Undamped Single Degree-of-Freedom

System 1

The analysis and design of structures to resist the effect produced by time dependent forces or
motions requires conceptual idealizations and simplifying assumptions through which the physical
system is represented by an idealized system known as the analytical or mathematical model. These
idealizations or simplifying assumptions may be classified in the following three groups:

1. Material assumptions. These assumptions or simplifications include material properties such as

homogeneity or isotrophy and material behaviors such as linearity or elasticity.
2. Loading assumptions. Some common loading assumptions are to consider concentrated forces to
be applied at a geometric point, to assume forces suddenly applied, or to assume external forces to
be constant or periodic.
3. Geometric Assumptions. A general assumption for beams, frames and trusses is to consider these
structures to be formed by unidirectional elements. Another common assumption is to assume that
some structures such as plates are two-dimensional systems with relatively small thicknesses. Of
greater importance is to assume that continuous structures may be analyzed as discrete systems by
specifying locations (nodes) and directions for displacements (nodal coordinates) in the structures
as described in the following section.

1.1 Degrees of Freedom

In structural dynamics the number of independent coordinates necessary to specify the configuration
or position of a system at any time is referred to as the number of degrees of freedom. In general, a
continuous structure has an infinite number of degrees of freedom. Nevertheless, the process of
idealization or selection of an appropriate mathematical model permits the reduction to a discrete
number of degrees of freedom. Figure 1.1 shows some examples of structures that may be represented
for dynamic analysis as one-degree-of-freedom-systems, that is, structures modeled as systems with a
single displacement coordinate.

© Springer Nature Switzerland AG 2019 3

M. Paz, Y. H. Kim, Structural Dynamics, https://doi.org/10.1007/978-3-319-94743-3_1
4 1 Undamped Single Degree-of-Freedom System

Fig. 1.1 Examples of structures modeled as one-degree-of-freedom systems

These one-degree-of-freedom systems may be conveniently described by the analytical model

shown in Fig. 1.2 which has the following elements:

1. A mass element m representing the mass and inertial characteristic of the structure.
2. A spring element k representing the elastic restoring force and potential energy storage of the
structure.
3. A damping element c representing the frictional characteristics and energy dissipation of the
structure.
4. An excitation force F(t) representing the external forces acting on the structural system.

The force F(t) is written this way to indicate that it is a function of time. In adopting the analytical
model shown in Fig. 1.2, it is assumed that each element in the system represents a single property;
that is, the mass m represents only the property of inertia and not elasticity or energy dissipation,
whereas the spring k represents exclusively elasticity and not inertia or energy dissipation. Finally, the
damper c only dissipates energy. The reader certainly realizes that such “pure” elements do not exist
in our physical world and that analytical models are only conceptual idealizations of real structures.
As such, analytical models may provide complete and accurate knowledge of the behavior of the
model itself, but only limited or approximate information on the behavior of the real physical system.
Nevertheless, from a practical point of view, the information acquired from the analysis of the
analytical model may very well be sufficient for an adequate understanding of the dynamic behavior
of the physical system, including design and safety requirements.

Fig. 1.2 Analytical model for one-degree-of-freedom systems

1.2 Undamped System 5

1.2 Undamped System

We start the study of structural dynamics with the analysis of a fundamental and simple system, the
one-degree-of-freedom system in which we disregard or “neglect” frictional forces or damping. In
addition, we consider the system, during its motion or vibration, to be free from external actions or
forces. Under these conditions, the system is said to be in free vibration and it is in motion governed
only by the influence of the so-called initial conditions, that is, the given displacement and velocity at
time t ¼ 0 when the study of the system is initiated. This undamped, one-degree-of-freedom system is
often referred to as the simple undamped oscillator. It is usually represented as shown in Fig. 1.3a or
Fig. 1.3b or any other similar arrangement. These two figures represent analytical models that are
dynamically equivalent. It is only a matter of preference to adopt one or the other. In these models the
mass m is restrained by the spring k and is limited to rectilinear motion along one coordinate axis,
designated in these figures by the letter u.

Fig. 1.3 Alternate representations of analytical models for one-degree-of-freedom systems

The mechanical characteristic of a spring is described by the relationship between the magnitude
of the force Fs applied to its free end and the resulting end displacement u as shown graphically in
Fig. 1.4 for three different springs.

Fig. 1.4 Force-displacement relationship: (a) Hard spring, (b) Linear spring, (c) Soft spring

The curve labeled (a) in Fig. 1.4 represents the behavior of a hard spring in which the force
required to produce a given displacement becomes increasingly greater as the spring is deformed. The
second spring (b) is designated a linear spring because the deformation is directly proportional to the
6 1 Undamped Single Degree-of-Freedom System

force and the graphical representation of its characteristic is a straight line. The constant of
proportionality between the force and displacement [slope of the line (b)] of a linear spring is referred
to as the stiffness or the spring constant, usually designated by the letter k. Consequently, we may
write the relationship between force and displacement for a linear spring as

Fs ¼ ku ð1:1Þ

A spring with characteristics shown by curve (c) in Fig. 1.4 is known as a soft spring. For such a
spring the incremental force required to produce additional deformation decreases as the spring
deformation increases. Undoubtedly, the reader is aware from his or her previous exposure to
analytical modeling of physical systems that the linear spring is the simplest type to manage
mathematically. It should not come as a surprise to learn that most of the technical literature on
structural dynamics deals with models using linear springs. In other words, either because the elastic
characteristics of the structural system are, in fact, essentially linear, or simply because of analytical
expediency, it is usually assumed that the force-deformation properties of the system are linear. In
support of this practice, it should be noted that in many cases the displacements produced in the
structure by the action of external forces or disturbances are small in magnitude (Zone E in Fig. 1.4),
thus rendering the linear approximation close to the actual structural behavior.

1.3 Springs in Parallel or in Series

Sometimes it is necessary to determine the equivalent spring constant for a system in which two or
more springs are arranged in parallel as shown in Fig. 1.5a or in series as in Fig. 1.5b.

Fig. 1.5 Combination of springs: (a) Springs in parallel (b) Springs in series

For two springs in parallel the total force required to produce a relative displacement of their ends
of one unit is equal to the sum of their spring constants. This total force is by definition the equivalent
spring constant ke and is given by

ke ¼ k1 þ k2 ð1:2Þ

In general for n springs in parallel

X
n
ke ¼ ki ð1:3Þ
i¼1

For two springs assembled in series as shown in Fig. 1.5b, the force P produces the relative
displacements in the springs
1.4 Newton’s Law of Motion 7

P
Δu1 ¼
k1

and
P
Δu2 ¼
k2
Then, the total displacement u of the free end of the spring assembly is equal to u ¼ Δu1 + Δu2, or
substituting Δu1 and Δu2
P P
u¼ þ ð1:4Þ
k1 k2
Consequently, the force ke necessary to produce one unit displacement (equivalent spring constant) is
given by
P
ke ¼
u
Substituting u from this last relation into Eq. (1.4), we may conveniently express the reciprocal value
of the equivalent spring constant as
1 1 1
¼ þ ð1:5Þ
ke k1 k2

In general for n springs in series the equivalent spring constant may be obtained from

1 X n
1
¼ ð1:6Þ
ke k
i¼1 i

1.4 Newton’s Law of Motion

We continue with the study of the simple oscillator depicted in Fig. 1.3. The objective is to describe
its motion, that is, to predict the displacement or velocity of the mass m at any tine t, for a given set
of initial conditions at time t ¼ 0. The analytical relation between the displacement u, and time t, is
given by Newton’s Second Law of Motion, which in modern notation may be expressed as

