 , we get

−𝑐𝑤𝑛 2
𝑐 2 𝜔𝑛 2
𝑘𝜔𝑛
𝛼, 𝛽 = ±√ 2 − 2
2𝑚𝜔𝑛 4𝑚2 𝜔𝑛 𝑚𝜔𝑛

−𝑐 𝑐2 𝑘 𝑘
= 𝜔𝑛 [ ±√ 2 − 2] ∵ 𝜔𝑛2 =
2𝑚𝜔𝑛 4𝑚2 𝜔𝑛 𝑚𝜔𝑛 𝑚

−𝑐 𝑐2 𝑘 𝑚
= 𝜔𝑛 [ ± √(2𝑚𝜔 )2
− . ] ∵ 2𝑚𝜔𝑛 = 𝑐𝑐
2𝑚𝜔𝑛 𝑛 𝑚 𝑘

−𝑐 𝑐2 𝑐
= 𝜔𝑛 [ ± √(𝑐 )2 − 1] ∵ 𝜌=
𝑐𝑐 𝑐 𝑐𝑐

⟹ 𝛼, 𝛽 = 𝜔𝑛 [−𝜌 ± √𝜌2 − 1]
The general solution of an overdamped system is

(−𝜌𝜔𝑛 +𝜔𝑛 √𝜌2 −1)𝑡 (−𝜌𝜔𝑛 −𝜔𝑛 √𝜌2 −1)𝑡

𝑥 = 𝐴𝑒 𝛼𝑡 + 𝐵𝑒 𝛽𝑡 = 𝐴𝑒 + 𝐵𝑒

(𝜔𝑛 √𝜌2 −1)𝑡 (−𝜔𝑛 √𝜌2 −1)𝑡

⟹ 𝑥 = 𝑒 (−𝜌𝜔𝑛 𝑡) [𝐴𝑒 + 𝐵𝑒 ] ................ (3.3)

Figure 2.8 The displacement plot of an overdamped system

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

The motion of an overdamped system is non-periodic, regardless of the initial conditions.

If the damping is higher, it will take long time to decay from an initial disturbance. 21

Equation of motion for underdamped system (ρ < 1)

For an underdamped system, the discriminant is negative. The discriminant is negative

only when the damping coefficient is less than the critical damping coefficient. Thus the
roots of the characteristic equation are complex (or imaginary). They are

𝛼, 𝛽 = 𝜔𝑛 (−𝜌 ± √(𝜌2 − 1))

Figure The displacement plot of the underdamped system

⟹ 𝛼 = ±𝑖𝛽 = −𝜌𝜔𝑛

And 𝛽 = 𝜔𝑛 √𝜌2 − 1

Therefore the general solution of an underdamped system is

√𝜌2 −1)𝑡 √𝜌2 −1)𝑡

𝑥 = 𝑒 −𝜌𝜔𝑛𝑡 [𝐴𝑒 (𝑖𝜔𝑛 + 𝐵𝑒 (−𝑖𝜔𝑛 ]

It can also be written as

𝑥 = 𝑒 −𝜌𝜔𝑛𝑡 [𝐴𝑒 𝑖𝜔𝑑 𝑡 + 𝐵𝑒 𝑖𝜔𝑑 𝑡 ] ......................................... (3.4)

Where, 𝜔𝑑 = 𝜔𝑛 √1 − 𝜌2 ; 𝜔𝑑 is called damped natural frequency.

Equation (3.4) can also be written as

𝑥 = 𝑒 −𝜌𝜔𝑛𝑡 [𝐴 cos 𝜔𝑑 𝑡 + 𝐵 sin 𝜔𝑑 𝑡]

Or,
𝑥 = 𝑋𝑒 −𝜌𝜔𝑛𝑡 [sin(𝜔𝑑 𝑡 + 𝜙)]

In which, the term sin(𝜔𝑑 𝑡 + 𝜙) indicates periodic motion and causes vibration and the
term 𝑋𝑒 −𝜌𝜔𝑛𝑡 indicates exponential decay.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

2𝜋 2𝜋
Damping period, 𝑇𝑑 = 𝜔 =
𝑑 𝜔𝑛 √1−𝜌2 22
Amplitude = 𝑋𝑒 −𝜌𝜔𝑛𝑡

II. Active damping or negative damping

Active damping refers to energy dissipation from the system by external means, such
as controlled actuators, etc. In this damping, the amplitude tends to increase which
lead to instability of the system. This type of supplying energy to the system is known
as negative damping.
Examples: for active or negative damping are suspension bridges, transmission line
wires.
III. Passive damping
Passive damping refers to energy dissipation within the structure by damping devices
such as isolator, by structural joints and supports or by structural member’s internal
element.

2.5 UNDAMPED AND DAMPED FORCED VIBRATION

ANALYSES
Undamped forced vibration
The equation for study is a forced spring–mass system
m𝑥̈ (t) +kx(t) =f(t).
The model originates by equating the Newton’s second law force m𝑥̈ (t) to the sum of the
Hooke’s force −kx(t) and the external force f(t). The physical model is a laboratory box
containing an undamped spring–mass system, transported on a truck as in Figure (a), with
external force f(t) =F0cosωt induced by the speed bumps.

Figure (a) An undamped spring-mass system in a box is transported on a truck.

Speed bumps on the shoulder of the road induce periodic vertical oscillations to the box.
Undamped Spring-Mass System
The forced spring-mass equation without damping is
𝐹 𝑘
𝑥̈ (t) +𝜔02 x(t) =𝑚0 cosωt, ω0=√𝑚
The general solution x(t) always presents itself in two pieces, as the sum of the
homogeneous solution xh and a particular solution xp. For ω = ω0, the general solution is

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

x(t) =xh(t) +xp(t),

xh(t) =c1cosω0t + c2sinω0t, c1,c2 constants, 23
𝐹0⁄
𝑚
xp(t) =A1cosωt, A1=𝜔2−𝜔 2
0
A general statement can be made about the solution decomposition: The solution is a sum
of two harmonic oscillations, one of natural frequency ω0 due to the spring and the other
of natural frequency ω due to the external force F0cosωt.

Damped forced vibration

Real systems do not exhibit idealized harmonic motion, because damping occurs. A
watch balance wheel submerged in oil is a key example: frictional forces due to the
viscosity of the oil will cause the wheel to stop after a short time. The same wheel
submerged in air will appear to display harmonic motion, but indeed there is friction
present, however small, which slows the motion. Consider a spring–mass system
consisting of a mass m and a spring with Hooke’s constant k, with an added dashpot or
dampener, depicted in Figure (b) as a piston inside a cylinder attached to the mass. A
useful physical model, for purposes of intuition, is a screen door with door–closer: the
closer has a spring and an adjustable piston–cylinder style dampener.

Figure (b) a spring-mass system with dampener

The dampener is assumed to operate in the viscous domain, which means that the force
due to the dampener device is proportional to the speed that the mass is moving:
F=c𝑥̇ (t).
The number c ≥ 0 is called the damping constant. Three forces act:
(1) Newton’s second law F1=m𝑥̈ (t),
(2) Viscous damping F2=c𝑥̇ (t) and
(3) The spring restoring force F3=kx(t).
The sum of the forces F1+F2+F3 acting on the system must equal the external force f(t),
which gives the equation for a damped spring–mass system
m𝑥̈ (t) +c𝑥̇ (t) +kx(t) =f(t).
The motion is called damped if c >0 and undamped if c= 0. If there is no external force,
f(t) = 0, then the motion is called free or unforced and otherwise it is called forced.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

2.6 EVALUATION OF VISCOUS DAMPING RATIO 24

Logarithmic Decrement Method

Figure 2.9 Amplitude decay for viscous damping

This method is used to measure damping in time domain. In this method, the free
vibration displacement amplitude history of a system is measured and recorded.
Figure 3.11 shows the amplitude decay curve. Logarithmic decrement is defined as the
natural logarithmic value of the ratio of two adjacent peak values of displacement in free
vibration. It is a dimensionless parameter. It is denoted by a symbol δ.
𝑥1
𝛿 = ln
𝑥2
In case of underdamped system, the general solution is given by,
𝑥 = 𝑋𝑒 −𝜌𝜔𝑛𝑡 (𝑠𝑖𝑛𝜔𝑑 𝑡 + 𝜙)
Let the displacement within one cycle is x0
Amplitude x0=𝑥𝑒 −𝜌𝜔𝑛 𝑡
Let x1 is the displacement after one cycle
Amplitude 𝑥1 = 𝑥𝑒 −𝜌𝜔𝑛 (𝑡+𝑇𝑑 )
𝑥0 𝑥𝑒 −𝜌𝜔𝑛 𝑡
⟹ =
𝑥1 𝑥𝑒 −𝜌𝜔𝑛 (𝑡+𝑇𝑑)
𝜌𝜔𝑛 𝑇𝑑
=𝑒
Taking logarithm on both sides
𝑥0
log = 𝜌𝜔𝑛 𝑇𝑑
𝑥1

