Mechanical Vibrations

Lecture Slides

Addis Ababa Science and Technology University

College of Engineering

Electromechanical Engineering

Delbante Alebachew
 1. Introduction
1.1. Definition:
Vibration is a dynamic behavior of physical system in
which a system oscillates about a certain equilibrium
position.

A vibratory system, in general, includes

• a means for storing potential energy (spring or elasticity),
• a means for storing kinetic energy (mass or inertia), and
2
• a means by which energy is gradually lost (damper).
Number of Degrees of Freedom (DoF): The minimum
number of independent coordinates required to
determine completely the positions of all parts of a
system at any instant of time defines the number of
degrees of freedom of the system.

3
1.2 Classification of Vibration
Vibration can be classified in several ways. Some of the important
classifications are as follows
• Free and Forced Vibration
• Undamped and Damped Vibration
• Linear and Nonlinear Vibration
• Deterministic (Periodic) and Random Vibration

4
• Periodic function: is any function that repeats
itself in time, i.e. any function for which there
exists a fixed time T called the period such that

f(t) = f(t+T) for all values of t.

– Example: forcing function that is the sum of two
sinusoids of different frequencies
f(t) T f(t) is a general periodic
function of period T

5
• Fourier Theory: Any periodic function f(t) with
period T, may be represented by an infinite
series of the form:
a0 
f (t )  +  (an cos n T t + bn sin n T t )
2 n =1
Where T = 2
T
2 T
a0 =  f (t ) dt
T 0
It is twice the
2 T
average of the an =  f (t ) cos nT tdt Fourier
function f(t) over T 0 Coefficients
one cycle
2 T
bn =  f (t ) sin nT tdt
T 0
for n =1, 2,....
6
Simplification:
• Orthogonality:
– Integral of the products of two functions is zero.
T 0 m  n
0 sin nT t sin mT t dt = T m = n
 2
T 0 m  n
0 cos nT t cos mT t dt = T m = n and
 2
T
0
cos nT t sin mT t dt = 0
Where m & n are integers.

7
• Symmetry
– If f(t)=f(-t):
• Symmetry about the origin (y axis)
• Called even function
• The Fourier Series contains only cos components, i.e. bn=0
– If f(t)=-f(-t):
• Called an odd function
• The Fourier Series contains only sin components, i.e. an=0

Home reading ………

8
1.3. Modeling of Dynamic Systems for Vibration
– Machines could be simple or complex
• Algorithm for vibration analysis:

Vibration
Problem

Produce
physical
modeling

Produce Modify the physical

Mathematical model in light of
Modeling the solution

Solve the
Mathematical
Model

Check the
solution with
reality
Yes No
9
Stop
Model:
– Simplified but sufficiently sophisticated
– Depends on an individual perspective
– Depends on the accuracy of analysis needed
Elements of Modeling a mechanical system:

– Mass
– Spring
– Damper Building blocks in the development of
– Rod physical models of dynamic systems
– Beam
– Plate

10
• Physically these elements are of continuous type
(In-Dof) and possess non linear characteristic.
– take simplified assumptions and identify the basic parts
with relevance to the problem
– model by lumped/discretized equivalent (finite No
DoF):
• Mass
• Spring and With linear Characteristics
• Damper
• After physical modeling follows mathematical
Modeling
– Define physical quantities PDE for Continuous Model

– Establish equation of motion ODE for Discretized Models

• Then solution of equation of motion gives the

motion of the dynamic system as function of time.
11
Example ………
 Discrete Truck Model

model consists of seven masses with ten spring-damper-systems.

12
1.3. Characteristics of Discrete System Components
• Essential properties of a machine or part to
experiences vibration are:
– Inertia:
✓ associated with mass moving in translation or rotation (m or J)
– Elasticity:
✓ Capability to deform and return to original form (deformation
before plastic limit)
✓ Symbolized by spring
✓ Accumulation of elastic potential energy
– Damping
✓ Represents energy dissipation in system
✓ Symbolized by a dashpot (damper)

13
In real mechanical system
• Inertia & elasticity are distributed
– Examples
i. Distributed mass & tortional stiffness of crank shaft

Actual

c1 c2 c3 c4 c5 c6
Mechanical
Model
k1 J k2 J k3 J k4 J k5 J k6
1 2 3 4 5

ii. Distributed mass and bending stiffness of a bridge

Actual Mechanical
Model

14
Note:
• Distributed properties are represented by
equivalent lumped masses interconnected to
one another through lumped damping and
stiffness elements.
– This gives a discrete vibration model or discrete system.
• The lumped elements should posses the same
properties like the continuous element, which is
satisfied through the equivalence in kinetic,
potential & dissipation energies.