F ¼ ma ð1:7Þ

Where F is the resultant force acting on a particle of mass m and a its resultant acceleration. The
reader should recognize that Eq. (1.7) is a vector relation and as such it can be written in equivalent
form in terms of its components along the coordinate axes x, y, and z, namely,
X
Fx ¼ max ð1:8aÞ
X
Fy ¼ may ð1:8bÞ
X
Fz ¼ maz ð1:8cÞ

The acceleration is defined as the second derivative of the position vector with respect to time; it
follows that Eqs. (1.8) are indeed differential equations. The reader should also be reminded that these
equations as stated by Newton are directly applicable only to bodies idealized as particles, that is,
bodies assumed to possess mass but no volume. However, as is proved in elementary mechanics,
8 1 Undamped Single Degree-of-Freedom System

Newton’s Law of Motion is also directly applicable to bodies of finite dimensions undergoing
translatory motion.
For plane motion of a rigid body that is symmetric with respect to the reference plane of motion
(x-y plane), Newton’s Law of Motion yields the following equations:
X
Fx ¼ mðaG Þx ð1:9aÞ
X
Fy ¼ mðaG Þy ð1:9bÞ
X
MG ¼ I G α ð1:9cÞ

In the above equations (aG)x and (aG)y are the acceleration components, along the x and y axes, of
the center of mass G of the body; α is the angular acceleration; IG is the mass moment of inertia of the
body with respect to an axis through G, the center of mass; and ∑MG is the sum with respect to an axis
through G, perpendicular to the x-y plane of the moments of all the forces acting on the body.
Equations (1.9) are certainly also applicable to the motion of a rigid body in pure rotation about a
fixed axis, alternatively, for this particular type of plane motion, Eq. (1.9c) may be replaced by
X
M0 ¼ I 0 α ð1:9dÞ

in which the mass moment of inertia I0 and the moment of the forces M0 are determined with respect
to the fixed axis of rotation. The general motion of a rigid body is described by two vector equations,
one expressing the relation between the forces and the acceleration of the mass center, and another
relating the moments of the forces and the angular motion of the body. This last equation expressed in
its scalar components is rather complicated, but seldom needed in structural dynamics.

1.5 Free Body Diagram

At this point, it is advisable to follow a method conducive to an organized and systematic analysis in
the solution of dynamics problems. The first and probably the most important practice to follow in
any dynamic analysis is to draw a free body diagram of the system, prior to writing a mathematical
description of the system.
The free body diagram (FBD), as the reader may recall, is a sketch of the body isolated from all
other bodies, in which all the forces external to the body are shown. For the case at hand, Fig. 1.6b
depicts the FBD of the mass m of the oscillator, displaced in the positive direction with reference to
coordinate u and acted upon by the spring force Fs ¼ ku (assuming a linear spring). The weight of the
body mg and the normal reaction N of the supporting surface are also shown for completeness, though
these forces, acting in the vertical direction, do not enter into the equation of motion written for the
u direction. The application of Newton’s Law of Motion gives

ku ¼ m€u ð1:10Þ

where the spring force acting in the negative direction has a minus sign, and where the acceleration
has been indicated by u€. In this notation, double overdots denote the second derivative with respect
to time and obviously a single overdot denotes the first derivative with respect to time, that is, the
velocity.
1.6 D’Alembert’s Principle 9

Fig. 1.6 Alternate free body diagrams: (a) Single degree-of-freedom system. (b) Showing only external forces, (c)
Showing external and inertial forces

1.6 D’Alembert’s Principle

An alternative approach to obtain Eq. (1.10) is to make use of D’Alembert’s Principle which states
that a system may be set in a state of dynamic equilibrium by adding to the external forces a fictitious
force that is commonly known as the inertial force.
Figure 1.6c shows the FBD with inclusion of the inertial force m€u. This force is equal to the mass
multiplied by the acceleration, and should always be directed negatively with respect to the corres-
ponding coordinate. The application of D’Alembert’s Principle allows us to use equations of
equilibrium in obtaining the equation of motion. For example, in Fig. 1.6c, the summation of forces
in the u direction gives directly
m€u þ ku ¼ 0 ð1:11Þ

which obviously is equivalent to Eq. (1.10).

The use of D’Alembert’s Principle in this case appears to be trivial. This will not be the case for a
more complex problem, in which the application of D’Alembert’s Principle, in conjunction with
the Principle of Virtual Work, constitutes a powerful tool of analysis. As will be explained later, the
Principle of Virtual Work is directly applicable to any system in equilibrium. If follows then that
this principle may also be applied to the solution of dynamic problems, provided that D’Alembert’s
Principle is used to establish the dynamic equilibrium of the system.

Illustrative Example 1.1

Show that the same differential equation is obtained for a body vibrating along a horizontal axis or for
the same body moving vertically, as shown in Fig. 1.7a, b.
Solution:
The FBDs for these two representations of the simple oscillator are shown in Fig. 1.7c, e, in which the
inertial forces have been included. Equating to zero in Fig. 1.7c the sum of the forces along the
direction u, we obtain
m€u þ ku ¼ 0 ðaÞ

When the body in Fig. 1.7d is in the static equilibrium position, the spring is stretched u0 units and
exerts a force ku0 ¼ W upward on the body, where W is the weight of the body. When the body is
displaced a distance u downward from this position of equilibrium the magnitude of the spring force
10 1 Undamped Single Degree-of-Freedom System

Fig. 1.7 Two representations of the simple oscillator and corresponding free body diagrams. (a) Idealized single
degree of freedom system, (b) Alternative idealized single degree of freedom system, (c) dynamic equilibrium with
inertial force, (d) static displacement due to gravity load, (e) dynamic equilibrium with inertial force of alternative
model

is given by Fs ¼ k(u0 + u) or Fs ¼ W + ku since ku0 ¼ W. Using this result and applying it to the body
in Fig. 1.7e, we obtain from Newton’s Second Law of Motion
ðW þ kuÞ þ W  m€u ¼ 0 ðbÞ

or
m€u þ ku ¼ 0

which is identical to Eq. (a).

1.7 Solution of the Differential Equation of Motion

The next step toward our objective is to find the solution of the differential Eq. (1.11). We should
again adopt a systematic approach and proceed first to classify this differential equation. Since the
dependent variable u and second derivative u€ appear in the first degree in Eq. (1.11), this equation is
classified as linear and of second order. The fact that the coefficients of u and of u€ (k and m,
respectively) are constants and that the second member (right-hand side) of the equation is zero
further classifies this equation as homogenous with constant coefficients. We should recall, probably
with a certain degree of satisfaction, that a general procedure exists for the solution of linear
1.7 Solution of the Differential Equation of Motion 11

differential equations (homogenous or non-homogenous) of any order. For this simple, second-order
differential equation we may proceed directly by assuming a trial solution given by

u ¼ A cos ωt ð1:12Þ

or by
u ¼ B sin ωt ð1:13Þ

where A and B are constants depending on the initiation of the motion while ω is a quantity denoting a
physical characteristic of the system as it will be shown next. The substitution of Eq. (1.12) into
Eq. (1.11) gives
 
mω2 þ k A cos ωt ¼ 0 ð1:14Þ

If this equation is to be satisfied at any time, the factor in parentheses must be equal to zero, or
k
ω2 ¼ ð1:15Þ
m
The reader should verify that Eq. (1.13) is also a solution of the differential Eq. (1.11), with ω also
satisfying Eq. (1.15).
The positive root of Eq. (1.15),
rffiffiffiffi
k
ω¼ ð1:16aÞ
m
is known as the natural frequency of the system for reasons that will soon be apparent.
The quantity ω in Eq. (1.16a) may be expressed in terms of the static displacement resulting from
the weight W ¼ mg applied to the spring. The substitution into Eq. (1.16a) of m ¼ W/g results in
rffiffiffiffiffi
kg
ω¼ ð1:16bÞ
W
Hence
rffiffiffiffiffi
g
ω¼ ð1:16cÞ
ust

where ust ¼ W/k is the static displacement of the spring due to the weight W.
Since either Eq. (1.12) or Eq. (1.13) is a solution of Eq. (1.11), and since this differential equation
is linear, the superposition of these two solutions, indicated by Eq. (1.17) below, is also a solution.
Furthermore, Eq. (1.17), having two constants of integration, A and B, is, in fact, the general solution
for this linear second-order differential equation.
u ¼ A cos ωt þ B sin ωt ð1:17Þ