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

𝑐 𝑐
𝛿 = 𝜌𝜔𝑛 𝑇𝑑 = 𝑐 𝜔𝑛 𝑇𝑑 = 2𝑚𝜔 𝜔𝑛 𝑇𝑑
𝑐
𝑐 𝑛 25
𝛿 = 2𝑚 𝑇𝑑
2𝜋
Where, 𝑇𝑑 =
𝜔𝑛 √1−𝜌2
𝑐 2𝜋 𝑐 2𝜋
𝛿 = 2𝑚 =𝑐
𝜔𝑛 √1−𝜌2 𝑐 √1−𝜌2

2𝜋𝜌
𝛿= ............................................. (3.5)
√1−𝜌2

For small values of the damping ratio, the above equation can be approximated by
𝛿 = 2𝜋𝜌 ............................................... (3.6)
Let the amplitude after n cycles be xn. This can be written as,
𝑥𝑛 = 𝑥𝑒 −𝜌𝜔𝑛(𝑡+𝑛𝑇𝑑 )
𝑥0 𝑥𝑒 −𝜌𝜔𝑛𝑡
= −𝜌𝜔 (𝑡+𝑛𝑇 ) = 𝑒 𝜌𝜔𝑛𝑛𝑇𝑑
𝑥1 𝑥𝑒 𝑛 𝑑

= 𝑒 𝑛𝛿
𝑥𝑛 = 𝑥0 𝑒 −𝑛𝛿 ..................................................... (3.7)

Half Power Bandwidth Method

This method is used to determine the damping ratio from frequency domain. Half power
bandwidth BD is defined as the ratio of the frequency range between the two half power
points to the natural frequency at this mode.
Frequency response curve is used to define the half-power bandwidth. In which, the
damping ratio is determined from the frequencies at which the response amplitude is
reduced to 1/√2 times the maximum amplitude or resonant amplitude.
Let 𝜔1 and 𝜔2 are the forcing frequencies on either side of the resonant frequency at
which the response amplitude is equal to 1/√2 times the resonant amplitude. The
bandwidth 𝜔1 and 𝜔2 are also called as half-power points.
1
We know that,𝑀𝑓 =
√(1−𝛽 2 )2 +(2𝜌𝛽)2
1
Maximum resonant amplitude =
2𝜌√1−𝜌2

Figure 2.10 Half power bandwidth

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

To obtain the frequency ratio, equating the resonant amplitude into 1/√2 times peak
value. 26

1 1 1
= [ ]
√(1 − 𝛽 2 )2 + (2𝜌𝛽)2 √2 2𝜌√1 − 𝜌2

Squaring on both sides of this equation

1 1 1
= [ 2 ]
(1 − 𝛽 2 )2 + (2𝜌𝛽) 2 2 4𝜌 (1 − 𝜌2 )

After solving, 𝛽 2 = 1 − 2𝜌2 + 2𝜌√1 − 𝜌2

In which for small values of damping, neglect the term 𝜌2 in the radical. The two roots
are
𝛽1 2 = 1 − 2𝜌2 + 2𝜌 ⟹ 𝛽1 = 1 − 𝜌 − 𝜌2

And 𝛽2 2 = 1 − 2𝜌2 − 2𝜌 ⟹ 𝛽2 = 1 + 𝜌 − 𝜌2

Subtracting 𝛽1 and 𝛽2, we get

𝛽2 − 𝛽1 = 1 + 𝜌 − 𝜌2 − (1 − 𝜌 − 𝜌2 ) = 2𝜌 ............................. (3.8)

While adding 𝛽1 and 𝛽2, we get

𝛽2 + 𝛽1 = 1 + 𝜌 − 𝜌2 + (1 − 𝜌 − 𝜌2 ) = 2(1 − 𝜌2 ) .................. (3.9)

For small values of ρ,

𝛽2 + 𝛽1 = 2

𝛽2 −𝛽1 𝛽2 −𝛽1
From eq. (3.8), 𝜌 = =
2 𝛽2 +𝛽1

𝜔2 −𝜔1
𝜌= ................................................................................... (3.10)
𝜔2 +𝜔1

Where, 𝜔1 𝑎𝑛𝑑 𝜔2 are the forcing frequencies at which the amplitudes of response are equal to 1/√2
times the maximum amplitude.

Resonant Amplification Method

This method is based on the observations (in the form of amplitudes) made from a forced vibration
test on a structure within a range of frequencies closer to resonant frequency. From this observation,
the dynamic magnification factor and frequency ratio will be calculated by using the formula. By
using these data, a frequency response curve is plotted as shown in figure 2.11.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

27

Figure 2.11 Graph magnification factor vs frequency ratio.

From this curve, the magnification factor corresponding to maximum amplitude will be used to
determine the damping ratio

1
𝜌=
2(𝑀𝑓 )𝑚𝑎𝑥
(𝑋𝑑𝑦𝑛 )𝑚𝑎𝑥
Where, 𝑀𝑓 =
𝑋𝑠𝑡𝑎𝑡𝑖𝑐

Transducers are used for measuring the amplitudes, whereas vibration generators are used to provide
a source of harmonic excitation for structures.

2.7 FORCED HARMONIC VIBRATION

Figure 2.12 Viscously damped system with harmonic excitation

Consider the SDOF system with a sinusoidally varying forcing function applied to the mass as
shown F=F0sinωt

From the Newton’s Second Law, ∑f = ma ⟹ 𝑚𝑥̈ + 𝑐𝑥̇ + 𝑘𝑥 = 𝐹0 sin 𝜔𝑡 .....................(X)

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

The solution consists of the complementary solution (homogeneous solution) and the particular
solution. The complementary part of the solution has already discussed in undamped vibration. 28

Since the vibration of the response is at the same frequency as the excitation, the particular solution
will be of the form 𝑥 = 𝑋𝑠𝑖𝑛 (𝜔𝑡 − 𝜙).

Substituting this into the differential equation, the solution is of the form
𝐹0 𝑐𝜔
𝑋= and 𝜙 = tan−1 (𝑘−𝑚𝜔2 )
√(𝑘−𝑚𝜔 2 )2 +(𝑐𝜔)2

Note that this is also seen graphically as

This is expressed in non-dimensional form as

𝐹0 𝑐𝜔
𝑋= 𝑘
and 𝜙 = tan−1 ( 𝑘
𝑚𝜔2
)
2 2 2 1−
√(1−𝑚𝜔 ) +(𝑐𝜔) 𝑘
𝑘 𝑘

and can be further reduced recalling the following expressions for a SDOF

𝑘 𝑐
𝜔𝑛 = √𝑚 , 𝑐𝑐 = 2𝑚𝜔𝑛 , 𝜌 = 𝑐
𝑐

The non-dimensional expression is

𝜔
𝑋𝑘 1 2𝜌( )
−1 𝜔𝑛
= and 𝜙 = tan ( 𝜔 2
)
𝐹0 2 2
2
1−( )
√(1−( 𝜔 ) ) +(2𝜌( 𝜔 )) 𝜔𝑛
𝜔𝑛 𝜔𝑛

This yields the popular plot of forced response,

Figure 2.12 Plot of non-dimensional expression

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

The complex force vector also yields useful information for interpretation of the results.
29

Figure 2.13 Vector relationship in forced vibration

2.8 ROTATING UNBALANCE

The effect of unbalance is a common problem in vibrating systems.

Figure 2.14 Harmonic disturbing force resulting from rotating unbalance

Consider a one-dimensional system with an unbalance represented by an eccentric mass, m, with

offset, e, rotating at some speed, ω, as shown.

