CHAPTER

2
Theory of Vibrations

2.1 Objective and Overview

In order to understand the behaviour of a structure subjected to dynamic load lucidly, one must study
the mechanics of vibrations caused by the dynamic load. The pattern of variation of a dynamic load
with respect to time may be either periodic or transient. The periodical motions can be resolved into
sinusoidally varying components e.g. vibrations in the case of reciprocating machine foundations.
Transient vibrations may have very complicated non-periodic time history e.g. vibrations due to
earthquakes and quarry blasts.
A structure subjected to a dynamic load (periodic or transient) may vibrate in one of the
following four ways of deformation or a combination there-of:
(i) Extensional (Fig. 2.1a)
(ii) Shearing (Fig. 2.1b)
(iii) Bending (Fig. 2.1c)
(iv) Torsional (Fig. 2.1d)

Fig. 2.1. Different types of vibrations

The forms of vibration mainly depend on the mass, stiffness distribution and end conditions of
the system.
To study the response of a vibratory system, in many cases it is satisfactory to reduce it to an
idealized system of lumped parameters. In this regard, the simplest model consists of mass, spring
and dashpot. This chapter is framed to provide the basic concepts and dynamic analysis of such
systems. Actual field problems which can be idealized to mass-spring-dashpot systems, have also
been included.
16 Dynamics of Soils and Their Engineering Applications

2.2 Definitions
2.2.1 Vibrations
If the motion of the body is oscillatory in character, it is called vibration.

2.2.2 Degrees of Freedom

The number of independent co-ordinates which are required to define the position of a system during
vibration, is called degrees of freedom (Fig. 2.2).

Fig. 2.2. Systems with different degrees of freedom

2.2.3 Periodic Motion

If motion repeats itself at regular intervals of time, it is called periodic motion.

2.2.4 Free Vibration

If a system vibrates without an external force, then it is said to undergo free vibrations. Such
vibrations can be caused by setting the system in motion initially and allowing it to move freely
afterwards.

2.2.5 Natural Frequency

This is the property of the system and corresponds to the number of free oscillations made by the
system in unit time.
Theory of Vibrations 17

2.2.6 Forced Vibrations

Vibrations that are developed by externally applied exciting forces are called forced vibrations.
These vibrations occur at the frequency of the externally applied exciting force.

2.2.7 Forcing Frequency

This refers to the periodicity of the external forces which act on the system during forced vibrations.
This is also termed as operating frequency.

2.2.8 Frequency Ratio

The ratio of the forcing frequency and natural frequency of the system is referred as frequency ratio.

2.2.9 Amplitude of Motion

The maximum displacement of a vibrating body from the mean position is amplitude of motion.

2.2.10 Time Period

Time taken to complete one cycle of vibration is known as time period.

2.2.11 Resonance
A system having n degrees of freedom has n natural frequencies. If the frequency of excitation
coincides with any one of the natural frequencies of the system, the condition of resonance occurs.
The amplitudes of motion are very excessive at resonance.

2.2.12 Damping
All vibration systems offer resistance to motion due to their own inherent properties. This resistance
is called damping force and it depends on the condition of vibration, material and type of the
system. If the force of damping is constant, it is termed Coulomb damping. If the damping force is
proportional to the velocity, it is termed viscous damping. If the damping in a system is free from
its material property and is contributed by the geometry of the system, it is called geometrical or
radiation damping.

2.2.13 Principal Modes of Vibration

In a principal mode, each point in the system vibrates with the same frequency. A system with n
degrees of freedom possesses n principal modes with n natural frequencies.

2.2.14 Normal Mode of Vibration

If the amplitude of a point of the system vibrating in one of the principal modes is made equal to
unity, the motion is then called the normal mode of vibration.

2.3 Harmonic Motion

Harmonic motion is the simplest form of vibratory motion. It may be described mathematically by
the following equation:
Z = A sin(wt – q) (2.1)
18 Dynamics of Soils and Their Engineering Applications

Equation (2.1) is plotted as function of time in Fig. 2.3. The various terms of this equation are
as follows:
Z = Displacement of the rotating mass at any time t
A = Displacement amplitude from the mean position, sometimes referred as single amplitude. The
distance 2A represents the peak-to-peak displacement amplitude, sometimes referred to as
double amplitude, and is the quantity most often measured from vibration records.
w = Circular frequency in radians per unit time. Because the motion repeats itself after 2p radians,
the frequency of oscillation in terms of cycles per unit time will be w/2p. It is denoted by f.
q = Phase angle. It is required to specify the time relationship between two quantities having the
same frequency when their peak values having like sign do not occur simultaneously. In Eq.
(2.1) the phase angle is a reference to the time origin.

Fig. 2.3. Quantities describing harmonic motion

More commonly, the phase angle is used as a reference to another quantity having the same
frequency. For example, at some reference point in a harmonically vibrating system, the motion may
be expressed by
Z1 = A1 sin wt (2.2)
Motion at any other point in the system might be expressed as
Zi = Ai sin (wt – qi) (2.3)
with p ≥ q ≥ – p.
For positive values of q the motion at point i reaches its peak within one half cycle after the
peak motion occurs at point 1. The angle q is then called phase lag. For negative values of q the
peak motion at i occurs within one half cycle ahead of motion at 1, and q is called as phase lead.
The time period, T is given by
1 2π
T= = (2.4)
f ω
The velocity and acceleration of motion are obtained from the derivatives of Eq. (2.1).
dZ
Velocity = = Z = ωA cos (ωt − θ) (2.5)
dt
π
= ωA sin(ωt − θ + )
2
d 2 Z 
Acceleration = = Z = − ω 2 A cos (ωt − θ) (2.6)
dt 2
  = ω2 A sin (ωt – q + π)
Theory of Vibrations 19

Equations (2.5) and (2.6) show that both velocity and acceleration are also harmonic and can be
represented by vectors wA and w2A, which rotate at the same speed as A, i.e. w rad/unit time. These,
however, lead the displacement and acceleration vectors by p/2 and p respectively. In Fig. 2.4 vector
representation of harmonic displacement, velocity and acceleration is presented considering the
displacement as the reference quantity (q = 0).
When two harmonic motions having little different frequencies are superimposed, a non-
harmonic motion as shown in Fig. 2.5 occurs. It appears to be harmonic except for a gradual increase
and decrease in amplitude. The displacement of such a vibration is given by:
Z = A1 sin (w1t – q1) + A2 sin (w2t – q2) (2.7)

Fig. 2.4. Vector representation of harmonic displacement, velocity and acceleration

Fig. 2.5. Motion containing a beat

The dashed curve (Fig. 2.5) representing the envelop of the vibration amplitudes oscillates
at a frequency, called the beat frequency, which corresponds to the difference in the two source
frequencies:
20 Dynamics of Soils and Their Engineering Applications

1 | ω1 − ω 2 |
fb =
= (2.8)
Tb 2π
The frequency of the combined oscillations is the average of the frequencies of the two
components and is given by
1  1   ω + ω2 
f = =  1
 (2.9)
T  2   2π 
The maximum and minimum amplitudes of motion are the sum and difference of the amplitudes
of the two sources respectively.
Zmax = A1 + A2
(2.10a)

Zmin = |A1 – A2|

(2.10b)
If the drive systems of two machines designed to operate at the same speed are not synchronized,
they may result vibrations having the beat frequency.

2.4 Vibrations of a Single Degree Freedom System

The simplest model to represent a single degree of freedom system consisting of a rigid mass m
supported by a spring and dashpot is shown in Fig. 2.6a. The motion of the mass m is specified by
one co-ordinate Z. Damping in this system is represented by the dashpot, and the resulting damping
force is proportional to the velocity. The system is subject to an external time dependent force F(t).
Figure 2.6b shows the free body diagram of the mass m at any instant during the course of
vibrations. The forces acting on the mass m are:

Fig. 2.6. Single degree freedom system

(i) Exciting force, F (t): It is the externally applied force that causes the motion of the system.
(ii) Restoring force, Fr: It is the force exerted by the spring on the mass and tends to restore the
mass to its original position. For a linear system, restoring force is equal to K · Z, where K is
the spring constant and indicates the stiffness. This force always acts towards the equilibrium
position of the system.
(iii) Damping force, Fd: The damping force is considered directly proportional to the velocity and
given by C · Ż where C is called the coefficient of viscous damping; this force always opposes
Theory of Vibrations 21

the motion. In some problems in which the damping is not viscous, the concept of viscous
damping is still used by defining an equivalent viscous damping which is obtained so that the
total energy dissipated per cycle is same as for the actual damping during a steady state of
motion.
 According to De-
(iv) Inertia force, Fi: It is due to the acceleration of the mass and is given by m Z.
Alembert’s principle, a body which is not in static equilibrium by virtue of some acceleration
which it possess, can be brought to static equilibrium by introducing on it an inertia force.
This force acts through the centre of gravity of the body in the direction opposite to that of
acceleration.
The equilibrium of mass m gives
mZ + CZ + KZ = F (t ) (2.11)
which is the equation of motion of the system.

2.4.1 Undamped Free Vibrations

For undamped free vibrations, the damping force and the exciting force is equal to zero. Therefore
the equation of motion of the system becomes
mZ + KZ = 0 (2.12a)
 K
or Z +   Z = 0 (2.12b)
 m
The solution of this equation can be obtained by substituting
Z = A1 cos wnt + A2 sin wnt (2.13)

where A1 and A2 are both constants and wn is undamped natural frequency.

Substituting Eq. (2.13) in Eq. (2.12b), we get,
 K
−ω 2n ( A1 cos ω n t + A2 sin ω n t ) +   ( A1 cos ω n t + A2 sin ω n t ) = 0
 m

K
or ωn = ± (2.14)
m
The values of constants A1 and A2 are obtained by substituting proper boundary conditions. We
may have the following two boundary conditions:
(i) At time t = 0, displacement Z = Z0, and
(ii) At time t = 0, velocity Ż = V0
Substituting the first boundary condition in Eq. (2.13)
A1 = Z0 (2.15)
   Now Z = − A1 ω n sin ω n t + A2 ω n cos ω n t (2.16)
Substituting the second boundary condition in Eq. (2.16)
V0
A2 = (2.17)
ωn
V0
   Hence Z = Z 0 cos ω n t + sin ω n t (2.18)
ωn
   Now let Z0 = AZ cosq (2.19)
22 Dynamics of Soils and Their Engineering Applications

V0
and = AZ sin θ (2.20)
ωn
Substitution of Eqs. (2.19) and (2.20) into Eq. (2.18) yields
Z = AZ cos (wnt – q)
(2.21)

 V 
θ = tan −1  0 
where (2.22)
 ωn Z0 
2
V 
Az = Z 02 +  0  (2.23)
 ωn 
The displacement of mass given by Eq. (2.21) can be represented graphically as shown in Fig. 2.7.
   It may be noted that

At time t equal to Displacement Z is

0 AZ cosq
θ AZ
ωn

π
+θ
2 0
ωn

π+θ
–AZ
ω
3
π+θ
2 0
ωn

2π + θ
AZ
ωn

It is evident from Fig. 2.7 that nature of foundation displacement is sinusoidal. The magnitude
of maximum displacement is AZ. The time required for the motion to repeat itself is the period of
vibration, T and is therefore given by
2π
T=
(2.24)
ωn
The natural frequency of oscillation. fn is given by
1 ω 1 K
fn = = n =
(2.25)
T 2π 2π m
mg W
   Now = = δ st (2.26)
K K
where g = Acceleration due to gravity, 9.81 m/s2
W = Weight of mass m
δst = Static deflection of the spring
Theory of Vibrations 23

1 g
   Therefore fn = (2.27)
2π δ st
Equation (2.27) shows that the natural frequency is a function of static deflection. The relation
of fn and dst given by Eq. (2.27) gives a curve as shown in Fig. 2.8.
The nature of variation of the velocity and acceleration of the mass is also shown in Fig. 2.7.