The expression for velocity, u_ , is simply found by differentiating Eq. (1.17) with respect to time,
that is,

u_ ¼ Aω sin ωt þ Bω cos ωt ð1:18Þ

Next, we should determine the constants of integration A and B. These constants are determined
from known values for the motion of the system which almost invariably are the displacement u0 and
12 1 Undamped Single Degree-of-Freedom System

the velocity υ0 at the initiation of the motion, that is, at time t ¼ 0. These two conditions are referred to
as initial conditions, and the problem of solving the differential equation for the initial conditions is
called an initial value problem.
After substituting, for t ¼ 0, u ¼ u0, and u_ ¼ υ0 into Eqs. (1.17) and (1.18) we find that

u0 ¼ A ð1:19aÞ

υ0 ¼ Bω ð1:19bÞ

Finally, the substitution of A and B from Eq. (1.19) into Eq. (1.17) gives
υ0
u ¼ u0 cos ωt þ sin ωt ð1:20Þ
ω
which is the expression of the displacement u of the simple oscillator as a function of the time
variable t. Thus, we have accomplished our objective of describing the motion of the simple
undamped oscillator modeling structures with a single degree of freedom.

1.8 Frequency and Period

An examination of Eq. (1.20) shows that the motion described by this equation is harmonic and
therefore periodic, that is, it can be expressed by a sine or cosine function of the same frequency ω.
The period may easily be found since the functions sine and cosine both have a period of 2π. The
period T of the motion is determined from

ωT ¼ 2π

or
2π
T¼ ð1:21Þ
ω
The period is usually expressed in seconds per cycle or simply in seconds, with the tacit
understanding that it is “per cycle”. The reciprocal value of the period is the natural frequency f.
From Eq. (1.21)
1 ω
f ¼ ¼ ð1:22Þ
T 2π
The natural frequency f is usually expressed in hertz or cycles per second (cps). Because the
quantity ω differs from the natural frequency f only by the constant factor 2π, ω also is sometimes
referred to as the natural frequency. To distinguish between these two expressions for natural
frequency, ω may be called the circular or angular natural frequency. Most often, the distinction is
understood form the context or from the units. The natural frequency f is measured in cps as indicated,
while the circular frequency ω should be given in radians per second (rad/sec).

Illustrative Example 1.2

Determine the natural frequency of the beam-spring system shown in Fig. 1.8 consisting of a weight
of W ¼ 50.0 lb attached to a horizontal cantilever beam through the coil spring k2. The cantilever
beam has a thickness h ¼ ¼ in, a width b ¼ 1 in. modulus of elasticity E ¼ 30  106 psi, and length
L ¼ 12.5 in. The coil spring has a stiffness k2 ¼ 100 (lb/in).
1.8 Frequency and Period 13

Fig. 1.8 System for Illustrative Example 1.2

Solution:
The deflection Δ at the free end of a uniform cantilever beam acted upon by a static force P at the free
end is given by
PL3
Δ¼
3EI
The corresponding spring constant k1 is then
P 3EI
k1 ¼ ¼ 3
Δ L
where the cross-section moment of inertia I ¼ 12
1
bh3 (for a rectangular section). Now, the cantilever
and the coil spring of this system are connected as springs in series. Consequently, the equivalent
spring constant as given from Eq. (1.5) is
1 1 1
¼ þ ðrepeatedÞ ð1:5Þ
ke k1 k2

Substituting corresponding numerical values, we obtain

 3
1 1 1
I¼ 1 ¼ ðinÞ4
12 4 768
3  30  106
k1 ¼ ¼ 60 lb=in
ð12:5Þ3  768

and
1 1 1
¼ þ
ke 60 100
ke ¼ 37:5 lb=in

The natural frequency for this system is then given by Eq. (1.16a) as
pffiffiffiffiffiffiffiffiffiffi
ω ¼ ke =m ðm ¼ W=g and g ¼ 386 in= sec 2 Þ
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ω ¼ 37:5  386=50:0
ω ¼ 17:01 rad= sec

or using Eq. (1.22)

f ¼ 2:71 cps ðAnsÞ
14 1 Undamped Single Degree-of-Freedom System

1.9 Amplitude of Motion

Let us now examine in more detail Eq. (1.20), the solution describing the free vibratory motion of the
undamped oscillator. A simple trigonometric transformation may show us that we can rewrite this
equation in the equivalent forms, namely
u ¼ C sin ðωt þ αÞ ð1:23Þ
or

u ¼ C cos ðωt  βÞ ð1:24Þ

where
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C ¼ u20 þ ðυ0 =ωÞ2 ð1:25Þ

u0
tan α ¼ ð1:26Þ
υ0 =ω

and

υ0 =ω
tan β ¼ ð1:27Þ
u0

The simplest way to obtain Eq. (1.23) or Eq. (1.24) is to multiply and divide Eq. (1.20) by the factor
C defined in Eq. (1.25) and to define α (or β) by Eq. (1.26) [or Eq. (1.27)]. Thus

Fig. 1.9 Definition of angle α or angle β

 
u0 υ0 =ω
u¼C cos ωt þ sin ωt ð1:28Þ
C C

With the assistance of Fig. 1.9, we recognize that

u0
sin α ¼ ð1:29Þ
C
and

υ0 =ω
cos α ¼ ð1:30Þ
C
The substitution of Eqs. (1.29) and (1.30) into Eq. (1.28) gives
u ¼ Cð sin α cos ωt þ cos α sin ωtÞ ð1:31Þ
1.9 Amplitude of Motion 15

The expression within the parentheses of Eq. (1.31) is identical to sin(ωt + α), which yields
Eq. (1.23). Similarly, the reader should verify without difficulty, the form of solution given by
Eq. (1.24).
The value of C in Eq. (1.23) (or Eq. (1.24)) is referred to as the amplitude of motion and the angle α
(or β) as the phase angle. The solution for the motion of the simple oscillator is shown graphically in
Fig. 1.10.

Fig. 1.10 Undamped free-vibration response

Illustrative Example 1.3

Consider the steel frame shown in Fig. 1.11a having a rigid horizontal member to which a horizontal
dynamic force is applied. As part of the overall structural design it is required to determine the natural
frequency of this structure. Two assumptions are made:

1. The masses of the columns are neglected.

2. The horizontal members are sufficiently rigid to prevent rotation at the tops of the columns.

These assumptions are not mandatory for the solution of the problem, but they serve to simplify the
analysis. Under these conditions, the frame may be modeled by the spring-mass system shown in
Fig. 1.11b.