Let x be the displacement of the non-rotating mass (M-m) about the equilibrium point, then the
displacement of the eccentric mass is,

𝑥 + 𝑒 sin 𝜔𝑡

and the equation of motion becomes

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

𝑑2
(𝑀 − 𝑚)𝑥̈ + 𝑚 2 (𝑥 + 𝑒 sin 𝜔𝑡) = −𝑘𝑥 − 𝑐𝑥̇
𝑑𝑡 30
This can easily be cast as

𝑀𝑥̈ + 𝑐𝑥̇ + 𝑘𝑥 = (𝑚𝑒𝜔2 ) sin 𝜔𝑡

which is essentially identical to (X) with the substitution of F0=meω2.

The steady-state solution just developed is applicable for this solution,

𝑚𝑒𝜔 2 𝑐𝜔
𝑋= and 𝜙 = tan−1 (𝑘−𝑀𝜔2)
√(𝑘−𝑀𝜔 2 )2 +(𝑐𝜔)2

The differential equation describing the system,

𝐹
𝑥̈ + 2𝜌𝜔𝑛 𝑥̇ + 𝜔𝑛2 𝑥 = 𝑚0 sin 𝜔𝑡 .................................... (3.11)

and the complete solution of this problem is given as,

𝐹0 sin(𝜔𝑡−𝜙)
𝑥(𝑡) = + 𝑋1 𝑒 −𝜌𝜔𝑛𝑡 sin(√1 − 𝜌2 𝜔𝑛 𝑡 + 𝜙1 ) ................ (3.12)
𝑘 2 2
2
√(1−( 𝜔 ) ) +(2𝜌( 𝜔 ))
𝜔𝑛 𝜔𝑛

Manipulating into non-dimensional form,

𝜔 2 𝜔
𝑀𝑋 ( ) 2𝜌( )
𝜔𝑛 −1 𝜔𝑛
= ............(3.13) and 𝜙 = tan ( 𝜔 2
) ............ (3.14)
𝑚𝑒 2 2
2
1−( )
√(1−( 𝜔 ) ) +(2𝜌( 𝜔 )) 𝜔𝑛
𝜔𝑛 𝜔𝑛

This yields the popular plot of forced response

Figure 2.14 Plot of above non-dimensional form equations

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

2.9 BASE EXCITATION

31
Many times a system is excited at the location of support commonly called ‘base excitation’.

Figure 2.15 System excited by motion of support point

With the motion of the base denoted as ‘y’ and the motion of the mass relative to the intertial
reference frame as ‘x’, the differential equation of motion becomes

𝑚𝑥̈ = −𝑘(𝑥 − 𝑦) − 𝑐(𝑥̇ − 𝑦̇ ) ...................................... (3.15)

Substitute

𝑧 =𝑥−𝑦

into the equations to give

𝑚𝑧̈ + 𝑐𝑧̇ + 𝑘𝑧 = −𝑚𝑦̈ = 𝑚𝜔2 𝑌 sin 𝜔𝑡 ........................... (3.16)

This is identical in form to equation (X) where z replaces x and mω2Y replaces F0.

Thus the solution can be written by inspection as

𝑚𝜔 2 𝑌
𝑍= ........................................................... (3.17) and
√(𝑘−𝑚𝜔 2 )2 +(𝑐𝜔)2

𝑐𝜔
𝜙 = tan−1 (𝑘−𝑚𝜔2 ) ....................................................................... (3.18)

The steady state amplitude and phase from this equation can be written as

𝑋 𝑘 2 −(𝑐𝜔)2
|𝑌 | = √(𝑘−𝑚𝜔2 )2 +(𝑐𝜔)2 .............................................................. (3.19)

𝑚𝑐𝜔 3
tan 𝜙 = (𝑘(𝑘−𝑚𝜔2 )−(𝑐𝜔)2 ) ............................................................... (3.20)

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

32

Figure 2.16 Plots of eqns. (3.19) and (3.20)

2.10 VIBRATION ISOLATION

Vibration isolation is an important technique in seismic resistant design of structures and
machine foundation. There are two different types of problem involved in vibration isolation.
They are displacement isolation and force isolation.

Displacement isolation (or) Passive isolation

In case of earthquakes, the strong ground motion is transmitted to the building as a shaking force
which is resting on ground. This will produce deformations or sometimes can lead to the collapse
of building. Thus, isolators are provided to reduce or eliminate the ground motion from the
ground to the building. This is called passive isolation.

Mathematical expression for displacement isolation:

The effective loading, 𝐹𝑒𝑓𝑓 = 𝑚𝜔2 𝑥0

The steady state response is given by

𝐹0 𝑚𝜔 2 𝑥0 1
𝑥𝑅 = 𝑀𝑓 =
𝑘 𝑘 √(1−𝛽 2 )2 +(2𝜌𝛽)2

𝜔2
( 2 )𝑥0
𝜔 𝑛
=
√(1−𝛽 2 )2 +(2𝜌𝛽)2

𝛽 2 𝑥0
𝑥𝑅 = ..................................................... (3.21)
√(1−𝛽 2 )2 +(2𝜌𝛽)2

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

2𝜌𝛽
We know that, tan 𝜙 = 1−𝛽2
33
Polygon law is used to find the displacement of seismic mass. From figure (below),

Figure 2.17 Vector form

The displacement of mass is given by

𝑥 2 = 𝑥𝑅2 + 𝑥𝑆2 + 2𝑥𝑆 + 𝑥𝑅 cos 𝜙

2
𝛽 2 𝑥0 𝛽 2 𝑥0 (1−𝛽 2 )
=[ ] + 𝑥𝑆2 + 2𝑥𝑆 [ ][ ] .................. (3.22)
√(1−𝛽 2 )2 +(2𝜌𝛽)
2 √(1−𝛽 2 )2 +(2𝜌𝛽) 2 √(1−𝛽 2 )2 +(2𝜌𝛽)2

Force Isolation (or) Active Isolation

In machine foundation, the unbalanced force in a machine is finally transmitted to the ground surface
as a shaking force through its mounts on the foundation. The periodic forces thus transmitted to the
ground can excite the machines and their parts and may cause large amplitude. Proper isolators can
be used to reduce or eliminate shaking forces transmitted to the ground. Such a system is called
active isolation.

Mathematical expression for force isolation:

Consider the damped oscillator with a harmonic force F= F0 sin 𝜔𝑡 as shown in figure 2.17(a). The
differential equation of motion is given by, 𝑚𝑥̈ + 𝑐𝑥̇ + 𝑘𝑥 = 𝐹0 sin 𝜔𝑡

The steady state response is given by,

𝐹0
𝑘
𝑋𝑑𝑦𝑛 = sin(𝜔𝑡 − 𝜙) ....................................... (3.23)
√(1−𝛽 2 )2 +(2𝜌𝛽)2

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

34

Figure 2.17(a) Isolator

2𝜌𝛽
We know that, tan 𝜙 = 1−𝛽2

The force transmitted to the support through the spring and damper. Thus the total force transmitted
to the foundation is

𝐹𝑇 = 𝑘𝑥 + 𝑐𝑥̇ ............................................. (3.24)

Differentiating eq. (q) and substitute in eq. (w)

𝐹 𝐹 1
𝐹𝑇 = 𝑘 +𝑐𝑘𝜔 cos(𝜔𝑡 − 𝜙)
𝑘√(1−𝛽 2 )2 +(2𝜌𝛽)2 √(1−𝛽 2 )2 +(2𝜌𝛽)2

2𝜌𝛽
We know that, tan 𝜙 =
1−𝛽 2

From fig. 3.17(b),

Figure 3.17 (b)

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

(1−𝛽 2 )
cos 𝜙 =
√(1−𝛽 2 )2 +(2𝜌𝛽)2 35
𝛽4 2𝛽 2 (1−𝛽 2 )
𝑥 2 = 𝑥𝑆2 [1 + (1−𝛽2 )2 +(2𝜌𝛽)2 + (1−𝛽2 )2 +(2𝜌𝛽)2 ]

(1−𝛽 2 )2 +(2𝜌𝛽)2 +𝛽4 +2𝛽2 (1−𝛽2 )

= 𝑥𝑆2 [ (1−𝛽 2 )2 +(2𝜌𝛽)2
]

1+𝛽 4 −2𝛽 2 +(2𝜌𝛽)2 +𝛽 4 +2𝛽 2 −2𝛽2

= 𝑥𝑆2 [ (1−𝛽 2 )2 +(2𝜌𝛽)2
]

1+(2𝜌𝛽)2
𝑥 = 𝑥𝑆 √( ......................................................................................... (3.25)
1−𝛽 2 )2 +(2𝜌𝛽)2

Thus the above equation gives relative transmission of the base motion to the oscillator. The ratio of
the maximum amplitude of motion of the oscillator to the amplitude of support motion is known as
transmissibility of support motion and it is denoted by TR.
𝑥 1+(2𝜌𝛽)2
𝑇𝑅 = 𝑥 [(1−𝛽2)2 +(2𝜌𝛽)2 ] ............................................. (3.26)
𝑆

Thus transmissibility mainly depends on damping ratio and frequency ratio.