Fig. 2.7. Plot of displacement, velocity and acceleration for the free vibration of a mass-spring system

Fig. 2.8. Relationship between natural frequency and static deflection

2.4.2 Free Vibration with Viscous Damping

For damped free vibration system (i.e., the excitation force F0 sin w t on the system is zero), the
differential equation of motion can be written as
mZ + CZ + KZ = 0
(2.28)
24 Dynamics of Soils and Their Engineering Applications

where C is the damping constant or force per unit velocity. The solution of Eq. (2.28) may be written
as
Z = Aeλ t
(2.29)
where A and l are arbitrary constants. By substituting the value of Z given by Eq. (2.29) in Eq.
(2.28), we get
m A l2 el t + C A lel t + K A el t = 0

 C K
or λ2 +   λ + = 0 (2.30)
 m m
By solving Eq. (2.30)
2
C  C  K
λ1, 2 = −
±   − (2.31)
2m  2m m
The complete solution of Eq. (2.28) is given by
Z = A1e λ1t + A2 e λ 2t
(2.32)
The physical significance of this solution depends upon the relative magnitudes of (C/2m)2 and
(K/m), which determines whether the exponents are real or complex quantities.
To proceed further, it is convenient to define here a new term critical damping coefficient (Cc).
It is that value of the damping coefficient (C) that makes the square root term of Eq. (2.31) equal to
zero. Therefore, when
2
 C  K

  = ; C = Cc (2.32a)
2m  m
   Then Cc = 2 Km (2.32b)
The ratio of the actual damping constant C to the critical damping constant Cc is defined as
damping ratio.
C
ξ= (2.33a)
Cc
C C Cc C ⋅ 2 Km C K
= . = = . = ωnξ (2.33b)
2m Cc 2m Cc ⋅ 2m Cc m

λ1, 2 = (−ξ ± ξ 2 − 1)ω n (2.34)

   Therefore, Eq. 2.32 may be written as:

Z = e − ξωn t  A1e ξ −1ωn t + A2 e ξ −1 ωn t 

2 2

  (2.35)
The physical significance of this solution depends upon the relative magnitude of x with respect
to 1, which determines whether the exponents are real or complex quantities.
Case 1: x > 1 (Over damped system)
In this system, the damping is comparatively large and the exponents of Eq. (2.35) are real. The
values of arbitrary constants A1 and A2 may be obtained consideing that the body (Fig. 2.6) is
displaced by distance Z0 from equilibrium position and released without any initial velocity. Then,
boundary conditions become:
(i) At t = 0; Z = Z0 (2.36a)
Theory of Vibrations 25

(ii)             At t = 0; Ż = 0 (2.36b)

dZ
Z = = e − ξωn t × ( − ξω n )  A1e ξ2 − 1 ω n t
+ A2 e ξ2 − 1 ω n t 
dt  

+ e − ξωn t  A1ω n ξ 2 − 1 e ξ −1 ωn t − A2 ω n ξ 2 − 1 e − 
2
ξ2 − 1 ω n t

  (2.37)
Apply boundary conditions in Eqs. (2.36) and (2.37),
A1 + A2 = Z0 (2.38a)

− ( A1 + A2 )ξ ω n + ( A1 − A2 ) ξ 2 − 1 ω n = 0 (2.38b)
Solving Eqs. (2.38a) and (2.38b)

( ξ + ξ 2 − 1)
A1 = Z0 (2.39a)
2 ξ 2 −1

( − ξ + ξ 2 − 1)
A2 = Z0 (2.39b)
2 ξ2 −1
Equation (2.35) can be written as
e − ξωntξ(e
Z= ξ2 − 1 ω n t
− e− ξ2 − 1 ω n t
)+ ξ 2 − 1 (e ξ2 − 1 ω n t
+ e− ξ2 − 1 ω n t
) Z 0 (2.40a)
2 ξ − 1 
2

Z 1 (ξ + ξ 2 − 1)e( − ξ +
or = ξ2 − 1 ) ω n t
+ (− ξ + ξ 2 − 1)e( − ξ − ξ2 − 1 ) ω n t 
(2.40b)
Z 0 2 ξ 2 − 1  

Since the power of e is negative in both the terms of Eq. (2.40b), they both decreases exponentially
with t; therefore Z will decrease exponentially with the increase in t. Z will become zero when t will
Z
become infinity. Plots of versus wnt are shown by firm lines in Fig. 2.9 for different values of
Z0
damping ratio (x > 1). In all these curves the system does not cross the equilibrium position. Higher
the damping, more sluggish is the response of the system. Hence if x > 1, the motion is aperiodic
and no oscillation occurs.
Case II: x = 1 (Critically damped system)
In this case, from Eq. (2.34)
l1 = l2 = – wn
The solution of Eq. (2.30) is given by
− ωnt
Z = ( A1 + A2 t )e (2.41)
The above equation is the solution of a system having critical damping.
Using same boundary conditions as given in Eq. (2.36),
A1 = Z0, and A2 = wnZo (2.41)
Therefore,
Z = (1 + ω n t )e − ωn t . Z o
(2.42a)
or
Z
= (1 + ω n t )e − ωn t (2.42b)
Z0
26 Dynamics of Soils and Their Engineering Applications

The value of Z/Z0 in the above equation also decreases with the increase in t as shown by dotted
line curve in Fig. 2.9. This is also an aperiodic motion, and in this case the displacement-time curve
lies below any of the curves for over-damped system.

Fig. 2.9. Displacement time plots of over-damped and critically damped systems

Case III: x < 1 (Underdamped system)

The roots l1 and l2 are complex and are given by

λ1.2 =  − ξ ± i 1 − ξ 2  (2.43)
 
The complete solution of Eq. (2.28) is given by

Z = A1e( − ξ ± i 1 − ξ2 ) ω n t
+ A2 e( − ξ − i 1 − ξ2 ) ω n t
(2.44a)
or

Z = e − ξωn t  A1ei 1 − ξ2 ω n t
+ A2 e − i 1 − ξ2 ω n t 
 (2.44b)
The Eq. (2.44b) can be written as

Z = e − ξωn t C1 sin(ω n 1 − ξ 2 t ) + C2 cos(ω n 1 − ξ 2 t ) (2.44c)

 

or Z = e − ξωn t [C1 sin ω nd t + C2 cos ω nd t ] (2.44d)

2
where     ω nd = ω n 1 − ξ = Damped natural frequency
     C1 = A1 + A2 and C2 = i(A1 – A2)
In Eq. (2.44d), constants C1 and C2 are real which make A1 and A2 complex conjugate quantities.
The motion of the system is oscillatory (Fig. 2.10) and the amplitude of vibration goes on
decreasing in an exponential fashion.
Theory of Vibrations 27

Fig. 2.10. Free vibrations of a viscously underdamped system

As a convenient measure of damping, we may compute the ratio of amplitudes of the successive
cycles of vibration
Z1 e − ωn ξt
= − ωn ξ (t + 2 π / ωnd ) (2.45a)
Z2 e

Z1
or = e ωn ξ. 2 π / ωnd (2.45b)
Z2

Z1 1 − ξ2
or = e 2 πξ / (2.45c)
Z2

Z1 2πξ
log e =
or Z2 1 − ξ 2 (2.45d)
  Z  
Natural logarithm of ratio of two successive peak amplitudes i.e. log e  1   is called as
logarithmie decrement.   Z 2  

1 Z
or         ξ = log e 1 , as for small values of ξ, 1− ξ 2 = 1 (2.45e)
2π Z2
Thus, damping of a system can be obtained from a free vibration record by knowing the
successive amplitudes which are one cycle apart.
If the damping is very small, it may be convenient to measure the differences in peak amplitudes
for a number of cycles, say n.
In such a case, if Zn is the peak amplitudes of the nth cycle, then
Z 0 Z1 Z 2 Zn − 1
= = = ... = = eδ where d = 2px
Z1 Z 2 Z 3 Zn
Z 0  Z 0   Z1   Z 2   Z n −1  nδ
   Therefore, =       ...  =e
Z n  Z1   Z 2   Z 3   Z n 

1 Z
   Hence δ= log e 0 (2.45f)
n Zn
1 Z
or ξ= log e 0 (2.45g)
2πn Zn
Variation of damped natural frequency with damping is shown in Fig. 2.11. The decrease in
damped natural frequency with increase in damping ratio is small initially but is very steep as x
increases further. Therefore the damped natural frequency may be taken approximately equal to
undamped natural frequency for lower values of x (<0.4).
28 Dynamics of Soils and Their Engineering Applications

Fig. 2.11. Variation of damped natural frequency with damping

Therefore, a system is:

   over damped if x > 1;
   critically damped if x = 1 and
   under damped if x < 1.