Fig. 1.11 One-degree-of-freedom frame and corresponding analytical model for Illustrative Example 1.3
16 1 Undamped Single Degree-of-Freedom System

Solution:
The parameters of this model may be computed as follows:
W ¼ 200  25 ¼ 5000 lb
I ¼ 82:5 in4
E ¼ 30  106 psi
12Eð2I Þ 12  30  106  165
k¼ ¼
L3 ð15  12Þ3

k ¼ 10, 185 lb=in ðAnsÞ

Note: A unit displacement of the top of a fixed column requires a force equal to 12EI/L3.
Therefore, the natural frequency from Eqs. (1.16b) and (1.22) is
rffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1 kg 1 10, 185  386
f ¼ ¼
2π W 2π 5000
f ¼ 4:46 cps ðAnsÞ

Illustrative Example 1.4

The elevated water tower tank with a capacity for 5000 gallons of water shown in Fig. 1.12a has a
natural period in lateral vibration of 1.0 sec when empty. When the tank is full of water, its period
lengthens to 2.2 sec. Determine the lateral stiffness k of the tower and the weight W of the tank.
Neglect the mass of the supporting columns (one gallon of water weighs approximately 8.34 lb).

Fig. 1.12 (a) Water tower tank of Illustrative Example 1.4. (b) Analytical model

Solution:
In its lateral motion, the water tower is modeled by the simple oscillator shown in Fig. 1.12b in which
k is the lateral stiffness of the tower and m is the vibrating mass of the tank.

(a) Natural frequency ωE (tank empty):

1.9 Amplitude of Motion 17

rffiffiffiffiffi
2π 2π kg
ωE ¼ ¼ ¼ ðaÞ
T E 1:0 W
(b) Natural frequency ωF (tank full of water)

Weight of water Ww:

W w ¼ 5000  8:34 ¼ 41, 700 lb
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2π 2π kg ðbÞ
ωF ¼ ¼ ¼
T F 1:0 W þ 41, 700
Squaring Eqs. (a) and (b) and dividing correspondingly the left and right sides of these equations,
results in

ð2:2Þ2 W þ 41, 700

2
¼
ð1:0Þ W

and solving for W

W ¼ 10, 860 lb ðAnsÞ

Substituting in Eq. (a), W ¼ 10,860 lb and g ¼ 386 in/sec2, yields

rffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2π k386
¼
1:00 10, 860

and
k ¼ 1110 lb=in ðAnsÞ

Illustrative Example 1.5

The steel frame shown in Fig. 1.13a is fixed at the base and has a rigid top W that weighs 1000 lb.
Experimentally, it has been found that its natural period in lateral vibration is equal to 1/10 of a
second. It is required to shorten or lengthen its period by 20% by adding weight or strengthening the
columns. Determine needed additional weight or additional stiffness (neglect the weight of the
columns).

Fig. 1.13 (a) Frame of Illustrative Example 1.5. (b) Analytical model
18 1 Undamped Single Degree-of-Freedom System

Solution:
The frame is modeled by the spring-mass system shown in Fig. 1.13b. Its stiffness is calculated from
rffiffiffiffiffi
2π kg
ω¼ ¼
T W
as
rffiffiffiffiffiffiffiffiffiffi
2π kg  
¼ g ¼ 386 in= sec 2
0:1 1000
or

k ¼ 10, 228 lb=in

(a) Lengthen the period to TL ¼ 1.2  0.10 ¼ 0.12 sec by adding weight ΔW:

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2π 10, 228  386
ω¼ ¼
0:12 1000 þ ΔW
Solve for ΔW:

ΔW ¼ 440 lb ðAnsÞ

(b) Shorten the period to Ts ¼ 0.8  0.1 ¼ 0.08 sec by strengthening columns in Δk:

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2π ð10; 228 þ ΔK Þð386Þ
ω¼ ¼
0:08 1000
Solve for Δk:
Δk ¼ 5753 lb=in ðAnsÞ

1.10 Response of SDF Using MATLAB Program

Plot the displacement as a function of time, u(t) ranging from 0 to 5 sec.

Given:

• Initial conditions: u0 ¼ 1 in. and u_ 0 ¼ 0:2 in:= sec

• Natural period: T ¼ 0.5 sec (Fig. 1.14).
1.10 Response of SDF Using MATLAB Program 19

clear all
clc

%%%%-GIVEN VALUES-%%%%
%%% Set Initial Conditions

u0=1; %Initial Displacement

v0=2; %Initial Velocity

%%%Define period and frequency

T=0.5; %Natural Period

omega=2*pi/T; %Natural Frequency

%%%%-ESTIMATION-%%%%
%%% Generate time stamp equally between 0 to 5 sec with a total of 500 %%% data

t=linspace(0,5,500);

%%%Calculate the displacement response

A = u0; % Eq. 19a

B = v0/omega; % Eq. 19b

u= A*cos(omega*t)+B*sin(omega*t); % Eq. 20

C= sqrt(A^2+B^2); %Maximum amplitude (Eq. 1.25)

udot = -A*omega*sin(omega*t)+B*omega*cos(omega*t); %Velocity

%%%Plot the response curve

plot(t, u);
title ('Response');
xlabel ('Time (sec)'); %Label the x-axis of the plot
ylabel ('Displacement (in.)'); %Label the y-axis of the plot

Response
1.5

1
Displacement (in.)

0.5

-0.5

-1

-1.5
0 1 2 3 4 5
Time (sec)

Fig. 1.14 Response of SDF using MATLAB

20 1 Undamped Single Degree-of-Freedom System

1.11 Summary

Several basic concepts were introduced in this chapter:

1. The analytical or mathematical model of a structure is an idealized representation for its analysis.
2. The number of degrees of freedom of a structural system is equal to the number of independent
coordinates necessary to describe its position.
3. The free body diagram (FBD) for dynamic equilibrium (to allow application of D’Alembert’s
Principle) is a diagram of the system isolated from all other bodies, showing all the external forces
on the system, including the inertial force.
4. The stiffness or spring constant of a linear system is the force necessary to produce a unit
displacement.
5. The differential equation of the undamped simple oscillator in free motion is

m€u þ ku ¼ 0

and its general solution is

u ¼ A cos ωt þ B sin ωt

where A and B are constants of integration determined from initial conditions of the displacement
u0 and of the velocity υ0:
A ¼ u0
B ¼ υ0 =ω
pffiffiffiffiffiffiffiffiffi
ω ¼ k=m is the natural frequency in rad= sec
ω
f ¼ is the natural frequency in cps
2π
1
T ¼ is the natural period in seconds
f

6. The equation of motion may be written in the alternate forms:

u ¼ C sin ðωt þ αÞ

or

u ¼ C cos ðωt  βÞ

where
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C ¼ u20 þ ðυ0 =ωÞ2

and
u0
tan α ¼
υ0 =ω
υ0 =ω
tan β ¼
u0
1.12 Problems 21

1.12 Problems

Problem 1.1
Determine the natural period for the system in Fig. P1.1. Assume that the beam and springs
supporting the weight W are massless.

Fig. P1.1

Problem 1.2
The following numerical values are given in Problem 1.1: L ¼ 100 in. EI ¼ 108 (lb.in2), W ¼ 3000 lb,
k ¼ 2000 lb/in. If the weight W has an initial displacement of u0 ¼ 1.0 in and an initial velocity of
u0 ¼ 20 in/sec, determine the displacement and the velocity 1 sec later.

Problem 1.3
Determine the natural frequency for horizontal motion of the steel frame in Fig. P1.3. Assume the
horizontal girder to be infinitely rigid and neglect the mass of the columns.

Fig. P1.3

Problem 1.4
Calculate the natural frequency in the horizontal mode of the steel frame in Fig. P1.4 for the following
cases:
22 1 Undamped Single Degree-of-Freedom System

(a) The horizontal member is assumed to be rigid.

(b) The horizontal member is flexible and made of steel sections-- W 8  24.