The isolation system is effective only when β < √2. Hence, the transmissibility values for β > √2 are
less than those for β < √2.

Important points from transmissibility curves:

Figure 2.18 Vibration-Transmissibility ratio (applied load or displacement)

(𝜉 𝑖𝑛 𝑓𝑖𝑔𝑢𝑟𝑒 𝑖𝑠 𝑡𝑒𝑟𝑚𝑒𝑑 𝑎𝑠 𝜌 𝑖. 𝑒. 𝑑𝑎𝑚𝑝𝑖𝑛𝑔 𝑟𝑎𝑡𝑖𝑜)

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

1. All the curves for different value of ρ intersect TR = 1 line at frequency ratio β = 0 and √2.
For frequency ratios in the range (0 < β <√2 ), the transmissibility ratio > 1 for all values of
36
ρ. Thus, larger the damping ratio, smaller is the transmissibility ratio value.

2. At β = 1, the TR value is controlled by the damping ratio ρ. For undamped case, the TR is
indeed very large. In the range of β > √2, TR goes on decreasing as the frequency ratio
increases for all values of ρ.

𝐹2 𝐹2 (2𝜌𝛽)2 1+(2𝜌𝛽)2
𝐹𝑇 = √(1−𝛽2 )2 +(2𝜌𝛽)2 + (1−𝛽2 )2 +(2𝜌𝛽)2 = 𝐹√(1−𝛽2 )2 +(2𝜌𝛽)2

𝐹𝑇 1+(2𝜌𝛽)2
⟹ = 𝑇𝑅 = √(1−𝛽2 )2 +(2𝜌𝛽)2 ........................................... (3.27)
𝐹

Where, TR is transmissibility ratio. Thus ratio of the maximum amplitude, i.e. force transmitted to the
foundation to the amplitude of the applied force is known as transmissibility of the support system.

Thus force isolation and displacement isolation have the same transmissibility ratio.

In general, effectiveness of the vibration-isolation system is = 1− 𝑇𝑅

∴ 𝑇𝑅 = 1 only when 𝐹 = 𝐹𝑇

1+(2𝜌𝛽)2
1 = √(1−𝛽2 )2 +(2𝜌𝛽)2

⟹ (1 − 𝛽 2 )2 + (2𝜌𝛽)2 = 1 + (2𝜌𝛽)2

⟹ 𝛽 = ±√2 ................................................. (3.28)

2.12 VIBRATION MEASURING INSTRUMENTS

Vibration measuring instruments are also called as seismic instruments or seismometer.
Measurement of ground movement during an earthquake gives the necessary data for seismic
analysis and the records of the ground motions of a structure helps us to evaluate how the
structures behave during earthquakes. Seismograph is a graphical plot which is used to measure
either the displacement or acceleration.

SEISMOMETER

Basic element of seismometer is in the form of a transducer which consists of a mass-spring-

damper system mounted on a rigid frame and that is attached to the surface where the motion is
to be measured as shown in figure below.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

37

Figure 2.19 Seismometer

When the system is made to vibrate, the mass moves relative to the frame and this relative
displacement is recorded. This measured relative displacement provides the corresponding
support motion, i.e. acceleration or displacement. One of disadvantages of the seismometer is
its large size.

Displacement measurement: Let us consider the response of the instrument to a harmonic

base displacement of

𝑥𝑆 = 𝑋𝑆 sin 𝜔𝑡

The corresponding acceleration is given by

𝑥𝑆̈ = −𝜔2 𝑋𝑆 sin 𝜔𝑡

The equation of motion for this system is

𝑚𝑥̈ + 𝑐𝑥̇ + 𝑘𝑥 = −𝑚𝑥𝑆̈

Effective loading 𝐹𝑒𝑓𝑓 = −𝑚𝑥𝑆̈ = 𝑚(𝜔2 𝑋𝑆 sin 𝜔𝑡) = 𝑚𝜔2 𝑋𝑆 sin 𝜔𝑡

Therefore, the relative dynamic steady-state response is

𝐹𝑒𝑓𝑓 𝑚𝜔2 𝑋𝑆 𝑀𝑓
𝑥𝑑𝑦𝑛 𝑜𝑟𝑋𝑅 = 𝑀𝑓 = sin 𝜔𝑡 = 1
𝑘 𝑘
𝜔2
= 𝜔2 𝑋𝑆 𝑀𝑓
𝑛

or 𝑋 = 𝛽 2 𝑋𝑆 𝑀𝑓 ........................................... (3.29)

Thus the dynamic steady state response is a function of 𝛽 2 𝑀𝑓 and it is plotted against β for
different damping values as shown in figure below.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

38

Figure 2.20 Response of seismometer to harmonic base displacement

This graph shows that 𝛽 2 𝐷 is constant when frequency ratio β > 1 for a damping ratio of 0.5.
Hence, the response of this instrument is proportional to the base displacement amplitude for
high frequency excitation. Since this displacementmeter is used for only high frequency
ranges. Its usage will be increased by reducing the natural frequency by means of increasing
the mass or decreasing the spring stiffness.

ACCELEROMETER

This instrument essentially consists of a viscous damped oscillator as shown in figure below.
The entire system is mounted inside a rigid frame which is attached to the base where the
vibration is to be measured. Let us consider the response of the instrument to a harmonic base
acceleration of

𝑥̈ (𝑡) = 𝑥̈ 0 sin 𝜔𝑡

The response is measured in terms of the amplitude x(t) of the mass relative to the frame. The
governing differential equation for this system is written as

𝑚𝑥̈ + 𝑐𝑥̇ 𝑅 + 𝑘𝑥𝑅 = −𝑚𝑥̈ 𝑆

Where xR is the relative displacement.

We know that,

𝑥 − 𝑥𝑅 = 𝑥𝑆

𝑥 = 𝑥𝑅 + 𝑥𝑆

𝑚(𝑥̈𝑆 + 𝑥̈ 𝑅 ) + 𝑐𝑥̇ 𝑅 + 𝑘𝑥𝑅 = −𝑚𝑥̈ 𝑆 = 𝐹𝑒𝑓𝑓

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

39

Figure 2.21 Accelerometer

Therefore, effective loading of mass m is

𝐹𝑒𝑓𝑓 = 𝑚𝑥𝑆 𝜔2 sin 𝜔𝑡

We know that, 𝑋𝑑𝑦𝑛 = 𝑋𝑠𝑡𝑎𝑡𝑖𝑐 × 𝑀𝑓

𝑚𝑋̈𝑆
𝑋𝑅 = 𝑀𝑓 .................................................... (3.40)
𝑘

1
Where, 𝑀𝑓 = ................................... (3.41)
√(1−𝛽 2 )2 +(2𝜌𝛽)2

XR is the dynamic steady-state response of the system. From the frequency response curve, as
shown in figure below, it is seen that for a damping ratio of 0.7, the value of magnification
factor is almost constant at a particular frequency range of 0 < β < 0.6. From the above eq.
(3.40) the response produced by this instrument will be directly proportional to the base or
support acceleration amplitude for forcing frequencies upto 0.6 times that the natural
frequency of the instrument.

Figure 2.22 Phase angle vs. β

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

Thus, this type of instrument will work effectively as an accelerometer for relatively low
frequencies. This instrument may be widely used when its natural frequency relatively increased 40
to the exciting frequency (resonant frequency) by means of increasing the stiffness of spring or
by decreasing the mass of oscillator.