2.4.3 Forced Vibrations of Single Degree Freedom System

In many cases of vibrations caused by rotating parts of machines, the systems are subjected to
periodic exciting forces. Let us consider the case of a single degree freedom system which is acted
upon by a steady state sinusoidal exciting force having magnitude F and frequency w (i.e. F(t) = F0
sin wt). For this case the equation of motion (Eq. 2.11) can be written as:

mZ + CZ + K Z = F0 sin ω t (2.46)

Equation (2.46) is a linear, non-homogeneous, second order differential equation. The solution
of this equation consists of two parts namely (i) complementary function, and (ii) particular integral.
The complementary function is obtained by considering no forcing function. Therefore the equation
of motion in this case will be:
mZ1 + CZ1 + K Z1 = 0 (2.47)
The solution of Eq. (2.47) has already been obtained in the previous section and is given by,
Z1 = e − ξ ωn t (C1 sin ω nd t + C2 cos ω nd t ) (2.48)
Here Z1 represents the displacement of mass m at any instant t when vibrating without any
forcing function.
The particular integral is obtained by rewriting Eq. (2.45) as
m Z2 + C Z 2 + K Z 2 = F0 sin ωt (2.49)
where Z2 is the displacement of mass m at any instant t when vibrating with forcing function.
The solution of Eq. (2.49) is given by
Z 2 = A1 sin ω t + A2 cos ω t (2.50)
where A1 and A2 are two arbitrary constants.
Substituting Eq. (2.50) in Eq. (2.49)
m(–A1w2 sin w t – A2 w2cos w t) + C(A1 w cos w t – A2 w sin w t) +
K(A1 sin wt + A2 cos wt) = F0 sin w t (2.51)
Theory of Vibrations 29

Considering Sine and Cosine functions in Eq. (2.50) separately,

(–mA1 ω2 + KA1 – CA2 w) sin wt = F0 sin w t (2.52a)
(–mA2 ω2 + KA2 – CA1 w) cos wt = 0 (2.52b)
From Eq. (2.52a),
K  C  F
A1  − ω 2  − A2  − ω = 0 (2.53a)
m  m  m
and from Eq. (2.52b)
K  C 
A1  − ω + A2  − ω 2  = 0 (2.53b)
m  m 
   Solving Eqs. (2.53a) and (2.53b), we get
( K − m ω 2 ) F0
A1 = (2.53c)
( K − mω 2 ) + C 2 ω 2
− C ω F0
and A2 = (2.53d)
( K − mω 2 ) 2 + C 2 ω 2
By substituting the values of A1 and A2 in Eq. (2.50),
F0
Z2 = {( K − m ω 2 ) sin ω t − C ω cos ω t} (2.54)
( K − mω 2 ) 2 + C 2 ω 2
Cω
let, tan θ = (2.55)
K − mω 2
By substituting Eq. (2.55) in Eq. (2.54), one can obtain
F0
Z2 = .sin (ω t − θ) (2.56)
( K − m ω 2 )2 + C 2 ω 2
Eq. (2.56) may be written as
F0 / K
Z2 = .sin (ω t − θ) (2.57)
(1− η2 ) + (2ηξ) 2
ω
where h = frequency ratio =
ωn
C C
x = Damping ratio = =
Cc 2 Km
The complete solution is obtained by adding the complimentary function and the particular
integral. Since the complimentary function is an exponentially decaying function, it will die out
soon and the motion will be described by only the particular integral (Fig. 2.12). The system will
vibrate harmonically with the same frequency as the forcing frequency, and the peak amplitude is
given by
F0 / K
AZ = (2.58)
2 2 2
(1 − η ) + (2 ηξ )

The quantity F0/K is equal to the static deflection of the mass under force F0. Dynamic
magnification factor, m is defined as the ratio of the dynamic amplitude AZ to the static deflection,
and is given by
30 Dynamics of Soils and Their Engineering Applications

Fig. 2.12. Superposition of transient and steady state vibrations

1
µ= (2.59)
(1 − η ) + (2 ηξ) 2
2 2

The variation of m versus h is shown in Fig. 2.13 for different values of damping ratio x. It
would be seen that near h = 1, the value of m is maximum. This is called resonance and the forcing
frequency f at which it occurs is called the resonant frequency.
Differentiating Eq. (2.59) with respect to h and equating to zero, it can be shown that resonance
will occur at a frequency ratio given by

h= 1 − 2ξ 2 (2.60a)
which is approximately equal to unity for small values of x.
2
or wnd = ωn 1− 2ξ (2.60b)

where wnd = Damped resonant frequency

Theory of Vibrations 31

Fig. 2.13. Magnification factor (m) versus frequency ratio (h)

By substituting Eq. (2.60) in Eq. (2.59), the maximum value of magnification factor is obtained.
It is given by
1
mmax = (2.61)
2ξ 1− ξ 2
1
= (For small values of x) (2.62)
2ξ
Assuming a damping of 5% in a structure, its amplitude at resonance will be 10 times the static
deflection. This indicates that systems will be subjected to very large amplitudes at resonance which
should be avoided.
The phase angle q given by Eq. (2.55) indicates the phase difference between the motion and
the exciting force. It can be written as

 2 ηξ 
q = tan −1  (2.63)
 1 − η2 
Variation of q with respect to h is shown in Fig. 2.14.
It is important to remember that the damping ratio and any of these parameters are simply the
parameters used to understand the effects of damping. They allow the effects of energy dissipation
to be represented in a convenient manner mathematically. For most soils and structures, however,
energy is dissipated hysteorically (i.e. by yielding or plastic straining of material).In such cases the
behavior is more accurately characterized by evaluating the non-linear response of the system.
32 Dynamics of Soils and Their Engineering Applications

Fig. 2.14. Phase lag versus frequency ratio for different amounts of damping.

2.4.3.1 Rotating Mass Type Excitation

Machines with rotating masses develop alternating force as shown in Fig. 2.15a. Since horizontal
forces on the foundation at any instant cancel, the net vibrating force on the foundation is vertical
and equal to 2 me ew2 sin wt, where me is the mass of each rotating element, placed at eccentricity e
from the centre of rotating shaft and w is the angular frequency of masses. Figure 2.15b shows such
a system mounted on elastic supports with dashpot representing viscous damping.

Fig. 2.15. Single degree freedom system with rotating mass type excitation

The equation of motion can be written as

mZ + CZ + K Z = 2 me eω 2 sin ωt (2.64)
Theory of Vibrations 33

where m is the mass of foundation including 2 me. Equations (2.64) and (2.46) are similar, except
that 2 me e w2 appears in Eq. (2.64) in place of Fo. The solution of Eq. (2.64) may therefore be
written as
Z = AZ sin (w t + q) (2.65)

(2me e / m) ⋅ η2
where AZ = (2.66)
(1 − η2 ) 2 + (2ξη) 2
Since Fo = 2 me · ew2

Fo ω2 ω2  e
or = 2 me ⋅ e = 2me ⋅ e =  2me  ⋅ η2
K K 2
(mω n )  m

−1  2 ηξ 
q = tan  (2.67)
 1 − n 
Equation (2.66) can be expressed in non-dimensional form as given below:
Az η2
= (2.68)
(2me e / m) (1− η2 ) 2 + (2ηξ) 2
   The value of Az/(2mee/m) is plotted against frequency ratio h in Fig. 2.16a. The curves are
similar in shape to those in Fig. 2.13 except that these start from origin. The variation of phase angle
q with h is shown in Fig. 2.16 b. Differentiating Eq. (2.68) with respect to h and equating to zero, it
can be shown that resonance will occur at a frequency ratio given by

1
h= (2.69a)
1 − 2ξ 2
ωn
wnd = (2.69b)
1− 2ξ 2

By substituting Eq. (2.69a) in Eq. (2.68), we get

 Az  1
  = (2.70)
 2me e / m  max 2ξ 1− ξ 2
1
≈
for small damping (2.71)
2ξ

2.5 Vibration Isolation

In case a machine is rigidly fastened to the foundation, the force will be transmitted directly to
the foundation and may cause objectionable vibrations. It is desirable to isolate the machine from
the foundation through a suitable designed mounting system in such a way that the transmitted
force is reduced. For example, the inertial force developed in a reciprocating engine or unbalanced
forces produced in any other rotating machinery should be isolated from the foundation so that
the adjoining structure is not set into heavy vibrations. Another example may be the isolation of
delicate instruments from their supports which may be subjected to certain vibrations. In either
case the effectiveness of isolation may be measured in terms of the force or motion transmitted to
the foundation. The first type is known as force isolation and the second type as motion isolation.
34 Dynamics of Soils and Their Engineering Applications

Fig. 2.16. Response of a system with rotating unbalance

2.5.1 Force Isolation

Figure 2.17 shows a machine of mass m supported on the foundation by means of an isolator having
an equivalent stiffness K and damping coefficient C. The machine is excited with unbalanced
vertical force of magnitude 2 me ew2 sin w t. The equation of motion of the machine can be written
as:
mZ + CZ + KZ = 2me eω 2 sin ω t (2.72)
The steady state motion of the mass of machine can be worked out as
2me eω 2 / K
Z= ⋅ sin (ωt − θ) (2.73)
(1 − η2 ) 2 + (2 ηξ) 2
Theory of Vibrations 35

Fig. 2.17. Machine-isolator-foundation system

 2 ηξ 
where q = tan −1  (2.74)
2
1 − η 
The only force which can be applied to the foundation is the spring force KZ and the damping
force CŻ; hence the total force transmitted to the foundation during steady state forced vibration is
Ft = KZ + CZ
(2.75)
Substituting Eq. (2.73) in Eq. (2.75), we get
2me eω 2 C ⋅ 2me eω 2 / K
Ft = ⋅ sin (ωt − θ) + ⋅ ω cos (ωt − θ) (2.76)
(1 − η2 ) 2 + (2 ηξ) 2 (1 − η2 ) 2 + (2 ηξ) 2
Equation (2.76) can be written as:

1 + (2 ηξ) 2
Ft = 2me eω 2 ⋅ sin (ωt − β) (2.77)
(1 − η2 ) 2 + (2 ηξ) 2
where b is the phase difference between the exciting force and the force transmitted to the foundation
and is given by

−1  C ω 
b = θ − tan  (2.78)
 K 
Since the force 2 me e w2 is the force which would be transmitted if springs were infinitely rigid,
a measure of the effectiveness of the isolation mounting system is given by

Ft 1 + (2 ηξ) 2
mT = = (2.79)
2 me e ω 2 (1 − η2 ) 2 + (2 ηξ) 2
mT is called the transmissibility of the system. A plot of mT versus h for different values of
x is shown in Fig. 2.18. It will be noted from the figure that for any frequency ratio greater than
2 , the force transmitted to the foundation will be less than the exciting force. However in this
case, the presence of damping reduces the effectiveness of the isolation system as the curves for
damped case are above the undamped ones for η > 2 . A certain amount of damping, however, is
essential to maintain stability under transient conditions and to prevent excessive amplitudes when
36 Dynamics of Soils and Their Engineering Applications

Fig. 2.18. Transmissibility (mT) versus frequency ratio (h)

the vibrations pass through resonance during the starting or stopping of the machine. Therefore, for
the vibration isolation system to be effective h should be greater than 2 .

2.5.2 Motion Isolation

In many situations, it would be necessary to isolate structure or mechanical systems from vibrations
transmitted from the neighbouring machines. Again we require a suitable mounting system so
that least vibrations are transmitted to the system due to the vibrating base. We consider a system
mounted through a spring and dashpot and attached to the surface which vibrates harmonically with
frequency w and amplitude Y0 as shown in Fig. 2.19.
Let Z be the absolute displacement of mass; the equation of motion of the system can be written
as:
m Z + C ( Z − Y ) + K ( Z − Y ) = 0 (2.80)

or m Z + K Z + K Z = C Y + K Y = C w Y0 cos w t + K Y0 sin w t

m Z + C Z + K Z = Y0 K 2 + (C ω ) 2 sin (ω t + α)
or (2.81)
Theory of Vibrations 37

Fig. 2.19. Motion isolation system

Cω
where a = tan −1 (2.82)
K
The solution of Eq. (2.81) will give the maximum amplitude as:

1+ (2ηξ) 2
Zmax = Y0 . (2.83)
(1− η2 ) 2 + (2ηξ) 2
The effectiveness of the mounting system (transmissibility) is given by

Z max 1 + (2 ηξ) 2
mT = = (2.84)
Y0 (1 − η2 ) 2 + (2 ηξ) 2

Equation (2.84) is the same expression as Eq. (2.79) obtained earlier. Transmissibility of such
system can also be studied from the response curves shown in Fig. 2.18. It is again noted that for the
vibration isolation to be effective, it must be designed in such a way that η > 2 .