Hint: When the girder stiffness needs to be considered to determine the effective stiffness of column
fixed on the ground, the following formula is useful.
ke ¼ kc ðleftÞ þ kc ðrightÞ
24Ec I c ð1 þ 6γ Þ
¼
h3 ð4 þ 6γ Þ
where,
I g =L
γ¼
I c =h
I g and L are the moment of inertia and span length for girder:
I c and h are the moment of inertia and height of column:

Fig. P1.4

Problem 1.5
Determine the natural frequency of the fixed beam in Fig. P1.5 carrying a concentrated weight W at its
center. Neglect the mass of the beam.

Fig. P1.5

Problem 1.6
The numerical values for Problem 1.5 are given as: L ¼ 120 in. EI ¼ 109 (lb.in2), W ¼ 5000 lb, with
initial conditions u0 ¼ 0.5 in and υ0 ¼ 15 in/sec. Determine the displacement, velocity, and
acceleration of W at t ¼ 2 sec later. Plot the responses (i.e., displacement, velocity, and acceleration)
using MATLAB and determine the maximum amplitude.
1.12 Problems 23

Problem 1.7
Consider the simple pendulum of weight W illustrated in Fig. P1.7. If the cord length is L, determine
the motion of the pendulum. The initial angular displacement and initial angular velocity are θ0 and
θ_ 0 , respectively (Assume the angle θ is small).
Note: A simple pendulum is a particle of concentrated weight that oscillates in a vertical arc and is
supported by a weightless cord. The only forces acting are those of gravity and the cord tension (i.e.,
frictional resistance is neglected).

Fig. P1.7

Problem 1.8
A diver standing at the end of a diving board that cantilevers 2 ft oscillates at a frequency 2 cps.
Determine the flexural rigidity EI of the diving board. The weight of the diver is 180 lb. (Neglect the
mass of the diving board).

Problem 1.9
A bullet weighing 0.2 lb is fired at a speed of 100 ft/sec into a wooden block weighing W ¼ 50 lb
and supported by a spring of stiffness 300 lb/in (Fig. P1.9). Determine the displacement u(t) and
velocity u(t) of the block after t sec.

Fig. P1.9

Problem 1.10
An elevator weighing 500 lb is suspended from a spring having a stiffness of k ¼ 600 lb/in. A weight
of 300 lb is suspended through a cable to the elevator as shown schematically in Fig. P1.10.
Determine the equation of motion of the elevator if the cable of the suspended weight suddenly
breaks.
24 1 Undamped Single Degree-of-Freedom System

Fig. P1.10

Problem 1.11
Write the differential equation of motion for the inverted pendulum shown in Fig. P1.11 and
determine its natural frequency. Assume small oscillations, and neglect the mass of the rod.

Fig. P1.11

Problem 1.12
Show that the natural frequency for the system of Problem 1.11 may be expressed as
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
W
f ¼ f0 1 
W cr

where W ¼ mg, Wcr is the critical buckling weight, and f0 is the natural frequency neglecting the effect
of gravity.

Problem 1.13
A vertical pole of length L and flexural rigidity EI carries a mass m at its top, as shown in Fig. P1.13.
Neglecting the weight of the pole, derive the differential equation for small horizontal vibrations of
the mass, and find the natural frequency. Assume that the effect of gravity is small and neglect
nonlinear effects.
1.12 Problems 25

Fig. P1.13

Problem 1.14
Show that the natural frequency for the system in Problem 1.13 may be expressed as
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
W
f ¼ f0 1 
W cr

where f0 is the natural frequency calculated neglecting the effect of gravity and Wcr is the critical
buckling weight.

Problem 1.15
Determine an expression for the natural frequency of the weight W in each of the cases shown in
Fig. P1.15. The beams are uniform of cross-sectional moment of inertia I and modulus of elasticity
E. Neglect the mass of the beams.

Fig. P1.15
26 1 Undamped Single Degree-of-Freedom System

Problem 1.16
A system is modeled by two freely vibrating masses m1 and m2 interconnected by a spring having a
constant k as shown in Fig. P1.16. Determine for this system the differential equation of motion for
the relative displacement ur ¼ u2 – u1 between the two masses. Also determine the corresponding
natural frequency of the system.

Fig. P1.16

Problem 1.17
Calculate the natural frequency for the vibration of the mass m shown in Fig. P1.17. Member AE is
rigid with a hinge at C and a supporting spring of stiffness k at D. (Problem contributed by Professors
Vladimir N. Alekhin and Alesksey A. Antipin of the Urals State Technical University, Russia.)

Fig. P1.17

Problem 1.18
Determine the natural frequency of vibration in the vertical direction for the rigid foundation
(Fig. P1.18) transmitting a uniformly distributed pressure on the soil having a resultant force
Q ¼ 2000 kN. The area of the foot of the foundation is A ¼ 10 m2. The coefficient of elastic
compression of the soil is k ¼ 25,000 kN/m3. (Problem contributed by Professors Vladimir N. Alekhin
and Alesksey A. Antipin of the Urals State Technical University, Russia.)

Fig. P1.18
1.12 Problems 27

Problem 1.19
Calculate the natural frequency of free vibration of the chimney on elastic foundation (Fig. P1.19),
permitting the rotation of the structure as a rigid body about the horizontal axis x-x. The total weight
of the structure is W with its center of gravity at a height h from the base of the foundation. The mass
moment of inertia of the structure about the axis x-x is I and the rotational stiffness of the soil is
k (resisting moment of the soil per unit rotation). (Problem contributed by Professors Vladimir
N. Alekhin and Alesksey A. Antipin of the Urals State Technical University, Russia.)

Fig. P1.19
 Damped Single Degree-of-Freedom
System 2

We have seen in the preceding chapter that the simple oscillator under idealized conditions of no
damping, once excited, will oscillate indefinitely with a constant amplitude at its natural frequency.
However, experience shows that it is not possible to have a device that vibrates under these ideal
conditions. Forces designated as frictional or damping forces are always present in any physical
system undergoing motion. These forces dissipate energy; more precisely, the unavoidable presence
of these frictional forces constitute a mechanism through which the mechanical energy of the system,
kinetic or potential energy, is transformed to other forms of energy such as heat. The mechanism of
this energy transformation or dissipation is quite complex and is not completely understood at this
time. In order to account for these dissipative forces in the analysis of dynamic systems, it is
necessary to make some assumptions about these forces, on the basis of experience.

2.1 Viscous Damping

In considering damping forces in the dynamic analysis of structures, it is usually assumed that these
forces are proportional to the magnitude of the velocity, and opposite to the direction of motion. This
type of damping is know as viscous damping; it is the type of damping force that could be developed
in a body restrained in its motion by a surrounding viscous fluid.
There are situations in which the assumption of viscous damping is realistic and in which the
dissipative mechanism is approximately viscous. Nevertheless, the assumption of viscous damping is
often made regardless of the actual dissipative characteristics of the system. The primary reason for
such wide use of this assumed type of damping is that it leads to a relatively simple mathematical
analysis.

2.2 Equation of Motion

Let us assume that we have modeled a structural system as a simple oscillator with viscous damping,
as shown in Fig. 2.1a. In this figure m and k are, respectively, the mass and spring constant of the
oscillator and c is the viscous damping coefficient.