2.13 RESPONSE TO IMPULSIVE LOADING

We normally come across many practical situations like impact pile driving, landing of aircraft
on tarmac, blasting on structures, etc., in which the dynamic excitation is neither harmonic nor
periodic. The excitation in these cases is in the form of impulse which is different from harmonic
or periodic loadings. Impulsive loading is a product of the force and its duration. Impulsive or
shock loading provides a further class of dynamic loading which is characterized by a very short
time history. In this case the structure attains its maximum response very quickly. Therefore the
damping forces present in the system do not possess adequate time to resist the motion. Because
of this reason we consider here only the undamped vibration of a system. Moreover, the actual
shock loading is very complex. We, therefore, consider here a simpler idealized shock motions.
The results obtained as response of SDOF system to such simplified shock loading are
reasonably accurate and valid for engineering applications. Generally, the shock idealization
consists of discrete impulse, step impulse, half-sine wave impulse, triangular impulse, etc.

We study the response of structures under two phases, namely, first, the forced vibration phase
called Phase I and second, the free vibration phase which follows the forced vibration. The
second phase is termed as Phase II. The maximum response depends on the duration of the
impulse. If the pulse is longer, maximum response occurs under Phase I. In the case of short
impulse the maximum amplitude is reached in Phase II. In Phase II, the response of structure
depends on the displacement and velocity of the structures existing at the end of Phase I. We can
calculate the response under impulsive loading by two methods, namely, (i) Duhamel’s integral
and (ii) direct solution of differential equation. However, we confine our discussion here only on
the method by Duhamel’s integral.

2.14 DUHAMEL INTEGRAL

In theory of vibrations, Duhamel's integral is a way of calculating the response of linear
systems and structures to arbitrary time-varying external excitations.
The response of a linear, viscously damped single-degree of freedom (SDOF) system to a
time-varying mechanical excitation p(t) is given by the following second-order ordinary
differential equation
𝑑2 𝑥(𝑡) 𝑑𝑥(𝑡)
𝑚 2
+𝑐 + 𝑘𝑥(𝑡) = 𝑝(𝑡)
𝑑𝑡 𝑑𝑡
Where, m is the (equivalent) mass, x stands for the amplitude of vibration, t for time, c for
the viscous damping coefficient, and k for the stiffness of the system or structure.
If a system is initially rest at its equilibrium position, from where it is acted upon by a unit-
impulse at the instance t=0, i.e., p(t) in the equation above is a delta function δ(t), 𝑥(0) =
𝑑𝑥
| = 0, then by solving the differential equation one can get a fundamental solution
𝑑𝑡 t=0
(known as a unit-impulse response function),

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

1
𝑒 −𝜌𝜔𝑛 𝑡 sin 𝜔𝑑 𝑡 , 𝑡>0
ℎ(𝑡) = {𝑚𝜔𝑑 41
0, 𝑡 < 0
𝑐
where, 𝜌= is called the damping ratio of the system, ωn is the natural circular
2𝑚𝜔𝑛
frequency of the undamped system (when c=0) and 𝜔𝑑 = 𝜔𝑛 √1 − 𝜌2 is the circular
frequency when damping effect is taken into account (when c ≠ 0 ). If the impulse happens
at t=τ instead of t=0, i.e. p(t) = δ(t−τ), the impulse response is
1
ℎ(𝑡 − 𝜏) = 𝑒 −𝜌𝜔𝑛 𝑡 𝑠𝑖𝑛[𝜔𝑑 (𝑡 − 𝜏)] 𝑡≥𝜏
𝑚𝜔𝑑
Regarding the arbitrarily varying excitation p(t) as a superposition of a series of impulses:
then it is known from the linearity of system that the overall response can also be broken
down into the superposition of a series of impulse-responses:
𝑥(𝑡) ≈ ∑ 𝑝(𝜏) . Δ𝜏 . ℎ(𝑡 − 𝜏)
Letting Δ𝜏 → 0, and replacing the summation by integration, the above equation is strictly
valid
𝑡
𝑥(𝑡) = ∫ 𝑝(𝜏)ℎ(𝑡 − 𝜏)𝑑𝜏
0
Substituting the expression of h(t-τ) into the above equation leads to the general expression
of Duhamel's integral
𝑡
1
𝑥(𝑡) = ∫ 𝑝(𝜏)𝑒 −𝜌𝜔𝑛(𝑡−𝜏) sin[𝜔𝑑 (𝑡 − 𝜏)]𝑑𝜏
𝑚𝜔𝑑 0

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

CHAPTER 3 42
MULTI-DEGREE OF FREEDOM SYSTEM

3.1 INTRODUCTION
The system with one and two degree of freedom discussed in the first seven chapters represented
the simple mathematical models of complex physical systems. These simple models were able to
explain the dynamics behaviour of complex systems. Most often however such idealisation are not
possible and real structure such as multi-storey frames and bridges are continuous systems and as
such possess an infinite number of degrees of freedom. A multiple degrees of freedom system is
defined as a system whose motion is described by more than one independent coordinates with each
degree of freedom having a frequency of vibration similar to a SDOF oscillator. The systems shown
in Figure 3.1 is examples for MDOF.

Figure3.1 Three degrees of freedom spring mass system

The motion of MDOF systems is generally described by a finite set of simultaneous second order
differential equations. The analysis of vibrations of an MDOF systems is more complex and time
consuming than the analysis of the vibrations of a SDOF or a two degrees of freedom system.
Because the mass, stiffness and damping matrices are couples with more dependent variables, the
equations should be solved simultaneously. However, the coupling terms can be uncoupled by means
of a coordinate transformation which results in a set of independent second-order ordinary
differential equations of motion. In this chapter, free and forced vibration analysis of MDOF systems
are presented. Also the concept of modal superposition is discussed.

3.2 FREE VIBRATION ANALYSIS

In case of free vibration, the structure is not subjected to any dynamics excitation or external forces
or support motion. The motion of structure is influenced only by the initial conditions. For a
vibrating system of ‘n’ degrees of freedom, the motion is represented by ‘n’ differential equations of
motion. The method such as Newton’s second law, Lagrange’s equation and influence coefficients
are used to form the equations of motion. The solution of these equations may be obtained by matrix
method, Stodola method and Holzer method. These are called approximately methods of determining
natural frequencies. In this chapter, matrix method is used to solve the differential equations of
motion.

UNDAMPED SYSTEM
Let us consider the response of an un damped MDOF system subjected to free vibrations. The
equations of motion of free vibrations can be derived by using Newton’s second law of motion with
the help of free body diagram.
For n degrees of freedom system, there will be n number of equations of motion and these can be
written in matrix form as,

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

𝑚11 ⋯ 𝑚1𝑛 𝑥1 𝑘11 ⋯ 𝑘1𝑛 𝑥1

𝑚
[ 21 … 𝑚2𝑛 ] { ⋮ } + [𝑘21 … 𝑘2𝑛 ] { ⋮ } = {0} 43
𝑚𝑛1 … 𝑚𝑛𝑛 𝑥𝑛 𝑘𝑛1 … 𝑘𝑛𝑛 𝑥𝑛

Or [M] {ẍ}+[k]{x}={0}
In which [M] is mass matrix
[K] is stiffness matrix
{ẍ}is acceleration vector and
{x}is displacement vector

It should be noted that the mass matrix is a diagonal matrix. The elements of the stiffness matrix are
called as stiffness coefficients. Generally, the stiffness coefficients 𝑘𝑖𝑗 is define, as the forces applied
at coordinates i when a unit displacement is given at j, while all other coordinates are being fixed.
Mass matrix [M] is coupled with stiffness matrix [k] though acceleration. Coupling is not an
intrinsic property of a vibrating system. It is dependent upon the choice of the coordinate system of
depends on the analyst.

3.3 NATURAL FREQUENCIES AND NORMAL MODES

The equation of motion of an MDOF system subjected to free vibration is given as
[M]{ẍ}+[k]{x}={0}………………………(3.1)
For an un damped free vibration systems, the solutions equation of motions are in the form of
{x} = {X sin(𝝎𝒏 t + Φ)}………………….. (3.2a)
And {ẍ} = -𝝎𝟐𝒏 {X sin(𝝎𝒏 t + Φ)}……………... (3.2b)
Where X is the amplitude of motion. After substituting of Eq. (8.2) into Eq. (8.1), we get

-𝝎𝟐𝒏 [M] {X sin(𝝎𝒏 t + Φ)}+ [k]{X sin(𝝎𝒏 t + Φ)}={0}

Rearranging,
[[k]- 𝝎𝟐𝒏[M]]{X} = 0…………………… (3.3)
The problem of determining the constant (-𝝎𝟐𝒏 ) is an important mathematical problem known as an
characteristics value or eigen value problem. Its non-trivial solution is possible only when the
determinant of the coefficient matrix vanishes, that is
|[k]- 𝝎𝟐𝒏 [M]|=0………………………(3.4)
The expansion of the determinant in Eq.(3.4) results in a algebraic equation of nth order in which
𝝎𝟐𝒏 should be satisfied for n values of 𝝎𝟐𝒏 . This equation is known as characteristics are nothing but
natural frequencies 𝝎𝒊 of the MDOF system. The resulting simple harmonic motions can take place
in such a way that all the masses move in phase at the same frequency. Foe each natural frequency,
the resulting deflected shape is known as normal mode shape or Eigen vector. The mode
corresponding to the lowest frequency is called first mode or fundamental mode. The other modes
are called higher modes. The normal modes depends on mass and stiffness properties of the system.
The normal modes are completely determined by assuming unit values for the amplitude of motion at
the first degree of freedom. So that the normal modes of the remaining coordinate are computed
relatively. This process of normalizing each modes is called normalisation.