2.5.3 Materials Used in Vibration Isolation

Materials used for vibration isolation are rubber, felt, cork and metallic springs. The effectiveness of
each depends on the operating conditions.

2.5.3.1 Rubber
Rubber is loaded in compression or in shear, the latter mode gives higher flexibility. With loading
greater than about 0.6N per sq mm, it undergoes much faster deterioration. Its damping and stiffness
properties vary widely with applied load, temperature, shape factor, excitation frequency and the
amplitude of vibration. The maximum temperature upto which rubber can be used satisfactorily is
about 65ºC. It must not be used in presence of oil which attacks rubber. It is found very suitable for
high frequency vibrations.

2.5.3.2 Felt
Felt is used in compression only and is capable of taking extremely high loads. It has very high
damping and so is suitable in the range of low frequency ratio. It is mainly used in conjunction with
metallic springs to reduce noise transmission.
38 Dynamics of Soils and Their Engineering Applications

2.5.3.3 Cork
Cork is very useful for acoustic isolation and is also used in small pads placed underneath a large
concrete block. For satisfactory working it must be loaded from 10 to 25 N/sq mm. It is not affected
by oil products or moderate temperature changes. However, its properties change with the frequency
of excitation.

2.5.3.4 Metallic Springs

Metallic springs are not affected by the operating conditions or the environments. They are quite
consistent in their behaviour and can be accurately designed for any desired conditions. They have
high sound transmissibility which can be reduced by loading felt in conjunction with it. It has
negligible damping and so is suitable for working in the range of high frequency ratio.

2.6 Theory of Vibration Measuring Instruments

The purpose of a vibration measuring instrument is to give an output signal which represents, as
closely as possible, the vibration phenomenon. This phenomenon may be displacement, velocity
or acceleration of the vibrating system and accordingly the instrument which reproduces signals
proportional to these are called vibrometers, velometers or accelerometers.
There are essentially two basic systems of vibration measurement. One method is known as
the directly connected system in which motions can be measured from a reference surface which is
fixed. More often such a reference surface is not available. The second system, known as “Seismic
system” does not require a fixed reference surface and therefore is commonly used for vibration
measurement.
Figure 2.20 shows a Vibration measuring instrument which is used to measure any of the
vibration phenomena. It consists of a frame in which the mass m is supported by means of a spring
K and dashpot C. The frame is mounted on a vibrating body and vibrates along with it. The system
reduces to a spring mass dashpot system having base on support excitation as discussed in Section
2.5.2 illustrating motion isolation.

Fig. 2.20. Vibration measuring instrument

Let the surface S of the structure be vibrating harmonically with an unknown amplitude Yo
and an unknown frequency w. The output of the instrument will depend upon the relative motion
between the mass and the structure, since it is this relative motion which is detected and amplified.
Let Z be the absolute displacement of the mass, then the output of the instrument will be proportional
to X = Z – Y. The equation of motion of the system can be written as:
mZ + C ( Z − Y ) + K ( Z − Y ) = 0 (2.85)
Theory of Vibrations 39

Subtracting mŸ from both sides

m X + C X + K X = − mY = mY0 ω 2 sin ω t (2.86)
The solution can be written as:
η2
Y0 sin (ω t − θ)
X= (2.87)
(1− η2 ) 2 + (2ηξ) 2
ω
where h = = frequency ratio
ωn

x = damping ratio

 2ηξ 
and q = tan −1  2 
 1− η 
Equation (2.87) can be rewritten as:
X = h2 m Y0 sin(w t – q) (2.88)

1
where m=
(1− η ) + (2ηξ) 2
2 2

2.6.1 Displacement Pickup

The instrument will read the displacement of the structure directly if h2 m = 1 and q = 0. The
variation of h2 m with h and x is shown in Fig. 2.21. The variation of q with h is already given in
Fig. 2.14. It is seen that when h is large, h2 m is approximately equal to 1 and q is approximately
equal to 180º. Therefore to design a displacement pickup, h should be large which means that the
natural frequency of the instrument itself should be low compared to the frequency to be measured.
Or in other words, the instrument should have a soft spring and heavy mass. The instrument is
sensitive, flimsy and can be used in a weak vibration environment. The instrument cannot be used
for measurement of strong vibrations.

2.6.2 Acceleration Pickup (Accelerometer)

Equation (2.88) can be rewritten as:
1
X = 2 µ Y0 ω2 sin (ω t − θ)
(2.89)
ωn
The output of the instrument will be proportional to the acceleration of the structure if m is
constant. Figure 2.13 shows the variation of m with h and x. It is seen that m is approximately equal
to unity for small values of h. Therefore to design an acceleration pickup, h should be small which
means that the natural frequency of the instrument itself should be high compared to the frequency
to be measured. In other words, the instrument should have a stiff spring and small mass. The
instrument is less sensitive and suitable for the measurement of strong motion. The instrument size
is small.

2.6.3 Velocity Pickup

Equation (2.88) can be rewritten as:
1
X=
ηµ Y0 ω sin (ω t − θ) (2.90)
ωn
40 Dynamics of Soils and Their Engineering Applications

Fig. 2.21. Response of a vibration measuring instrument to a vibrating base

1
The output of the instrument will be proportional to velocity of the structure if h m is
constant. At h = 1. Eq. (2.90) can be written as ωn

1 1 1
X= Y0 ω sin (ω t − θ)    Q at h = 1, µ = (2.91)
ωn 2ξ 2ξ

Since wn and x are constant, the instrument will measure the velocity at h = 1.
It may be noted that the same instrument can be used to measure displacement, acceleration and
velocity in different frequency ranges.
X a Y if h > > 1 Displacement pickup (Vibrometers)
X a Ÿ if h < < 1 Acceleration pickup (Accelerometers)
X a Y if h = 1 Velocity pickup (Velometers)
Displacement and velocity pickups have the disadvantage of having rather a large size if
motions having small frequency of vibration are to be measured. Calibration of these pickups is not
simple. Further, corrections have to be made in the observations as the response is not flat in the
starting regions. From the point of view of small size, flat frequency response, sturdiness and ease
of calibration acceleration pickups are to be favoured. They are relatively less sensitive and this
disadvantage can easily be overcome by high gain electronic instrumentation.

2.6.4 Design of Acceleration Pickup

The relative displacement between the mass and the support would be a measure of the support
acceleration if h is less than 0.75 and x is of the order of 0.6 to 0.7. Of the various methods of
measurement of relative displacement, electrical gauging, in which the mechanical quantity is
converted into an equivalent electrical quantity is best suited for acceleration pickups. Electrical
gauging offers the possibility of high magnification of the signals which are usually weak because the
Theory of Vibrations 41

spring is stiff and the displacements are small. The mechanical quantity alters either the resistance,
or capacitance or the inductance of the circuit which consequently alters the current in the circuit.

2.7 Vibration of Multiple Degree Freedom Systems

In the preceding sections, vibrations of systems having single degree of freedom have been
discussed. In many engineering problems, one may come across the systems which may have more
than one degree of freedom. Two degrees freedom cases arise when the foundation of the system
is yielding thus adding another degree of freedom or a spring mass system is attached to the main
system to reduce its vibrations. In systems when there are a number of masses connected with each
other, even if each mass is constrained to have one degree of freedom, the system as a whole has as
many degrees of freedom as there are masses. Such an idealisation is done for carrying out dynamic
analysis of multistoreyed buildings.

Fig. 2.22. A four storeyed frame with mode shapes

Figure 2.22a shows the frame work of a four storeyed building. It is usual to lump the masses
at the floor levels and the lumped mass has a value corresponding to weight of the floor, part of
the supporting system (columns) above and below the floor and effective live load. The restoring
forces are provided by the supporting systems. Figure 2.22b shows such an idealisation and it gives
a four degrees of freedom system. In free vibration a system having four degrees of freedom has
four natural frequencies and the vibration of any point in the system, in general, is a combination of
four harmonics of these four natural frequencies respectively. Under certain conditions, any point
in the system may execute harmonic vibrations at any of the four natural frequencies, and these are
known as the principal modes of vibration. Figures 2.22c to 2.22f show the four modes of vibration.
If all the masses vibrate in phase (Fig. 2.22c), the mode is termed the first or lowest or fundamental
mode of vibration and the frequency associated with this mode would be the lowest in magnitude
compared to other modes. If all adjacent masses vibrate out of phase with each other (Fig. 2.22f), the
mode is termed the highest mode of vibration and the frequency associated with this mode would be
highest in magnitude compared to other modes.
In many aspects, the response of multi-degrees of freedom system, procedure of analysis is
analogous to those described above for single degree freedom system. Although the additional
degrees of freedom complicate the mathematics, the procedure is conceptually quite similar. In fact,
a very useful approach to the response of linear multi-degrees freedom system allows their response
to be computed as the sum of the responses as the series of single degree of freedom systems. One
has to keep in mind that in evaluating the response of a multi- degree freedom system, dynamic
equilibrium of all masses must be ensured simultaneously.
42 Dynamics of Soils and Their Engineering Applications

2.7.1 Two Degrees of Freedom Systems

2.7.1.1 Undamped Free Vibration

Figure 2.23 shows a mass-spring system with two degrees of freedom. Let Z1 be the displacement of
mass m1 and Z2 the displacement of mass m2. The equations of motion of the system can be written:
m1 Z1 + K1 Z1 + K 2 ( Z1 − Z 2 ) = 0 (2.92)
m2 Z2 + K 3 Z 2 + K 2 ( Z 2 − Z1 ) = 0 (2.93)
The solution of Eqs. (2.92) and (2.93) will be of the following form:
Z1 = A1 sin wn t (2.94)
Z2 = A2 sin wn t (2.95)
Substitution of Eqs. (2.94) and (2.95), into Eqs. (2.92) and (2.93), yields:
( K1 + K 2 − m1 ω2n ) A1 − K 2 A2 = 0 (2.96)
2
− K 2 A1 + ( K 2 + K 3 − m2 ω ) A2 = 0
n (2.97)

Fig. 2.23. Free vibration of a two degrees freedom system

For nontrivial solutions of wn in Eqs. (2.96) and (2.97),

K1 + K 2 − m1ω2n − K2
=0 (2.98)
− K2 K 2 + K 3 − m2 ω2n

 K + K 2 K 2 + K 3  2 K1 K 2 + K 2 K 3 + K 3 K1
or ω4n −  1 +  ωn + =0 (2.99)
 m1 m2  m1m2
n
   Equation (2.99) is quadratic in ω2 , and the roots of this equation are:

 1/ 2

1  K1 + K 2 K 2 + K 3   K1 + K 2 K 2 + K 3  4 K 22  
2
2
ω = +   − +  (2.100)
2  m1 m2   m1 m2 
n
m1 m2  
   
From Eq. (2.100), two values of natural frequencies wn1 and wn2 can be obtained. wn1 is
corresponding to the first mode and wn2 is of the second mode.
Theory of Vibrations 43

The general equation of motion of the two masses can now be written as
Z1 = A1(1) sin ωn1 t + A1( 2 ) sin ωn 2 t
(2.101)
(1) ( 2)
Z2 = A2 sin ωn1 t + A2 sin ωn 2 t
and (2.102)
The superscripts in A represent the mode.
The relative values of amplitudes A1 and A2 for the two modes can be obtained using Eqs. (2.96)
and (2.97).
A1(1) K2 K + K 3 − m2 ω2n1
Thus, (1) = 2
= 2 (2.103)
A2 K1 + K 2 − m1ωn1 K2

A1( 2 ) K2 K 2 + K 3 − m2 ω n2 2
= = (2.104)
A2( 2 ) K 2 + K 2 − m1ω 2n 2 K2

2.7.1.2 Undamped Forced Vibrations

Consider the system shown in Fig. 2.24 with excitation force Fo sin w t acting on mass m1. In this
case, equations of motion will be:
m1 Z1 + K1Z1 + K 2 ( Z1 − Z 2 ) = F0 sin w t (2.105)
m2 Z2 + K 3 Z 2 + K 2 ( Z 2 − Z1 ) = 0 (2.106)
For steady state, the solutions will be as
Z1 = A1 sin w t
(2.107)
Z2 = A2 sin w t
(2.108)

Fig. 2.24. Forced vibration of a two degrees freedom system

Substituting Eqs. (2.107) and (2.108) in Eqs. (2.105) and (2.106), we get
(K1 + K2 – m1 w2) A1 – K2A2 = F0 (2.109)
– K2A1 + (K2 + K3 – m2w2)A2 = 0 (2.110)
Solving for A1 and A2 from the above two equations, we get
( K 2 + K 3 − m2 ω2 ) F0
A1 = (2.111a)
  K + K 2 K 2 + K 3  2 K1 K 2 + K 2 K 3 + K 3 K1 
m1m2 ω4 −  1 + ω + 
  m1 m2  m1m2 
44 Dynamics of Soils and Their Engineering Applications

K 3 F0
and A2 = (2.111b)
  K + K 2 K 2 + K 3  2 K1 K 2 + K 2 K 3 + K 3 K1 
m1m2 ω4 −  1 + ω + 
  m1 m2  m1m2 

The above two equations give steady state amplitude of vibration of the two masses respectively,
as a function of w. The denominator of the two equations is same. It may be noted that:
(i) The expression inside the bracket of the denominator of Eqs. (2.111a) and (2.111b) is of the
same type as the expression of natural frequency given by Eq. (2.99), Therefore at w = wn1 and
w = wn2 values of A1 and A2 will be infinie as the denominator will become zero.
(ii) The numerator of the expression for A1 becomes zero when
( K 2 + K3 )
ω=
m2 (2.112)
Thus it makes the mass m1 motionless at this frequency. No such stationary condition exists for
mass m1. The fact that the mass which is being excited can have zero amplitude of vibration under
certain conditions by coupling it to another spring-mass system forms the principle of dynamic
vibration absorbers which will be discussed in Art. 2.8.

2.7.2 System with n Degrees of Freedom

2.7.2.1 Undamped Free Vibrations
Consider a system shown in Fig. 2.25 having n degrees of freedom. If Z1, Z2, Z3, ... Zn are the
displacements of the respective masses at any instant, then equations of motion are:
m1 Z1 + K1 Z1 + K 2 ( Z1 − Z 2 ) = 0 (2.113)
m2 Z2 − K 2 ( Z1 − Z 2 ) + K 3 ( Z 2 − Z 3 ) = 0 (2.114)
m3 Z3 − K 3 ( Z 2 − Z 3 ) + K 4 ( Z 3 − Z 4 ) = 0 (2.115)
...........................................
..........................................

mn Zn − K n ( Z n −1 − Z n ) = 0 (2.116)
The solution of Eqs. (2.113) to (2.116) will be of the following form:
Z1 = A1 sin wn t (2.117)
Z2 = A2 sin wn t (2.118)
Z3 = A3 sin wn t (2.119)
..........................
..........................
Zn = An sin wn t (2.120)
Substitution of Eqs. (2.117) to (2.120) into Eqs. (2.113) to (2.116), yields:
( K1 + K 2 ) − m1 ω2n  A1 − K 2 A2 = 0 (2.121)

− K 2 A1 + ( K 2 + K 3 ) − m1 ω2n  A2 − K 3 A3 = 0 (2.122)

Theory of Vibrations 45

Fig. 2.25. Undamped free vibrations of a multi-degree freedom system

− K 3 A2 + ( K 3 + K 4 ) − m3 ω2n  A3 − K 4 A4 = 0 (2.123)

...........................................
...........................................
− K n An −1 + ( K n − mn ω 2n ) An = 0 (2.124)

For nontrivial solutions of wn in Eqs. (2.121) to (2.124),

( K1 + K 2 ) − m1ω 2n  − K2 ... 0 0
− K2 ( K 2 + K 3 ) − m2 ω n  ...
2
0 0
0 − K3 ... 0 0 =0 (2.125)
... ... ... ... ...
0 0 ... − K n ( K n − mn ω n2

Equation (2.125) is of nth degree in wn2 and therefore gives n values of wn corresponding to n
natural frequencies. The mode shapes can be obtained from Eq. (2.121) to (2.124) by using, at one
time, one of the various values of wn as obtained from Eq. (2.125).
46 Dynamics of Soils and Their Engineering Applications

When the number of degrees of freedom exceeds three, the problem of forming the frequency
equation and solving it for determination of frequencies and mode shapes becomes tedius. Numerical
techniques are invariably resorted to in such cases.
Holzer’s numerical technique is a convenient method of solving the problem for the system
idealized as shown in Fig. 2.26. By summing forces at free end,
Inertia force at a level below mass mi – 1

∑ m j Z j
i −1
j =1
(2.126)

Spring force at that level corresponding to the difference of adjoining masses

= Ki – 1(Zi – Zi – 1) (2.127)

Equating Eqs. (2.126) and (2.127)

∑ m j Zi = Ki – 1 (Zi – Zi – 1)
i −1
j =1
(2.128)

Putting Zi = Ai sin w t in Eq. (2.128), we get

∑
i −1
j =1
m j (− Aj ω2n sin ωn t ) = Ki – 1 (Ai sin wnt – Ai – 1 sin wn t)

ωn2
∑
i −1
or Ai = Ai −1 − m j Aj (2.129)
Ki − r j =1

Fig. 2.26. An idealised multi-degree freedom system

Equation (2.129) gives a relationship between any two successive amplitudes. Starting with any
arbitrary value of A1, amplitude of all other masses can be determined. A plot of An + 1 versus wn2
would have the shape as shown in Fig. 2.27. Finally An + 1 should worked out to zero due to fixity at
the base. The intersection of the curve with wn2 axis would give various values of wn2. Mode shape
can be obtained by substituting the correct value of wn2 in Eq. (2.129).
Theory of Vibrations 47

Fig. 2.27. Residual as a function of frequency in Hotzer method

2.7.2.2 Forced Vfibration

Let an undamped n degree of freedom system be subjected to forced vibration, and Fi (t) represents
the force on mas mi. The equation of motion for the mass mi will be
n
mi Zi + ∑ K ij Z j = Fi (t) (2.130)
i =1
where i = 1, 2, 3, ... , n
The amplitude of vibration of a mass is the algebraic sum of the amplitudes of vibration in
various modes. The individual modal response would be some fraction of the total response with the
sum of fractions being equal to unity. If the factors by which the modes of vibration are multiplied
are represented by the coordinates d, then for mass mi
Zi = Ai(1) di + Ai(2) d2 + ... + Ai(r) dr + ... + Ai(n) dn (2.131)
Equation (2.131) can be written as
n

∑A
(r )
Zi = i dr (2.132)
r =1

Substituting Eq. (2.132) in Eq. (2.130)

n n n

∑m A
r =1
i i
(r )
dr + ∑
r =1
∑ j =1
K ij A(j r ) d r = Fi (t) (2.133)
Under free vibrations, it can be shown
n

∑
i =1
K ij A(j r ) = ω2nr mi Ai( r ) (2.134)
Substituting Eq. (2.134) in Eq. (2.133), we get
n n

∑m A
r =1
i i
(r )
dr + ∑ ωnr
2
mi Ai( r ) d r = Fi(t)
r =1
(2.135)

n
or ∑m A
r =1
i i
(r )
(dr + ω 2nr . d r ) = Fi(t) (2.136)
Since the left hand side is a summation involving different modes of vibration, the right hand
side should also be expressed as a summation of equivalent force contribution in corresponding
modes.
Let Fi(t) be expanded as:
n
(r )
Fi(t) = ∑ mi Ai f r (t ) (2.137a)
r =1

where fr(t) is the modal force and given by

48 Dynamics of Soils and Their Engineering Applications

∑ F (t ) ⋅ A
i i
(r )

fr(t) = i =1
(2.137b)
n
2
∑ m ( Ai( r ) )
i=
Substituting Eq. (2.137a) in Eq. (2.136), we get
d + ω2 d = f (t )
r nr r r (2.138)
Equation (2.138) is a single degree freedom equation and its solution can be written as
t
1
dr =
ωnr ∫f
0
r (τ)sin ωnr (t − τ)d τ where 0 < t < 1 (2.139)

It is observed that the co-ordinate d, uncouples the n degree of freedom system into n systems
of single degree of freedom. The d’s are termed as normal co-ordinates and this approach is known
as normal mode theory. Therefore the total solution is expressed as a sum of contribution of
individual modes.