© Springer Nature Switzerland AG 2019 29

M. Paz, Y. H. Kim, Structural Dynamics, https://doi.org/10.1007/978-3-319-94743-3_2
30 2 Damped Single Degree-of-Freedom System

Fig. 2.1 (a) Viscous damped oscillator. (b) Free body diagram

We proceed, as in the case of the undamped oscillator, to draw the free body diagram (FBD) and
apply Newton’s Law to obtain the differential equation of motion. Figure 2.1b shows the FBD of the
damped oscillator in which the inertial force m€u is also shown, so that we can use D’Alembert’s
Principle. The summation of forces in the u direction gives the differential equation of motion
m€u þ cu_ þ ku ¼ 0 ð2:1Þ
The reader may verify that a trial solution u ¼ A sin ω t or u ¼ B cos ω t will not satisfy Eq. (2.1).
However, the exponential function u ¼ Cept does satisfy this equation. Substitution of this function
into Eq. (2.1) results in the equation

mCp2 ept þ cCpept þ kCept ¼ 0

which, after cancellation of the common factors, reduces to an equation called the characteristic
equation for the system, namely

mp2 þ cp þ k ¼ 0 ð2:2Þ

The roots of this quadratic equation are

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
r
P1 c c 2 k
¼   ð2:3Þ
P2 2m 2m m
thus the general solution of Eq. (2.1) is given by the superposition of the two possible solutions,
namely

uðtÞ ¼ C1 ep1 t þ C2 ep2 t ð2:4Þ

where C1 and C2 are constant of integration to be determined from the initial conditions.
The final form of Eq. (2.4) depends on the sign of the expression under the radical in Eq. (2.3).
Three distinct cases may occur; the quantity under the radical may either be zero, positive or negative.
The limiting case in which the quantity under the radical is zero is treated first. The damping present
in this case is called critical damping.

2.3 Critically Damped System

For a system oscillating with critical damping (c ¼ ccr), the expression under the radical in Eq. (2.3) is
equal to zero, that is
 c 2 k
cr
 ¼0 ð2:5Þ
2m m
or
2.4 Overdamped System 31

pffiffiffiffiffiffi
ccr ¼ 2 km ð2:6Þ

where ccr designates the critical damping value. Since the natural frequency of the undamped system
pffiffiffiffiffiffiffiffiffi
is given by ω ¼ k=m, the critical damping coefficient given by Eq. (2.6) may also be expressed in
alternative expressions as
2k
ccr ¼ 2mω or ccr ¼ ð2:7Þ
ω
In a critically damped system the roots of the characteristic equation are equal, and from Eq. (2.3),
they are
ccr
p1 ¼ p 2 ¼  ð2:8Þ
2m
Since the two roots are equal, the general solution given by Eq. (2.4) would provide only one
independent constant of integration, hence, one independent solution, namely

u1 ðtÞ ¼ C1 eðccr =2mÞt ð2:9Þ

Another independent solution may be found by using the function

u2 ðtÞ ¼ C2 tu1 ðtÞ ¼ C2 teðccr =2mÞt ð2:10Þ

u2(t), as the reader may verify, also satisfies the differential Eq. (2.1). The general solution for a
critically damped system is then given by the superposition of these two solutions,

uðtÞ ¼ ðC1 þ C2 tÞeðccr =2mÞt ð2:11Þ

2.4 Overdamped System

In an overdamped system, the damping coefficient is greater that the value for critical damping, namely
c > ccr ð2:12Þ

Therefore, the expression under the radical of Eq. (2.3) is positive; thus the two roots of the
characteristic equation are real and distinct, and consequently the solution is given directly by
Eq. (2.4). It should be noted that for the overdamped or the critically damped system, the resulting
motion is not oscillatory; the magnitude of the oscillations decays exponentially with time to zero.
Figure 2.2 depicts graphically the response for the simple oscillator with critical damping. The
response of the overdamped system is similar to the motion of the critically damped system of
Fig. 2.2, but in the return toward the neutral position requires more time as the damping is increased.

Fig. 2.2 Free-vibration response with critical damping

32 2 Damped Single Degree-of-Freedom System

2.5 Underdamped System

When the value of the damping coefficient is less than the critical value (c < ccr), which occurs when
the expression under the radical is negative, the roots of the characteristic Eq. (2.3) are complex
conjugates, so that
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
P1 c k  c 2
¼ i  ð2:13Þ
P2 2m m 2m
pffiffiffiffiffiffiffi
where i ¼ 1 is the imaginary unit.
For this case, it is convenient to make use of Euler’s equations which relate exponential and
trigonometric functions, namely,

eix ¼ cos x þ i sin x

ð2:14Þ
eix ¼ cos x  i sin x

The substitution of the roots p1 and p2 from Eq. (2.13) into Eq. (2.4) together with the use of Eq. (2.14)
gives the following convenient form for the general solution of the underdamped system:

uðtÞ ¼ eðc=2mÞt ðA cos ωD t þ B sin ωD tÞ ð2:15Þ

where A and B are redefined constants of integration and ωD, the damping frequency of the system, is
given by
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
k  c 2
ωD ¼  ð2:16Þ
m 2m
or
qffiffiffiffiffiffiffiffiffiffiffiffiffi
ωD ¼ ω 1  ξ2 ð2:17Þ

This last result is obtained after substituting, in Eq. (2.16), the expression for the undamped natural
frequency
rffiffiffiffi
k
ω¼ ð2:18Þ
m
and defining the damping ratio of the system as
c
ξ¼ ð2:19Þ
ccr

where the critical camping coefficient ccr is given by Eq. (2.6).

Finally, when the initial conditions of displacement and velocity, u0 and υ0, are introduced, the
constants of integration can be evaluated and substituted into Eq. (2.15), giving
 
υ0 þ u0 ξω
uðtÞ ¼ eξωt u0 cos ωD t þ sin ωD t ð2:20Þ
ωD

Alternatively, this expression can be written as

uðtÞ ¼ Ceξωt cos ðωD t  αÞ ð2:21Þ

2.6 Logarithmic Decrement 33

where
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ðυ0 þ u0 ξωÞ2
C ¼ u20 þ ð2:22Þ
ω2D
and

ðυ0 þ u0 ξωÞ
tan α ¼ ð2:23Þ
ω D u0

A graphical record of the response of an underdamped system with initial displacement u0 but starting
with zero velocity (υ0 ¼ 0) is shown in Fig. 2.3. It may be seen in this figure that the motion is
oscillatory, but not periodic. The amplitude of vibration is not constant during the motion but
decreases for successive cycles; nevertheless, the oscillations occur at equal intervals of time. This
time interval is designated as the damped period of vibration and is given from Eq. (2.17) by
2π 2π
TD ¼ ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffi ð2:24Þ
ωD ω 1  ξ 2

Fig. 2.3 Free-vibration response for undamped system

The value of the damping coefficient for real structures is much less than the critical damping
coefficient and usually ranges between 2 and 10% of the critical damping value. Substituting for the
extreme value ξ ¼ 0.10 into Eq. (2.17),
ωD ¼ 0:995ω ð2:25Þ

It can be seen that the frequency of vibration for a system with as much as a 10% damping ratio is
essentially equal to the undamped natural frequency. Thus in practice, the natural frequency for a
damped system may be taken to be equal to the undamped natural frequency.