3.4 ORTHOGONALITY AND NORMALITY PRINCIPLES

The mode shapes or Eigen vectors are mutually orthogonal with respect to the mass and stiffness
matrices. Orthogonality is the important property of the normal modes or Eigen vectors and it is used
to uncouple the modal mass and stiffness matrices.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

As we know that, 𝝎𝟐 [M] {Φ} = [k] {Φ}

For 𝑖 𝑡ℎ mode (or) Eigen value, 44
𝝎𝟐𝒊 [M]{Φ}𝒊 = [k] {Φ}𝒊 ……………………………………..(3.5)
𝑡ℎ
For 𝑗 mode,
𝜔𝑗2 [M]{Φ}𝒋 = [k] {Φ}𝒋 ………………………………………(3.6)
Multiplying Eq. (3.5) by {Φ}𝑻𝒋 and Eq. (8.6) by {Φ}𝑻𝒊 to get

ω2i {Φ }𝑻𝒋 [M] {Φ}𝒊 = {Φ}𝑻𝒋 [k] {Φ}𝒊 ………………………….(3.7)

And ω2j {Φ}𝑻𝒊 [M]{Φ}j ={Φ}𝑻𝒊 [k]{Φ}j …………………………..(3.8)
By transposing LHS and RHS of Eq. (3.8), we get

ω2j {Φ}𝑻𝒋 [M] {Φ}𝒊 = {Φ}𝑻𝒋 [k] {Φ}𝒊 ………………… (3.9)

In which [M ]𝑻 = [M]
And [k]𝑻 = [k]
Subtract Eq. (3.7) from Eq. (3.9), we get
(ω2i - ω2j){Φ}𝑻𝒋 [M] {Φ}𝒊 = 0
For ωi ≠ ωj , we get .
{Φ}𝑻𝒊 [M] {Φ}𝒊 = 0 I ≠ j …………………….. (3.10)
Similarly,
{Φ}𝑻𝒊 [k] {Φ}j = 0
This condition is called the orthogonality principle.
Eigen vectors are unique; a scalar multiplier of an Eigen vector is also an Eigen vector.
For 𝑖 𝑡ℎ Eigen value,
{Φ}𝑻𝒊 [M]{Φ}𝒊 = scalar ………………………………………(3.11)

If the scalar is equal to unity, the resulting Eigen vectors are called normal modes, that is

{Φ}𝑻𝒊 [M] {Φ}𝒊 = I …………………………………(3.12)

Where I is the identity matrix which is a diagonal matrix with unit values along the main diagonal.
From Eq. (3.12), the natural modes are not only orthogonal but are normalized with respect to mass
matrix. These are also called mass orthonormal set.
For 𝑖 𝑡ℎ mode
𝜔𝑖2 [M]{Φ}𝒊 = [k] {Φ}𝒊
Multiplying by {Φ}𝑻 , 𝜔𝑖2 {Φ}𝑻𝒊 [M] {Φ}𝒊 = {Φ}𝑻𝒊 [k] {Φ}𝒊
Since, {Φ }𝑻𝒊 [M] {Φ}𝒊 = 1, ω2i (1) = {Φ}𝑻𝒊 [k] {Φ}𝒊

{Φ}𝑻𝒊 [k] {Φ}𝒊 = 𝝎𝟐𝒊 ……………………………………(3.13)

Equations (3.12) and (3.13) are called normality principle.

DAMPED SYSTEMS
The response of the damped MDOF system subjected to free vibration is governed by
[M]{ẍ}+ [c] {ẋ} + [k]{x}=0…………………………………(3.14)
In which [c] is damping matrix and {ẋ} is a velocity vector.
Generally small amount of damping is always present in real structure and it does not have much
influence on the determination of natural frequencies and mode shapes of the system.
Therefore, the natural frequencies and mode shapes for the damped systems are calculated by using
the sa.me procedure adopted for un-damped system.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

In Eq. (3.14) the damping matrix is coupled with mass and stiffness matrices. Thus orthogonality
principle is also applicable to damping matrix, i.e. 45
{Φ}𝑻𝒊 [c]{Φ}j = 0…………………........... (3.15)

Figure 3.2Mass-proportional damping and stiffness-proportional damping

Consider mass-proportional damping and stiffness-proportional damping as shown in figure 3.2.

Thus the damping matrix is of the form
[c] = α [M] + β[k]……………………………………………………(3.16)
In which α and β are arbitrary proportionality factors having units s-1 and s respectively and depend
on damping ratio and natural frequency. The factor α is selected to obtain a specified value of
damping ratio in any one mode say 𝜌𝑖 for 𝑖 𝑡ℎ mode.
As we know that ρ = 𝑪/𝑪𝒄 = 2m 𝝎𝒏
𝑪
Cc = 𝝆 = 2mωn
Cc = c = 2ρi m 𝝎𝒏

For 𝑖 𝑡ℎ mode, Cc = c = 2ρimωi = α [m]

α = 2ρiωi
Similarly factor β is selected to obtain a specified value of the damping ratio in any one mode say 𝑗 𝑡ℎ
mode.
As we know that, k = mω2n
c= β[k] = βmω2n
2ρωn m = βmω2n
𝟐𝝆
β = 𝝎𝒏
𝟐𝝆𝒋
For 𝑗 𝑡ℎ mode β = 𝝎𝒋

Multiplying by {Φ}, Eq. (8.16) we get

[c] {Φ} = α [M] {Φ} + β [k] {Φ}
𝑻
Pre-multiplied by {Φ} we get

{Φ}T [c] {Φ} = α {Φ}𝑻 [M]{Φ}+ β{Φ}T[k]{Φ}

= α [I] +βω2i
𝟐𝝆𝒋
= 2ρiωi [I] + 𝝎 𝝎𝟐𝒋
𝒋
𝑻
Or {Φ} [c]{Φ} = 2ρiωi[I] + 2ρjωj

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

3.5 DECOUPLING OF EQUATIONS/CONCEPT OF MODAL

46
SUPERPOSITION METHOD
The free vibration response of an MDOF system consists of coupled differential equations with
respect to mass, stiffness and damping. Solution of the coupled equations is a complicated problem
because it has dependent variables. To simplify the response of MDOF system, the coupled
differential equations must be transformed to independent or uncoupled equations which contain
only one dependent variable. This process of coupling the coupled differential equations is called
decoupling of equations. Thus the modal superposition method is used to uncouple the coupled
differential equations by means of transportation of coordinates which incorporates the
Orthogonality principles of the mode shapes. While decoupling, the response of a MDOF system is
reduced to the SDOF system. Total response may be obtained as the superposition of the solution of
SDOF system, i.e. independent modal equations.

Figure3.3

The free vibration response of MDOF system is given by [M] {ẍ}+[c]{ẋ}+ [k]{x}=0
In which {x} is the displacement vector which is also called geometric coordinates. In the
transformation of a coupled system of equations into a system of uncoupled equations, the normal
modes or normal coordinates are multiplied by some factors which are called generalized coordinates
or the natural coordinates.
Thus normal modes {Φ} helps to transform generalized coordinates {z} to geometric coordinates,
that is for the system shown in figure 3.3.