2.8 Undamped Dynamic Vibration Absorber

A system on which a steady oscillatory force is acting may
vibrate excessively, specially when close to resonance. Such
excessive vibrations can be eliminated by coupling a properly
designed spring mass system to the main system. This forms
the principle of undamped dynamic vibration absorber where
the excitation is finally transmitted to the auxiliary system,
bringing the main system to rest.
Let the combination of K and M be the schematic
representation of the main system under consideration with
the force F0 sin wt acting on it. A spring-mass (auxiliary)
absorber system is attached to the main system as shown in
Fig. 2.28. The equations of motion of the complete system
can be written as: Fig. 2.28. Vibration absorber
         MZ1 + KZ1 + K a ( Z1 − Z 2 ) = F0 sin wt (2.140)
ma Z2 + K a ( Z1 − Z 2 ) = 0 (2.141)
The forced vibration solution will be of the form
Z1 = A1 sin wt (2.142)
Z2 = A2 sin wt (2.143)
Substitution of Eqs. (2.142) and (2.143) in Eqs. (2.140) and (2.141) yields
A1 (– Mw2 + K + Ka) – KaA2 = F0 (2.144)
– Ka A1 + A2 (–maw + Ka) = 0 2 (2.145)
Substituting:
Fo
Z st =
= Static deflection of main system
K

2 Ka
ω na =
= Natural frequency of the absorber
ma
Theory of Vibrations 49

K
ω 2n = = Natural frequency of main system
M
m
µ m = a = Mass ratio = Absorber mas/Main mass
M
Equations (2.144) and (2.145) can be written as

 K ω2  K
A1 1+ a − 2  − A2 a = Zst (2.146)
 K ωn  K
A1
and A2 = (2.147)
 ω2 
1− 2 
 ωna 
Solving Eqs. (2.146) and (2.147) for A1 and A2, we get
ω2
1−
A1 ω2na
=
Z st  ω   K a ω2  K a
2 (2.148)
1− 2  1+ − −
 ωna   K ω2n  K
A2 1
= (2.149)
Z st  ω   K a ω2  K a
2

1− 2  1+ − −
 ωna   K ω2n  K

If the natural frequency wna of the absorber is chosen equal to w i.e.frequency of the excitation
force, it is evident from Eq. (2.148) that A1 = 0 indicating that the main mass does not vibrate at all.
Further Eq. (2.149) gives
A2 −K
=
Z st Ka

or A2Ka = – K Zst (2.150)

Thus the absorber system vibrate in such a way that its spring force at all instants is equal and
opposite to F0 sin w t. Hence, there is no net force acting on main mass M and the same therefore
does not vibrate.
The addition of a vibration absorber to a main system is not much meaningful unless the main
system is operating at resonance or at least near it. Under these conditions, w = wn. But for the
absorber to be effective, w should be equal to wna.
Therefore, for the effectiveness of the absorber at the operating frequency corresponding to the
natural frequency of the main system alone, we have
or wna = wn (2.151a)
Ka K
= (2.151b)
ma M
Ka ma
or = = µm (2.151c)
K M
When the condition enumerated in Eqs. (2.151) is fulfilled, the absorber is known as a tuned
absorber.
50 Dynamics of Soils and Their Engineering Applications

For a tuned absorber, Eqs. (2.148) and (2.149) become:

 ω2 
1−  2 
A1  ω na 
= (2.152)
Z st  ω2   ω2 
1 −
 ω 2   1 + µ − − µm
ω 2na 
m
na

A2 1
= (2.153)
Z st  ω 
2
ω2 
1− 2  1+ µ m − 2  − µm
 ωna   ωna 
The denominators of Eqs. (2.152) and (2.153) are identical. At a value of w when these
denominators are zero the two masses have infinite amplitudes of vibration. Let when w = wnl, the
denominators become zero. For this condition the expression for the denominators can be written
as
4 2
 ω nl   ω nl 

 ω  − (2 + µ m )  ω  + 1 = 0 (2.154)
na na

Equation (2.154) is quadratic in ω2nl , and therefore there are two values of wnl for which the
denominators of Eqs. (2.152) and (2.153) become zero. These two frequencies are the natural
frequencies of the system. Solution of Eq. (2.154) gives:
2
 ω nl   µm  µ 2m
 ω  =  1 +  ± µ m + (2.155)
na  2  4

The relationship of Eq. (2.155) is plotted in Fig. 2.29. From this plot, it is evident that greater
the mass ratio, greater is the spread between the two resonant frequencies. The frequency response

Fig. 2.29. Natural frequency ratio versus mass ratio

Theory of Vibrations 51

curve for the main system is shown in Fig. 2.30 for a value of mm = 0.2. The dotted curves shown
actually mean that the amplitude is negative or its phase difference with respect to the exciting force
is 180°. It can be noticed from this figure that by attaching a vibration absorber (wna = wn) to the
main system vibrating at resonance reduces its vibration to zero. Now if the exciting frequency is
absolutely constant, the system will work efficiently. Any change in the exciting frequency will shift
the operating point from the optimum point and the main system response will no longer be zero. It
may be noted that by adding the vibration absorber, we have introduced two resonant points instead
of one in the original system. Now if the variation of the exciting frequencyis such that the operating
point shifts near one of the new resonant points, then amplitudes will be excessive. Thus depending
upon the variation of the exciting frequencies the spread between the two resonant frequencies
has to be decided to remain reasonably away from the resonant points. After deciding the spread
between the resonant frequencies, a proper value of mm can be chosen from the curve of Fig. 2.29.
Undamped dynamic vibration absorbers are not suitable for varying forcing frequency excitation.
To make the vibration absorber effective over an extended range of frequencies of the disturbing
force, it is advantageous to introduce a damping device in the absorber system. Such an absorber
system is called a damped dynamic vibration absorber.

wnlnl
wna
na

Fig. 2.30. Response versus frequency of a vibration absorber

Equation (2.154) can also be written as

2
 ω2nl 
  − 1
 ωna 
mm = 2
(2.156)
 ωnl 
 
 ωna 
2.9 Dry Friction or Coulomb Damping
Systems discussed in Sections 2.4 to 2.6 have been analysed considering viscous damping which is
the most important type of damping, and it occurs for small velocities in lubricated sliding surfaces,
pistons with small clearances, etc. Eddy current damping is also of viscous nature. Dry friction or
coulomb damping occurs when two machine parts rub against each other, dry or un-lubricated. The
damping resistance in this case is practically constant, and is independent of the rubbing velocity.
52 Dynamics of Soils and Their Engineering Applications

Consider a single degree freedom system having a mass m and spring K, and let it move along
a rough horizontal plane which subjects to a constant force FD always opposing the motion (Fig.
2.31). The equation of motion becomes:
mZ + K Z ± FD = 0 (2.157)
or Z + (K/m)[Z ± (FD/K)] = 0 (2.158)
Let, Z ± (FD/K) = y
Z=Y
Equation (2.158) can be written as
Y + (K/m)y = 0 (2.159)
This is similar to Eq. (2.12b), therefore

wnd = ω n = K / m (2.160)
Hence the natural frequency of vibration of a system having Coulomb damping is the same as
that of undamped system.

Fig. 2.31. Free vibration with Coulomb damping

Equation (2.158) is non-linear because of sign change of FD. The period of free vibration
and variation of amplitude Z with time can be obtained as below. Let the mass be given an initial
displacement Zo and then released. The time required for the mass to attain its maximum displacement
on the other side of the equilibrium will be just one half of the period of vibration. The equation of
motion will be as follows:
mZ + KZ = FD (2.161)
The solution of Eq. (2.161) is
Z = A cos wnt + B sin wnt + (FD/K) (2.162)
Using the boundary conditions (At t = 0, Z = Zo and Ż = 0) we get, A = Zo – (FD/K), and B = 0.
   Therefore Z = [Zo – (FD/K] coswnt + (FD/K) (2.163)

   Now Ż = – [Zo – (FD/K)]wn. sin wn t (2.164)

At the extreme position on the opposite side, the velocity will be again zero and time required
for this half cycle will be:
Theory of Vibrations 53

wn t = p (2.165a)

or t = p/wn (2.165b)
Hence the period of free vibration will be twice the above value i.e.
T = 2p/wn (2.166)
The coulomb damping therefore does not change the natural frequency and time period, and
these remain the same as in undamped case. Equation (2.163) gives the variation of Z with time as
below:

t Z
0 Z0
π /(2 ωn ) FD / K
π / ωn − Z o + 2 FD / K

Similar analysis can be done for the other half cycle considering that the spring is compressed
by an amount (Zo – 2FD/K) and then left. It can be seen that after this half cycle, the displacement
becomes (Zo – 4FD/K). Hence after full one cycle the displacement reduces from Z to (Zo – 4FD/K).
Figure 2.32 shows a displacement-time plot of a system having free vibrations with coulomb
damping. It indicates that amplitude loss per cycle is 4FD/K.

Fig. 2.32. Displacement-time plot of a free vibration system having Coulomb damping

2.10 Systems under Transient Vibrations

As mentioned earlier, a system subjected to periodic excitation has two components of motion, the
transient and the steady state (Fig. 2.12). In most of such cases the transient part is not important
as it dies out soon and the steady part is the one that persists. However, where the excitation is of
non-periodic nature as caused by earthquakes, blasts, impacts, suddenly dropping of loads, etc. the
response of the system is purely transient. The maximum motion occurs within a relatively short
time after the application of force. For this reason, damping may be of little importance in transient
54 Dynamics of Soils and Their Engineering Applications

loads. However in the cases where force consists of series of pulses like in earthquakes, responses
are influenced by damping. In the subsequent paragraphs undamped transient vibrations have been
dealt in few typical cases.
Case 1: Suddenly Applied Load
Consider a mass-spring system shown in Fig. 2.33 subjected to a forcing function F (t) = FD. The
equation of motion for the mass m is given by:
m Z + K Z = F0 (2.167)
As explained in the earlier section, the system will have natural frequency wn as

wn = K / m (2.168)
The solution for displacement Z is given by
Z = A cos wn t + B sin wn t + F0/K (2.169)
The values of arbitrary constants A and B can be obtained considering two boundary conditions:
(i) At t = 0, Z = 0
(ii) At t = 0, Ż = 0
These give, A = – F0/K and B = 0
Therefore Eq. (2.169) becomes
Z = F0/K (1 – cos wn t) (2.170)
Magnification factor, m is given by
Z
m= = 1 − cos ω n t (2.171)
F0 / K
A plot of m versus time is given in Fig. 2.33c. It has a maximum value of 2, which occurs when
cos wn t = – 1. The first peak is reached when wn t = p or t = Tn/2, where Tn is the natural period of
vibration.

Fig. 2.33. System with suddenly applied load

Theory of Vibrations 55

Case 2: Rectangular Pulse

Consider a mass-spring system shown in Fig. 2.34a subjected to a rectangular pulse of uniform force
F0 for a given duration tp (Fig. 2.34b).
When t < tp, the equation of motion and m will be same as given by Eqs. (2.167) and (2.171)
respectively. At t = tp, the values of displacement and velocity are given by

Fig. 2.34. System with load having rectangular pulse

Zp = F0/K (1 – cos wn.tp)

(2.172a)
Żp = [(Fown)/K]sin wn tp
(2.172b)
For t > tp, the equation of motion will be
mZ + K Z = 0 (2.173)
The solution for displacement Z is given by
Z = A cos wn t + B sin wn t
(2.174)
The constants A and B can be obtained by using boundary conditions that at t = tp, Z = Zp, Ż = Żp.
Equation (2.174) becomes:
Z = F0/K [cos wn(t – tp) – cos wn t] (2.175a)
m = Z/(F0/K) = [cos wn (t – tp) – cos wn t]
(2.175b)
Plots of m versus time for two typical cases namely
(i) tp = Tn/3 = 2p/(3 wn) and
(ii) tp = Tn/4 = p/(2wn) are shown in Fig. 2.35.