2.6 Logarithmic Decrement

A practical method for determining experimentally the damping coefficient of a system is to initiate
free vibration, obtain a record of the oscillatory motion, such as the one shown in Fig. 2.4, and
measure the rate of decay of the amplitude of motion. The decay may be conveniently expressed by
the logarithmic decrement δ which is defined as the natural logarithm of the ratio of any two
successive peak amplitudes, u1 and u2, in the free vibration, that is
34 2 Damped Single Degree-of-Freedom System

u1
δ ¼ ln ð2:26Þ
u2
The evaluation of damping from the logarithmic decrement follows. Consider the damped vibration
motion represented graphically in Fig. 2.4 and given analytically by Eq. (2.21) as

uðtÞ ¼ Ceξωt cos ðωD t  αÞ ð(2.21) repeatedÞ

Fig. 2.4 Curve showing peak displacements and displacements at the points of tangency

We note from this equation that when the cosine factor is unity, the displacement is on points of the
exponential curve u(t) ¼ Ce–ξωt as shown in Fig. 2.4. However, these points are near but not equal to
the positions of maximum displacement. The points on the exponential curve appear slightly to the
right of the points of peak or maximum amplitude. For most practical problems, the discrepancy is
negligible and the displacement curve at points where the cosine is equal to one may be assumed to
coincide at the peak amplitude with the curve u(t) ¼ Ce–ξωt, so that we may write for two consecutive
peaks u1 at time t1 and u2 at time TD seconds later.

u1 ¼ Ceξωt1

and

u2 ¼ Ceξωðt1 þT D Þ

Dividing these two peak amplitudes and taking the natural logarithm, we obtain
u1
δ ¼ ln ¼ ξωT D ð2:27Þ
u2

or by substituting TD, the damped period, from Eq. (2.24),

2πξ
δ ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffi ð2:28Þ
1  ξ2

As we can see, the damping ratio ξ can be calculated from Eq. (2.28) after determining experimentally
the amplitudes of two successive peaks of the system in free vibration. For small values of the
damping ratio, Eq. (2.28) can be approximated by
δ ¼ 2πξ ð2:29Þ
2.6 Logarithmic Decrement 35

Alternatively, the logarithmic decrement may be calculated as the ratio of two consecutive peak
accelerations, which are easier to measure experimentally than displacements. In this case, taking the
first and the second derivatives in Eq. (2.21), we obtain

u_ ðtÞ ¼ Ceξωt ½ξω cos ðωD t  αÞ  ωD sin ðωD t  αÞ

€uðtÞ ¼ Ceξωt ½ξω cos ðωD t  αÞ  ωD sin ðωD t  αÞðξωÞ
þ ξωωD sin ðωD t  αÞ  ω2D cos ðωD t  αÞ

At a time t1, when cos(ωDt1 – α) ¼ 1 and sin(ωDt1 – α) ¼ 0

u1 ¼ Ceξωt1 ξ2 ω2  ω2D
€

and at time t2 ¼ t1 + TD, corresponding to a period later, when again the cosine function is equal to one
and the sine function is equal to zero,

€u2 ¼ Ceξωðt1 þT D Þ ξ2 ω2  ω2D

The ratio of the acceleration at times t1 and t2 is then

€u1
¼ eξωT D
€u2

and taking natural logarithmic results in the logarithmic decrement in terms of the accelerations as
€u1
δ ¼ ln ¼ ξωT D
€u2

which is identical to the expression for the logarithmic decrement given by Eq. (2.27) in terms of
displacement.

Illustrative Example 2.1

A vibrating system consisting of a weight of W ¼ 10 lb and a spring with stiffness k ¼ 20 lb/in is
viscously damped so that the ratio of two consecutive amplitudes is 1.00 to 0.85. Determine:

1. The natural frequency of the undamped system.

2. The logarithmic decrement.
3. The damping ratio.
4. The damping coefficient.
5. The damped natural frequency.

Solution:

1. The undamped natural frequency of the system in radians per second is

rffiffiffiffi s ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
k 386 in=sec 2
ω¼ ¼ 20 lb=in  Þ ¼ 27:78 rad=sec
m 10 lb

or in cycles per second

36 2 Damped Single Degree-of-Freedom System

ω
f ¼ ¼ 4:42 cps
2π
2. The logarithmic decrement is given by Eq. (2.26) as
u1 1:00
δ ¼ ln ¼ ln ¼ 0:163
u2 0:85
3. The damping ratio from Eq. (2.29) is approximately equal to
δ 0:163
ξ¼ ¼ ¼ 0:026
2π 2π
4. The damping coefficient is obtained from Eqs. (2.6 and 2.19) as
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi lb: sec
c ¼ ξccr ¼ 2  0:026 ð10  20Þ=386 ¼ 0:037
in
5. The natural frequency of the damped system is given by Eq. (2.17), so that
pffiffiffiffiffiffiffiffiffiffiffiffiffi
ωD ¼ ω 1  ξ 2
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ωD ¼ 27:78 1  ð0:026Þ2 ¼ 27:77 rad=sec

Ilustrative Example 2.2

A platform of weight W ¼ 4000 lb is being supported by four equal columns that are clamped to the
foundation as well as to the platform. Experimentally it has been determined that a static force of
F ¼ 1000 lb appliled horizontally to the platform produces a displacement of Δ ¼ 0.10 in. It is
estimated that damping in the structures is of the order of 5% of the critical damping. Determine for
this structure the following:

1. Undamped natural frequency.

2. Absolute damping coefficient
3. Logarithmic decrement.
4. The number of cycles and the time required for the amplitude of motion to be reduced from an
initial value of 0.1 to 0.01 in.

Solution:

1. The stiffness coefficient (force per unit displacement) is computed as

F 1000
k¼ ¼ ¼ 10, 000 lb=in
Δ 0:1
and the undamped natural frequency
rffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
kg 10, 000  386
ω¼ ¼ ¼ 31:06 rad=sec
W 4000
2.6 Logarithmic Decrement 37

2. The critical damping is

pffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi lb: sec
ccr ¼ 2 km ¼ 2 10, 000  4000=386 ¼ 643:8
in
and the absolute damping
lb: sec
c ¼ ξccr ¼ 0:05  643:8 ¼ 32:19
in
3. Approximately, the logarithmic decrement is
 
u0
δ ¼ ln ¼ 2πξ ¼ 2π ð0:05Þ ¼ 0:314
u1

and the ratio of two consecutive amplitudes

u0
¼ 1:37
u1
4. The ratio between the first amplitude u0 and the amplitude uk after k cycles may be expressed as
u0 u0 u1 uk1
¼  
uk u1 u2 uk

Then taking the natural logarithm, we obtain

u0
ln ¼ δ þ δ þ . . . þ δ ¼ kδ
uk
0:1
ln ¼ 0:314k
0:01
ln 10
k¼ ¼ 7:73 ! 8 cycles
0:314
The damped frequency ωD is given by
qffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ωD ¼ ω 1  ξ ¼ 31:06 1  ð0:05Þ2 ¼ 31:02 rad=sec
2

and the damped period TD by

2π 2π
TD ¼ ¼ ¼ 0:2025 sec
ωD 31:02
Then the time for eight cycles is

tð8 cyclesÞ ¼ 8T D ¼ 1:62 sec

Illustrative Example 2.3

A machine weighing 1000 lb is mounted through springs having a total stiffness k ¼ 2000 lb/in to a
simple supported beam as shown in Fig. 2.5a. Determine using the analytical model shown in
Fig. 2.5b the equivalent mass mE, the equivalent spring constant kE, and the equivalent damping
coefficient cE for the system assumed to have 10% of the critical damping. Neglect the mass of the
beam.
38 2 Damped Single Degree-of-Freedom System

Fig. 2.5 (a) System for Ilustrative Example 2.3. (b) Analytical model

Solution:
The spring constant kb for a uniform simple supported beam is obtained from the deflection δ resulting
for a force P applied at the center of the beam:

PL3
δ¼
48EI
Hence,
P 48EI
kb ¼ ¼ 3
δ L
48  107
¼ ¼ 7500 lb=in
403
The equivalent spring constant is then calculated using Eq. (1.5) for two springs in a series:
1 1 1
¼ þ
kE k kb
1 1 ðAnsÞ
¼ þ
2000 7500
kE ¼ 1579 lb=in