(1) (2) (3)

𝑥1 𝜙1 𝜙1 𝜙1
[𝑥] = {𝑥2 } = 𝑧1 {𝜙2 } + 𝑧2 {𝜙2 }+𝑧3 {𝜙2(3) }
(1) (2)

𝑥3 (1) (2) (3)

𝜙3 𝜙3 𝜙3
First mode- Second mode-Third mode
In general, {x} = 𝒛𝟏 {𝜱 } + 𝒛𝟐 {𝜱(𝟐) }+𝒛𝟑 {𝜱(𝟑) }
(𝟏)

Multiplying with [M], we get

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

[M]{x}=𝒛𝟏 [𝑴]{𝜱(𝟏) }+𝒛𝟐 [𝑴]{𝜱(𝟐) }+ 𝒛𝟑 [𝑴]{𝜱(𝟑) } 47

Pre multiply by {𝜱(𝟏) }𝑻

{𝜱(𝟏) } 𝑻 [M]{x}=𝒛𝟏 {𝜱(𝟏) }𝑻 [𝑴]{𝜱(𝟏) }+𝒛𝟐 {𝜱(𝟏) }𝑻 [𝑴]{𝜱(𝟐) }+ 𝒛𝟑 {𝜱(𝟏) }𝑻 [𝑴]{𝜱(𝟑) }

= 𝒛𝟏 {𝜱(𝟏) }𝑻 [𝑴]{𝜱(𝟏) }+0+0

{𝜱(𝟏) } 𝑻 [M]{x} = 𝒛𝟏 (1)

In general, for n number of Eigen values

𝒛𝒌 = {𝜱(𝒌) } 𝑻 [M] {x}

3.6 FORCED VIBRATION

Response of the forced vibration of an MDOF system may also be expressed in terms of the normal
modes of vibration. Modal superposition method is used to determine the total response of the
system. The equation of motion of an MDOF system is given by

[M]{ẍ}+ [c]{ẋ}+ [k]{x}= F(t) ……………………………(3.17)

Coordinate transformation is given by
{x}{t} = {Φ}z(t)
In which x(t) is geometric coordinate
z(t) is generalized coordinate
Substituting {x} (t) into Eq. (3.17), we get
[M]{Φ}{𝒛̈ } + [c] {Φ}{ż}+[k]{Φ}{z} = F(t)
Pre multiply by {Φ}T to get

{Φ}T[M]{Φ}{𝒛̈ }+{Φ}T [c]{Φ}{ż} +{Φ}T [k]{Φ}{z}= F(t)………………(3.18)

The Eq. (3.18) is solved by using orthogonality relations
{Φ}T[M]{Φ}= 1
{Φ}T[k] {Φ}= Diag(𝝎𝟐𝒊 )
𝟐𝝆𝒋
{Φ}T [c] {Φ} = 2ρiωi[I] + 𝝎𝟐𝒊
𝝎𝒋

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

CHAPTER 4 48
EARTHQUAKE RESPONSE OF LINEAR SYSTEMS
One of the most important applications of theory of structural dynamics is in analyzing the response
of structures to ground shaking caused by an earthquake. This chapter deals with linear systems,
which are elastic systems, so we will refer to them by linearly elastic systems.

4.1 EARTHQUAKE EXCITATION

For engineering, the time variation of ground acceleration is the most useful way of defining the
shaking of the ground during an earthquake. Ground acceleration appears on the right hand side of
equation of motion when a SDF system is subjected to ground excitation.

A basic instrument, called strong-motion accelerograph, records three components of motions

including two horizontal components and one vertical component. Acceleration is vector quantity in
3-D space that varies with time. It requires three components to define a vector in 3-D space.

Instrument does not record continuously all the time. It is triggered to start recording when the first
waves of earthquake arrive because there may not be any strong ground motion for months.

Basic element of an accelerograph is a transducer element, which is an SDF mass-spring-damper

system. It is characterized by its natural frequency fn and viscous damping ratio ρ; typically fn = 25
Hz and ρ = 60% for modern analog accelerograph and fn = 50 Hz and ρ = 70% for modern digital
accelerograph. These transducer parameters allow the instruments to record ground acceleration
containing frequency up to 60% of fn without excessive distortion.

Before 1990s, strong recorded ground motions were very rare because numbers of instruments were
limited and they were not located near the origin of earthquakes. After 1994 Northridge, California;
1995 Kobe, Japan; 1999 Turkey and Taiwan earthquakes, hundreds of strong motion records became
available.

Earthquake ground motions are irregular in nature and they vary widely in terms of amplitude,
duration, frequency content, and wave form. The peak acceleration can be more than 1g and duration
of the strong phase may be as short as a few seconds or as long as a few minutes.

The factors that affect characteristics of ground motion at a location are:

4.1 Source (magnitude, fault mechanism)

4.2 Path (distance from epicenter, geology, direction)
4.3 Site (soil condition at the location considered)

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

49

Figure 4.1 Ground motion recorded during several earthquakes. (based in part on Hudson
(1979).)

4.2 EQUATION OF MOTION

For a linear SDF system subjected to ground motion excitation 𝑢̈ 𝑔 (𝑡), the motion of the mass is
governed by the equation
𝑚𝑢̈ + 𝑐𝑢̇ + 𝑘𝑢 = −𝑚𝑢̈ 𝑔 (𝑡)
Divide by m on both sides
𝑢̈ + 2𝜌𝜔𝑛 𝑢̇ + 𝜔𝑛2 𝑢 = −𝑢̈ 𝑔 (𝑡)
For a given ground acceleration 𝑢̈ 𝑔 (𝑡), the response u(t) depends on the natural frequency ωn (or
period Tn) and damping ratio ρ of the system. Thus, two systems with the same Tn and ρ will have the
same response u(t) even if one of them has more mass and is stiffer than the other.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

50

Figure 4.2 Single degree of freedom systems

4.3 RESPONSE HISTORY

For a given ground motion 𝑢̈ 𝑔 (𝑡), the deformation response of an SDOF system depends on the
natural frequency and damping ratio. The response u(t) of an SDOF system can be determined from
numerical procedure.
Observe the responses u(t) of three SDOF systems with the same damping ratio 2%, but
different natural period of vibration. The time required for an SDF system to complete a cycle of
vibration when subjected to earthquake ground motion is very close to its natural period of vibration.
For these cases, the one with longest natural period has the largest peak deformation. This
trend is not necessarily true over the entire range of periods. Compare another three SDF systems, all
with the same natural period of vibration of 2 sec but different damping ratio of 0, 2, and 5%. The
time required for an SDF system to complete a cycle of vibration is similar for all three cases, but the
amplitude always decreases as damping ratio increases. This trend is consistent with the study of
response to harmonic and pulse excitations.

Figure 4.3 Deformation response of SDOF systems to El Centro ground motion

Once the deformation response history has been evaluated by dynamic analysis of structure, the
internal forces can be determined by static analysis of structure at each time instant. By the concept
of equivalent static force fs, which is related to the earthquake force specified in building code,

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

𝑓𝑠 = 𝑘𝑢(𝑡)
where k is the lateral stiffness of the frame. 𝑘 = 𝜔𝑛2 𝑚 51
2
𝑓𝑠 = 𝑚𝜔𝑛 𝑢(𝑡) = 𝑚𝐴(𝑡)
Where,
𝐴(𝑡) = 𝜔𝑛2 𝑢(𝑡)
A(t) is called pseudo-acceleration. The equivalent static force equal mass time pseudo-
acceleration, not the total acceleration 𝑢̈ 𝑡 (𝑡). A(t) is obtained by multiplying u(t) by 𝜔𝑛2 =
(2𝜋/𝑇𝑛 )2.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

4.4 RESPONSE SPECTRUM CONCEPT

52
First introduced by M. A. Biot (1932), the response spectrum method is a central concept in
earthquake engineering. It provides a convenient mean to summarize the peak response of all
possible SDOF systems to a particular component of ground motion.
A plot of peak value of a response quantity as a function the natural vibration period Tn or frequency
ωn or cyclic frequency fn is called “response spectrum”. Each plot is for a fixed damping ratio ρ.
Engineers prefer to use Tn rather than ωn because Tn is more familiar.
A variety of response quantity can be defined.
Deformation response spectrum 𝑢0 (𝑇𝑛 , 𝜌) ≡ max|𝑢(𝑡. 𝑇𝑛 , 𝜌)|
𝑡
Relative velocity response spectrum 𝑢̇ 0 (𝑇𝑛 , 𝜌) ≡ max|𝑢̇ (𝑡. 𝑇𝑛 , 𝜌)|
𝑡
Acceleration response spectrum 𝑢̈ 𝑡0 (𝑇𝑛 , 𝜌) ≡ max|𝑢̈ 𝑡 (𝑡. 𝑇𝑛 , 𝜌)|
𝑡

4.5 DEFORMATION RESPONSE SPECTRA

The peak value of deformation time history response of an SDOF system with natural period Tn due
to a particular ground excitation can be plotted as a point on the deformation response spectra. The
peak value of deformation is denoted by
𝐷 = 𝑢0 = max|𝑢(𝑡)|
𝑡
If many of such analyses are repeated for many SDOF systems with a fixed damping ratio ρ but
different natural periods Tn, the deformation response spectra can be constructed for the range of Tn
considered. Similar spectrum for other value of damping ratio can be constructed in a similar
manner.