Fig. 2.35. Response of systems with natural periods of a rectangular pulse

56 Dynamics of Soils and Their Engineering Applications

Illustrative Examples
Example 2.1
The motion of a particle is represented by the equation Z = 20 sin wt. Show the relative positions
and magnitudes of the displacement, velocity and acceleration vectors at time t = 0, and w = 2.0
rad/s and 0.5 rad/s.
Solution

Z = 20 sin wt

 π
Ż = 20 w cos w t = 20 ω sin  ωt + 
 2
Z = –20 w2 sin w t = 20 w2 sin(w t + p)

Fig. 2.36. Vector diagram (Example 2.1)

The magnitudes of displacement, velocity and acceleration vectors are 10, 10 w and 10 w2
respectively. The phase difference is such that the velocity vector leads the displacement vector by
p/2 and the acceleration vector leads the velocity vector by another p/2. Figures 2.36 a and 2.36 b
show the three vectors for w = 2.0 and 0.50 rad/s respectively.
2π 2 π
Time period = = = πs for w = 2.0 rad/s
ω 2
2π 2π
Time period = = = 4π s for w = 0.5 rad/s
ω (0.5)
Example 2.2
A body performs, simultaneously the motions
Z1 (mm) = 20 sin 8.0 t
Z2 (mm) = 21 sin 8.5 t
Determine the maximum and minimum amplitude of the combined motion, and the time period of
the periodic motion.
Solution Zmax = 21 + 20 = 41 mm
Zmin = 21 – 20 = 1 mm
The beat frequency is given by
8.5 − 8.0 0.5
f= = = 0.0795 Hz, and
2π 2π
Theory of Vibrations 57

1 2π
T= = = 4 π = 12.57 s
f 0.5
Example 2.3
A mass of 20 kg when suspended from a spring, causes a static deflection of 20 mm. Find the natural
frequency of the system.
W
Solution Stiffness of the spring, K =
δ st
20 × 9.81
K= ≈ 104 N/m
20 × 10 −3
1 K
Natural frequency, fn =
2π m

1 104
= = 3.6 Hz
2π 20
Example 2.4
For the system shown in Fig. 2.37, determine the natural frequency
of the system if
K1 = 1000 N/m        K2 = 500 N/m
K3 = 2000 N/m        K4 = K5 = 750 N/m`
Mass of the body = 5 kg
Solution
Let Ke1 and Ke2 represent respectively the effective stiffnesses of the
top three springs and the lower two springs, then
1 1 1 1 1 1 1
   = + + = + + = 0.0035
K e1 K1 K 2 K 3 1000 500 2000

      Ke1 = 285.7 N/m

      Ke2= K4 + K5 = 750 + 750 = 1500 N/m
Now Ke1 and Ke2 are two springs in parallel, therefore effective
stiffness,
      Ke = Ke1 + Ke2 = 285.7 + 1500 = 1785.7 N/m

1 K 1 1785.7
      fn = = = 3.0 Hz
2π m 2π 5.0
Fig. 2.37. Mass-springs system
Example 2.5
A vibrating system consists of a mass of 5 kg, a spring stiffness of 5 N/mm and a dashpot with a
damping coefficient of 0.1 N-s/m. Determine (i) damping ratio and (ii) logarithmic decrement.
Solution
−3
(i)       Cc = 2 k m = 2 5 × 10 × 5 = 0.319 N − s / m
C 0.1
       x = = = 0.313
Cc 0.319
58 Dynamics of Soils and Their Engineering Applications

2πξ 2π × 0.313
(ii) Logarithmic decrement = = = 2.07
2
1− ξ 1− 0.3132
Z1
log e = 2.07
Z2
Z1
i.e. = 7.92
Z2
Therefore the free amplitude in the next cycle decreases by 7.92 times.
Example 2.6
A mass attached to a spring of stiffness of 5 N/mm has a viscous damping device. When the mass
was displaced and released, the period of vibration was found to be 2.0 s, and the ratio of the
consecutive amplitudes was 10/3. Determine the amplitude and phase angle when a force F = 3 sin
4 t acts on the system. The unit of the force is Newton.
Solution

2πξ Z1 10
(i)      = log e = log e = 1.2
1− ξ Z2 3
or x = 0.195

(ii) Tn = 2.0 s

2π 2π
wn = = = 3.14 rad/s
T 2
w = 4.0 rad/s

ω 4.0
h= = = 1.273
ω n 3.14
F0 3.0
F0 = 3.0 N ; Ast = = = 0.6 mm
K 5.0
From Eq. (2.58),
Ast
Az = ; Ast = Static Deflection
(1 − η ) + (2ξη) 2
2 2

0.6
= = 0.755 mm
(1 − 1.273 ) + (2 × 0.195 × 1.273) 2
2 2

 2ηξ   2 × 1.273 × 0.195 

q = tan −1  2
= tan −1   = 141.4
o

 1− η   1 − 1.2732
Example 2.7
Show that in frequency-dependent excitation the damping factor x is given by the following
expression:

1  f 2 − f1 
x=  
2  2 fn 

where f1 and f2 are frequencies at which the amplitude is 1 / 2 times the peak amplitude.
Theory of Vibrations 59

Solution
In a forced vibration test, the system is excited with constant force of excitation and varying
frequencies. A response curve as shown in Fig. 2.38 is obtained.
At resonance, h = 1 and Az/Zst = 1/2x (for small values of x). If the frequency ratio is h when
amplitude of motion is 1 / 2 times the peak amplitude, then from Eq. 2.59, we get

Fig. 2.38. Determination of viscous damping in forced vibrations by Bandwidth method

1 1 1
⋅ =
2 2ξ (1− η ) + 4 η2 ξ 2
2 2

or h4 – 2h2(1 – 2x2) + (1 – 8x2) = 0

or η12.2 = 1  2(1− 2ξ 2 ) ± 4(1− 2ξ 2 ) 2 − 4(1− 8ξ 2 ) 

2 
2 2
= (1− 2ξ ) ± 2ξ 1+ ξ

   Now η22 − η12 = 4ξ 1− ξ 2 ≈ 4ξ [for small values of x]

η22 − η12 = f 2 − f1 =  f 2 − f1   f 2 + f1 
2 2

   Also, 2
fn  fn   fn 
 f −f  f +f
= 2  2 1  since 2 1 ≈ 2
 fn  fn

1  f 2 − f1 
   Therefore x=  
2  fn 

This method for determining viscous damping is known as the band width method.
60 Dynamics of Soils and Their Engineering Applications

Example 2.8
A machine of mass 100 kg is supported on springs of total stiffness of 784 N/mm. The machine
produces an unbalanced disturbing force of 392 N at a speed 50 c/s. Assuming a damping factor of
0.20, determine
(i) the amplitude of motion due to unbalance,
(ii) the transmissibility, and
(iii) the transmitted force.
Solution

784 × 103
(i) wn = K /m = = 87.7 rad/s
100
wn = 2p × 50 = 314 rad/s
ω 314
h= = = 3.58
ω n 87.7
F0 392
Zst = = = 0.5 mm
K 784
Z st 0.5
Now AZ = 2 2 2
=
(1− η ) + (2ηξ) (1− 3.58 ) + (2 × 3.58× 0.2) 2
2 2

= 0.042 mm

(1+ (2ηξ) 2 1+ (2 × 3.58× 0.2) 2

(ii) Transmissibility mT = =
(1− η2 ) 2 + (2ηξ) 2 (1− 3.582 ) 2 + (2 × 3.58× 0.2) 2
= 0.1467
(iii) Force transmitted = 392 × 0.1467 = 57.5 N
Example 2.9
The rotor of a motor having mass 2 kg was running at a constant speed of 30 c/s with an eccentricity
of 160 mm. The motor was mounted on an isolator with damping factor of 0.25. Determine the
stiffness of the isolator spring such that 15% of the unbalanced force is transmitted to the foundation.
Also determine the magnitude of the transmitted force.
Solution
(i) Maximum force generated by the motor
= 2 me ew2 = 2 × 2.0 × 0.16 × (2p × 30)2 = 22716 N = 22.72 kN
Force transmitted
(ii) mT = = 0.15
unbalanced force
1+ 4η2 ξ 2
i.e. = 0.15
(1− η2 ) 2 + (2ηξ) 2
or 1 + 4h2 × (0.25)2 = (0.15)2 [(1 – h2)2 + (2h × 0.25)2]
or h4 – 12.84 h2 – 43.44 = 0
ω
It gives h = 3.95 i.e. = 3.95
ωn
ω 60π
Therefore wn = K / m = = = 47.7 rad/s
3.95 3.95
Theory of Vibrations 61

K = m(47.7)2 = 2.0 × (47.7)2 = 4639 N/m

(iii) Force transmitted to the foundation = 0.15 × 22.72 = 3.4 kN.
Example 2.10
A seismic instrument with a natural frequency of 6 Hz is used to measure the vibration of a machine
running at 120 rpm. The instrument gives the reading for the relative displacement of the seismic
mass as 0.05 mm. Determine the amplitudes of displacement, velocity and acceleration of the
vibrating machine. Neglect damping.
Solution
(i) wn = 6 Hz = 37.7 rad/s
120 × 2π
w = 120 rpm = rad/s
= 12.57 rad/s
60
12.57
h= = 0.333
37.7
 1  1
m=  2
for ξ = 0  = = 1.125
 1 − η  1 − ( 0. 333) 2
(ii) For displacement pickup, Eq. (2.88) gives
X = h2 m Y0

0.05 = (0.333)2 × 1.125 × Y0

or Y0 = 0.40 mm

(iii) For velocity pickup, Eq. (2.91) gives

1
X= η µ(Y0 ω)
ωn
0.333
0.05 = ×1.125×Y0
(37.7)
or (Y0 w) = velocity = 5.03 mm/s
(iv) For acceleration pickup, Eq. (2.89) gives
µ
X= (Y0 ω2 )
ω2n
1.125
or 0.05 = (Y0 ω2 )
(37.7) 2
(37.7) 2 × 0.05
i.e. (Y0w2) = Acceleration = = 63.17 mm/s2
1.125
Example 2.11
Determine the natural frequencies and mode shapes of the system represented by a mathematical
model shown in Fig. 2.39a.
62 Dynamics of Soils and Their Engineering Applications

Fig. 2.39. Two degrees freedom system with mode shapes

Solution
(i) The system shown in Fig. 2.39a is a two degrees freedom system. The solution of such a
system has already been described in Section 2.7
(ii) The two natural frequencies of the system can be obtained using Eq. (2.100) by putting K1 =
K, K2 = 2 K and K3 = K, and m1 = m2 = m. By doing this, we get

1  3K 3K   4 × ( 2 K ) 2 
1/ 2
 K
ω2n1 =  +  −  =
2  m m   m2   m

1  6 K 4 K  5K
ω2n2 =  + =
2 m m  m
Hence, wn1 = K / m and wn2 = 5K / m
(iii) The relative values of amplitudes A1 and A2 for the two modes can be obtained using Eqs.
(2.103) and (2.104).
A1(1) K2 2K
(1) = = = +1
A2 K1 + K 2 − m1ω2n1 K + 2K − m × K / m

A1( 2 ) 2K
( 2)
= = −1
A2 K + 2 K − m × 5K / m

The mode shapes are shown in Figs. 2.39b and 2.39c.