The equivalent mass is:

 
W 1000 lb: sec 2
mE ¼ ¼ ¼ 2:59
g 386 in

The critical damping is calculated from Eq. (2.6):

pffiffiffiffiffiffiffiffiffiffiffi
ccr ¼ 2 kE mE
 
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi lb: sec ðAnsÞ
¼ 2 1579  2:59 ¼ 127:92
in

The damping coefficient cE is then calculated from Eq. (2.19):

cE
ξ¼
ccr  
lb: sec ðAnsÞ
cE ¼ ξccr ¼ 0:10  127:92 ¼ 12:79
in
2.7 Response of SDF Using MATLAB Program 39

2.7 Response of SDF Using MATLAB Program

Using finding values in illustrative Example 2.1, plot the responses with following damping ratios:
0.01, 0.03, 0.5, and 0.9. The initial conditions are the displacement of 1 in. and the velocity of 0 in./
sec.
40 2 Damped Single Degree-of-Freedom System

Fig. 2.6 Responses of damped SDF with varied damping ratios damping ratios: (a) 0.01, (b) 0.03, (c) 0.5, (d) 0.9

2.8 Summary

Real structures dissipate energy while undergoing vibratory motion. The most common and practical
method for considering this dissipation of energy is to assume that it is due to viscous damping forces.
These forces are assumed to be proportional to the magnitude of the velocity but acting in the
direction opposite to the motion. The factor of proportionality is called the viscous damping coeffi-
cient. It is expedient to express this coefficient as a fraction of the critical damping in the system
(ξ ¼ c/ccr). The critical damping may be defined as the least value of the damping coefficient for
which the system will not oscillate when disturbed initially, but it will simply return to the equilib-
rium position.
The differential equation of motion for the free vibration of a damped single degree-of-freedom
system is given by
m€u þ cu_ þ ku ¼ 0 ðEq. (2.1) repeatedÞ

The analytical expression for the solution of this equation depends on the magnitude of the damping
ratio. Three cases are possible:

1. Critically damped system (ξ ¼ 1).

2. Overdamped system (ξ > 1).
3. Underdamped system (ξ < 1).

For the underdamped system (ξ < 1), the solution of the differential equation of motion may be
written as
2.9 Problems 41


ξωt v0 þ u0 ξω
uðtÞ ¼ e u0 cos ωD t þ sin ωD t ðEq. (2.20) repeatedÞ
ωD

in which
pffiffiffiffiffiffiffiffiffi
ω¼ k=m is the undamped frequency
pffiffiffiffiffiffiffiffiffiffiffiffiffi
ωD ¼ ω 1  ξ2 is the damped frequency
ξ ¼ cccr is the damping ratio
pffiffiffiffiffiffi
ccr ¼ 2 km is the critical damping

and u0 and v0 are, respectively, the initial displacement and velocity.

A common method of determining the damping present in a system is to evaluate experimentally
the logarithmic decrement, which is defined as the natural logarithm of the ratio of two consecutive
peaks for the displacement or acceleration, in free vibration, that is,
u1 €u1
δ ¼ ln or δ ¼ ln ðEq. (2.26) repeatedÞ
u2 €u2

The damping ratio in structural systems is usually less than 10% of the critical damping (ξ < 0.1).
For such systems, the damped frequency is approximately equal to the undamped frequency.

2.9 Problems

Problem 2.1
Repeat Problem 1.2 assuming that the system has 15% of critical damping.

Problem 2.2
Repeat Problem 1.6 assuming that the system has 10% of critical damping.

Problem 2.3
The amplitude of vibration of the system shown in Fig. P2.3 is observed to decrease 5% on each
consecutive cycle of motion. Determine the damping coefficient c of the system k ¼ 200 lb/in and
m ¼ 10 lb.sec2/in.

Fig. P2.3
42 2 Damped Single Degree-of-Freedom System

Problem 2.4
It is observed experimentally that the amplitude of free vibration of a certain structure, modeled as a
single degree-of-freedom system, decreases in 10 cycles from 1 in to 0.4 in. What is the percentage of
critical damping?

Problem 2.5
Show that the displacement for critical and overcritical damped systems with initial displacement u0
and velocity v0 may be written as

u ¼ eωt ½u0 ð1 þ ωtÞ þ v0 t for ξ ¼ 1


0 v0 þ u0 ξω 0
u ¼ eξωt u0 coshωD t þ 0 sinhωD t for ξ > 1
ωD
qffiffiffiffiffiffiffiffiffiffiffiffiffi
0
where ωD ¼ ω ξ 2  1.

a) Write the MATLAB coding for these two cases.

b) Run the MATLAB file with the same cases in Problem 2.1 with damping ratios of 1 and 2.

Problem 2.6
A structure is modeled as a damped oscillator having a spring constant k ¼ 30 kip/in and undamped
natural frequency ω ¼ 25 rad/sec. Experimentally it was found that a force of 1 kip produced a
relative velocity of 1.0 in/sec in the damping element. Determine:

a) The damping ratio ξ.

b) The damped period TD.
c) The logarithmic decrement δ.
d) The ratio between two consecutive amplitudes.

Problem 2.7
In Fig. 2.4 it is indicated that the tangent points to the displacement curve corresponds to
cos(ωDt – α) ¼ 1. Therefore the difference in ωDt between any two consecutive tangent points is
2π. Show that the difference in ωDt between any two consecutive peaks is also 2π.

Problem 2.8
Show that for a damped system in free vibration the logarithmic decrement may be written as
1 ui
δ¼ ln
k uiþk

where k is the number of cycles separating the two measured peak amplitudes ui and ui+k.

Problem 2.9
It has been estimated that damping in the system of Problem 1.11 is 10% of the critical value.
Determine the damped frequency fD of the system and the absolute value of the damped coefficient c.
2.9 Problems 43

Problem 2.10
A single degree-of-freedom system consists of a mass with a weight of 386 lb and a spring of stiffness
k ¼ 3000 lb/in. By testing the system it was found that a force of 100 lb produces a relative velocity of
12 in/sec. Find:

a) The damping ratio, ξ.

b) The damped frequency of vibration, fD.
c) Logarithmic decrement, δ.
d) The ratio of two consecutive amplitudes.

Problem 2.11
Solve Problem 2.10 when the damping coefficient is c ¼ 2 lb.sec/in.

Problem 2.12
For each of the systems considered in Problem 1.15, determine the equivalent spring constant kE and
the equivalent damping coefficient cE in the analytical model shown in Fig. P2.12. Assume that the
damping in these systems is equal to 10% of critical damping.

Fig. P2.12

Problem 2.13
A vibration generator with two weights each of 30 lb with an eccentricity of 10 in rotating about
vertical axis in opposite directions is mounted on the roof of a one – story builsing with a roof that
weighs 300 kips. It is observed that the maximum lateral acceleration of 0.05 g occurs when the
vibrator generator is rotating at 400 rpm. Determine the equivalent viscous damping in the structure.

Problem 2.14
A system modeled by two freely vibrating masses m1 and m2 is interconnected by a spring and a
damper element as shown in Fig. P2.14. Determine for this system the differential equation of motion
in terms of relative motion of the masses ur ¼ u2 – u1.

Fig. P2.14
44 2 Damped Single Degree-of-Freedom System

Problem 2.15
Determine the relative motion ur ¼ u2 – u1 for the system shown in Fig. P2.14 in terms of the natural
frequency ω, damped frequency, ωD and relative damping, ξ. Hint: Define the equivalent mass as
M ¼ m1 m2 /(m1 + m2).