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

Figure 4.4 (a) ground acceleration; (b) deformation response of three SDOF systems with ρ =
2% and Tn = 0.5, 1 and 2 sec; (c) deformation response spectrum for ρ=2% 53
4.6 PSEUDO-VELOCITY RESPONSE SPECTRUM
Consider a quantity V for an SDF system with natural frequency ωn which has the peak deformation
D
2𝜋
𝑉 = 𝜔𝑛 𝐷 = 𝐷
𝑇𝑛
The quantity V has units of velocity. It is related to the peak value of strain energy Eso by
𝑚𝑉 2
𝐸𝑠𝑜 =
2
derived from
𝑘𝑢02 𝑘𝐷2 𝑘(𝑉 ⁄𝜔𝑛 )2 𝑚𝑉 2
𝐸𝑠𝑜 = = = =
2 2 2 2
V is called the peak relative pseudo-velocity or peak pseudo-velocity. It is not the same as peak
relative velocity uo. Pseudo-velocity response spectrum is a plot of V as a function of the natural
period Tn or natural frequency ωn of the system.

4.7 PSEUDO-ACCELERATION RESPONSE SPECTRUM

2𝜋 2
𝐴 = 𝜔𝑛2 𝐷 = ( ) 𝐷
𝑇𝑛
The quantity A has units of acceleration and is related to the peak value of base shear Vbo as
𝐴
𝑉𝑏𝑜 = 𝑓𝑠𝑜 = 𝑚𝐴 = 𝑤
𝑔
where w is the weight of the structure and g is the gravitational acceleration.
A/g may be interpreted as the base shear coefficient or lateral force coefficient. It is used in
building codes to represent the coefficient that is multiplied to the weight to obtain base shear force.
The pseudo-acceleration A is different from the total acceleration 𝑢̈ 𝑡 so it is called “pseudo.”

Figure 4.5 Response spectra (ρ = 0.02) for El Centro ground motion: (a) deformation response
spectrum; (b) pseudo-velocity response spectrum; (c) pseudo-acceleration response spectrum

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

4.8 PEAK STRUCTURAL RESPONSE FROM THE RESPONSE 54

SPECTRUM
If the response spectrum is available for a given ground motion, the peak value of deformation and
internal forces can be readily determined.
𝑇𝑛 𝑇𝑛 2
𝑢0 = 𝐷 = 𝑉=( ) 𝐴
2𝜋 2𝜋
𝑓𝑠𝑜 = 𝑘𝐷 = 𝑚𝐴

Figure 4.6 Peak value of equivalent static force

No further dynamic is required. Only static analysis of structure subjected to the equivalent static
force provides the peak values of internal forces during response to the given earthquake.
𝑉𝑏𝑜 = 𝑘𝐷 = 𝑚𝐴
𝑀𝑏𝑜 = ℎ𝑉𝑏𝑜

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

CHAPTER 5 55
DETERMINISTIC EARTHQUAKE RESPONSE
5.1 LUMPED SDOF ELASTIC SYSTEMS, TRANSLATIONAL
EXCITATION
The simplest form of earthquake response problem involves a SDOF lumped mass system subjected
to identical single-component translations of all support points. An example of such a system is
shown in Fig.5.1 to define earthquake response spectra. It is also used for formation of equation of
motion for system subjected to base translation,

Figure 5.1 Lumped SDOF system to rigid-base translation.

m 𝑣 𝑡̈ (t) + c𝑣̇ (t) + k v(t) = 0 ……….(5.1)

The superscript t in the first term of this equation denotes total displacement, the equation of motion
in terms of an effective loading as follows

m 𝑣 𝑡̈ (t) + c𝑣̇ (t) + k v(t) = peff(t) ……………(5.2)

where
peff(t) = -m𝑣̈ g(t)………………….. …………(5.3)
in this expression 𝑣̈ g(t) represents the free-field input acceleration applied at the base of the structure;
the negative sign has little significance in earthquake response analysis and generally is ignored.
It also is possible to express the effective earthquake force in terms of the free-field velocity and
displacement, 𝑣̇ g(t) and vg(t), if the equation of motion is formulated in terms of total rather than
relative motion. However, this form of the equation of motion seldom is used because it is so much
simpler to evaluate the base acceleration expression for effective loading, Eq. (5.3)

Asst. Prof. Amrendra Swarnakar

 STRUCTURAL DYNAMICS

For purposes of this discussion, it is convenient to express the earthquake displacement response
given by the solution of Eq. (5-2) in terms of the Duhamel integral expression for low damped 56
systems, as follows:
1 𝑡 1
v(t) = 𝜔 ∫0 𝑣̈ g(t) exp [-𝜌𝜔( t-𝜏 )d𝜏 = 𝜔V(t)…................ (5.4)

5.2 LUMPED MDOF ELASTIC SYSTEMS, TRANSLATIONAL

EXCITATION
The formulation of the earthquake response analysis of a lumped MDOF system can be carried out in
matrix notation in a manner entirely analogous to the foregoing development of the lumped SDOF
equations. Thus the equations of motion of the multistory shear building shown in Fig. can be written
by analogy with Eq (5-1)
as
m𝑣̈ 𝑡 (t) + c 𝑣̇ (t) + kv(t) = 0…………………..(5.5)

Figure 5.2 Discretized MDOF system with rigid-base translation

and again the effective earthquake force can be derived by expressing the total displacements as the
sum of the relative motions plus the displacements resulting directly from the support motions. For
the system of fig- this relationship may be written
vt (t)= v(t) +{1}vg(t)..………………… (5.6)
in which {1} represents a column of ones. This vector expresses the fact that a unit static translation
of the base of this structure produces directly a unit displacement of all degrees of freedom. Of
course, this simple relationship is a consequence of the type of support displacement which has been
applied as well as of the structural configuration; for other forms of structures or support motions this
Asst. Prof. Amrendra Swarnakar
 STRUCTURAL DYNAMICS

static-displacement vector would be different. Thus, the structure shown in Fig. should be
considered as a special case, even though a large number of practical analyses are assumed to be of 57
this type.
Substituting Eq.(5.6) into (5.6) leads to the relative-response equations of motion
m𝑣̈ 𝑡 (t) + c 𝑣̇ (t) + kv(t) = peff(t) ...…….(5.7)
in which
peff(t) =-m{1}vg(t)………….......………(5.8)
Equations (5.7) could be solved directly through the frequency domain or by numerical integration of
the coupled equations in the time domain; however, in analyzing the earthquake response of linear
structures, it generally is much more efficient to transform to a system of normal (modal) coordinates
because the support motions tend to excite strongly only the lowest modes of vibration. Thus good
approximations of the earthquake response of systems having dozens or even hundreds of degrees of
freedom can often be obtained by carrying out the analysis for only a few normal coordinates.
If it is assumed that the damping matrix is of a form which satisfies the same orthogonality
conditions as the mass and stiffness matrices, the result is a set of N uncoupled modal equations of
the form

Mn 𝑌̈n + Cn𝑌̇n + KnYn =Pn(t)………..(5.9)

By analogy with the derivation of the generalized SDOF response, it may be seen that the response
of each mode of the MDOF system is given by
𝐿𝑛
Yn(t)=𝑀 Vn(t)…..................…..(5.10)
𝑛 𝜔𝑛

Finally, the relative-displacement vector due to all modal responses is obtained by superposition, that
is,
𝐿𝑛
v(t)=ΦY(t)= Φ{𝑀 Vn(t)} ……...…(5.11)
𝑛 𝜔𝑛

in which Φ is made up of all mode shapes for which the modal response is excited significantly by
the earthquake, and the term in braces represents a vector of response terms defined for each mode
considered in the analysis.

Asst. Prof. Amrendra Swarnakar