Example 2.12
Determine the natural frequencies and mode shapes of the system represented by the mathematical
model shown in Fig. 2.40a.
Solution
(i) Equations of motion for the three masses can be written as
m Z1 + K Z1 + 2 K ( Z1 − Z 2 ) = 0 (2.157a)
m Z2 + 2 K ( Z 2 − Z1 ) + K ( Z 2 − Z 3 ) = 0 (2.157b)
Theory of Vibrations 63

Fig. 2.40. Three degrees freedom system with mode shapes

m Z3 + K ( Z 3 − Z 2 ) = 0 (2.157c)
For steady state, the solutions will be as
Z1 = A1 sin wn t (2.158a)
Z2 = A2 sin wn t (2.158b)
Z3 = A3 sin wn t (2.158c)
Substituting Eqs. (2.158) in Eqs. (2.157), we get
(3K − m ω2n ) A1 − 2 K A2 = 0 (2.159a)

− 2 K A1 + (3K − m ω2n ) A2 − K A3 = 0 (2.159b)

− K A2 + ( K − m ω2n ) A3 = 0 (2.159c)
For nontrivial solutions of wn in Eqs. (2.159)
3 K − m ω2n −2K 0
−2K 3K − m ω2n −K =0 (2.160)
0 −K K − m ωn2
m ω 2n
Putting λ = , Eq. (2.160) becomes
K
3− λ − 2 0
− 2 3 − λ −1 = 0 (2.161a)
0 −1 1 − λ
or l – 7 l + 10 l – 2 = 0
3 2 (2.161b)
Equation (2.161b) is cubic in l. The values of l are worked out as

l1 = 0.238, l2 = 1.637; and l3 = 5.129

Therefore, wn1 = 0.238 K / m ; wn2 = 1.637 K / m ; and wn3 = 5.129 K / m

64 Dynamics of Soils and Their Engineering Applications

(ii) Eqs. (2.159) in terms of l can be written as

(3 – l) A1 – 2A2 = 0 (2.162a)
–2A1 + (3 – l) A2 – A3 = 0 (2.162b)
– A2 + (1 – l) A3 = 0 (2.162c)
For I mode: l = 0.238
Equation (2.162a) gives
A1
(3 – 0.238)A1 – 2A2 = 0 or = 0.724
A2
Equation (2.162b) gives – 2 × 0.724 A2 + (3 – 0.238)A2 – A3 = 0
A2
or = 0.761
A3
Assuming A1 = a, A2 = 1.381 a and A3 = 1.815 a
A1 : A2 : A3 = a : 1.381 a : 1.815 a
= 1 : 1.381 : 1.815
= 0.551 : 0.761 : 1
Similarly,
For II mode; l = 1.637
A1 : A2 : A3 = – 0.933 : – 0.635 : 1, and
For III mode: l = 5.129
A1 : A2 : A3 = 3.891 : –4.14 : 1
The mode shapes are plotted in Figs. 2.40b, 2.40c and 2.40d.
Example 2.13
A small reciprocating machine weighs 50 kg and runs at a constant speed of 6000 rpm. After it was
installed, it was found that the forcing frequency is very close to the natural frequency of the system.
What dynamic absorber should be added if the nearest natural frequency of the system should be at
least 20 percent from the forcing frequency.
Solution
2 π N 2 π × 6000
(i) ω= = = 628 rad/s
60 60
At the time of installation of machine,
Forcing frequency ≈ Natural frequency of system
K
   Therefore, = 628
m
or K = m × 6282
= 50 × 6282 = 201 × 105 N/m
(ii) After adding the vibration absorber to the system, the natural frequency becomes (1 ± 0.2) 628
i.e. 753.6 rad/s or 502.4 rad/s
For tuned absorber:
ma K
= a = µM
M K
Theory of Vibrations 65

2
 ω  2 
 −1
nl

 ωna  
    Now from Eq. (2.156) mm = 
 ωn 
 
 ωna 
ωnl
for = 0.8
ωna
{(0.8) 2 −1}2
mm = = 0.2025
0.82
ωnl
and when = 1.2
ωna
{(1.2) 2 − 1}2
µm = = 0.134
1.22
Adopting the higher value of mm
Ka = 0.2025 × 201 × 105 = 40.7 × 105 N/m
ma = 0.2025 × 50 = 10.12 kg

Example 2.14
A horizontal mass-spring system with coulomb damping has a mass of 8.0 kg attached to a spring
of stiffness 1200 N/m. The coefficient of friction is 0.04. Determine (i) frequency of oscillation, (ii)
number of cycles corresponding to 60% reduction in amplitude if the initial amplitude is 60 mm, and
(iii) time taken to achieve this reduction.
Solution
(i) FD = m mg = 0.04 × 8.0 × 9.81 = 3.14 N

wn = K / m = 1200 / 8 = 12.25 rad/s

fn = wn/2p = 12.25/2p = 1.95 Hz

(ii) Initial amplitude = 60 mm = 0.06 m
Amplitude after 60% reduction = 0.036 m
Total reduction in amplitude = 0.06 – 0.036
= 0.024 m
Reduction in amplitude per cycle = 4FD /K = (4 × 3.14)/1200
= 0.0105 m
Number of cycles for 60% reduction = 0.024/0.0105 = 2.3 cycles
(iii) Time taken to achieve 60% reduction = Time taken to perform 2.3 cycles
= 2.3 × (2p/wn)
= 2.3 × (2p/12.25) = 1.18s

Practice Problems
2.1 A single degree (mass-spring-dashpot) system is subjected to a frequency dependent oscillatory
force (m e0 w2 sin w t). Proceeding from fundamentals, derive the expression of the amplitude
of the system.
2.2 ‘Presence of damping reduces the effectiveness of the isolation system’. Is this statement true?
If yes, explain with neat sketches.
66 Dynamics of Soils and Their Engineering Applications

2.3 Give two methods of determining ‘damping factor’ of a single degree freedom system.
2.4 Starting from fundamentals, explain the principles involved in the design of (i) Displacement
pickup, (ii) Velocity pickup, and (iii) Acceleration pickup. Illustrate your answer with neat
sketches.
2.5 Describe the principles involved in a ‘tuned dynamic vibration absorber’. Illustrate your
answer with neat sketches. Discuss clearly its limitations.
2.6 A mass of 25 kg when suspended from a spring, causes a static deflection of 25 mm. Find the
natural frequency of the system.
Ans. (20 rad/s)
2.7 A spring mass system (K1, m) has a natural frequency of f1. If a second spring of stiffness K2
is attached in series with the first spring, the natural frequency becomes f1/2. Determine K2 in
terms of K1.
Ans. (K1/3)
2.8 A mass of 5 kg is attached to the lower end of a spring whose upper end is fixed. The natural
period of this system is 0.40 s. Determine the natural period when a mass of 2.5 kg is attached
to the mid point of this spring with the upper and lower ends fixed.
Ans. (0.14 sec)
2.9 Determine the differential equation of motion of the system shown in Fig. 2.41. The moment
of inertia of weight W about the point O is J0. Show that the system becomes unstable when:
K ⋅a
b≥
W

Fig. 2.41. Mass-spring system

2.10 A body vibrating in a viscous medium has a period of 0.30 s and an inertial amplitude of 30
mm. Determine the logarithmic decrement if the amplitude after 10 cycles is 0.3 mm.
Ans. (0.46)
2.11 A vibration system consists of mass of 6 kg, a spring stiffness of 0.7 N/m and a dashpot with
a damping coefficient of 2 N-s/m. Determine
(a) Damping ratio
(b) Logarithmic decrement
Ans. (0.488, 3.55)
2.12 Write a differential equation of motion for the
system shown in Fig. 2.42 and determine the
natural frequency of damped oscillations and
the critical damping coefficient.

Fig. 2.42. Mass-spring dashpot system

Theory of Vibrations 67

2.13 A mass is attached to a spring of stiffness 6 N/mm with a viscous damping device. When the
mass was displaced and released, the period of vibration was found to be 1.8 s and the ratio of
consecutive amplitude was 4.2 to 1. Determine the amplitude and phase angle when a force F
= 2 sin 3t N acts on the system.
Ans. (0.708 mm, 56.4º)
2.14 A spring mass system is excited by a force F0 sin w t. At resonance the amplitude was measured
to be 100 mm. At 80% resonant frequency the amplitude was measured 80 mm. Determine the
damping factor of the system.
Ans. (0.1874)
2.15 Assuming small amplitudes, set up differential equation of motion for double pendulum using
the coordinates shown in Fig. 2.43. Show that the natural frequencies of the system as given
by the equation
g
ωn1, 2 = (2 ± 2 )
l
Determine the ratio of the amplitudes x1/x2.

Fig. 2.43. Double pendulum system

2.16 A motor weighs 220 kg and has rotating unbalance of 3000 N-mm. The motor is running at
constant speed of 2000 rpm. For vibration isolation, springs with damping factor of 0.25 is
used. Specify the springs for mounting such that only 20 percent of the unbalanced force is
transmitted to the foundation. Also determine the magnitude of the transmitted force.
Ans. (Ka = 931.22 kN/m, 26.3 kN)
2.17 A small reciprocating machine weighs 60 kg and runs at a constant speed of 5000 rpm. After it
was installed, it was found that the forcing frequency is very close to the natural frequency of
the system. What dynamic vibration absorber should be added if the nearest natural frequency
of the system should be at least 25 percent from the forcing frequency?
Ans. (15.3 kg, 4.2 × 106 N/m)
2.18 A mass of 1 kg is to be supported on a spring having a stiffness of 980 N/m. The damping
coefficient is 6.26 N-s/m. Determine the natural frequency of the system. Find also the
logarithmic decrement and the amplitude after three cycles if the initial displacement is 0.3
mm.
Ans. (31.14 rad/s, 0.628, 0.0456 mm)
2.19 A machine having a mass of 100 kg and supported on springs of total stiffness 7.84 × 105 N/m
has an unbalanced rotating element which results in a disturbing force of 392 N at a speed of
3000 rpm. Assuming a damping factor of 0.20, determine
68 Dynamics of Soils and Their Engineering Applications

(a) the amplitude of motion due to the unbalance,

(b) the transmissibility, and
(c) the transmitted force.
Ans. (0.043 mm, 0.148, 58.2 N)
2.20 The static deflection of the vibrometer mass is 20 mm. The instrument when attached to a
machine vibrating with a frequency of 125 cpm records a relative amplitude of 0.03 mm. Find
out for the machine,
(a) the amplitude of vibration.
(b) the maximum velocity of vibration, and
(c) the maximum acceleration of vibration.
Ans. (0.0576 mm, 0.754 mm/sec, 9.86 mm/sec2)