15
Discrete System Components
• The discrete mechanical system elements are:
1. Spring
• relates force to displacement
• Possess the property of elasticity
• Generally assumed to be massless
Fs stiffening Spring Fs Fs Linear Spring
(∆F>> ∆x)
Softening Spring
x1 x2
Fs=k(x2-x1)
(∆F<< ∆x)

∆x=x2-x1 kt
φ
Ms=kt φ Torsional Spring
Linear Range, Slope of k,
Spring Constant or stiffness Ms

• A spring operated in the linear range is

characterized by its stiffness, k [N/m]. 16
2. Damper
• relates force to velocity
• consists of a piston fitted loosely in a cylinder filled with
viscous fluid
• represented by viscous damper or dash pot
Fd
• assumed to be massless
Slope = C
Fd Fd
x1 x2 ( x2 − x1 )
Fd = C ( x2 − x1 )
• Fd is the damping force and resists an increase in
relative velocity.
• C [Ns/m] is the coefficient of viscous damping

17
3. Mass
• Relates force to acceleration
Fm
x
Slope = m
m Fm

Axial Vibration Torsional Vibration

Spring, k [N/m] Torsional Spring, kt [Nm]
Viscous damper, C [Ns/m] Torsional Viscous Damper, Ct [Nms]
Mass, m [kg] Mass moment of inertia, J [kg.m2]
Fs , Fd Ms , Md
x, x, x , , 18
End of chapter One

19
2. Differential equation of motion for 1st Order and
2nd Order Linear systems
→ Case i) System with m=0
F (t ) − Cx − kx = 0
 Cx + kx = F (t )

• This is a 1st order linear ODE with constant coefficients.

→ Case ii) Spring-Damper-Mass system
k FBD
Fs= kx
m F(t) m F(t)
Fd = cx
c x(t)
 F (t ) − Cx − kx = mx
2nd Order Linearized equation of
 mx + Cx + kx = F (t ) motion about equilibrium
• This is a 2nd order linear ODE with constant coefficients. 20
→ Case iii) Suspended Spring-Mass-Damper
Undeformed
F(t) position
FBD 1 F(t)
y(t)
y(t)
x(t)
m

k C Equilibrium Ky(t) Cy (t )
position mg
my (t ) + cy (t ) + ky (t ) + mg = F (t )
If y (t ) = x(t ) − xst , where xst is the static deformation
 y (t ) = x(t ) & y (t ) = x(t )
mx + cx + kx = F (t )
Since kxst will be cancelled by mg.

21
2.1. Force Free Response of 1st Order System

• The equation of motion is given by:

cx(t ) + kx(t ) = 0 [ F (t ) = 0].................(*)
Let x(t ) = Ae st  x(t ) = Ase st
Substituting in (*)
 cs Ae st + k Ae st = 0
k
s=−
c
k t
− t −
 x(t ) = Ae = Ae 
c

c
where  = and is called the Time Constant. 22
k
• The constant of integration A depends on the initial
conditions xo and vo.
Let x(t = 0) = xo
 xo = Ae0  A = xo
t
−

 x (t ) = xo e t>0
0 t<0
• This shows us that response decays
exponentially with time.
x(t)
xo The system returns to equilibrium without any
oscillation

t
23
 c with bigger  → slow return
=
k with smaller  → fast return

Hence, τ is the measure of the speed of return of the

system to equilibrium. That is, for a stiff spring, or for
light damping, the speed rate of return is fast and
vice versa.
x(t)

xo
Smaller τ, stiff or less
damped system
bigger τ, less stiff or more
damped system

t 24
2.2. Free Response of Single Degree of Freedom 2nd
Order Linear Systems (Harmonic Oscillations;
undamped C = 0)
• The equation of motion is given by:
mx + kx = 0 [F (t ) = 0 and c = 0]
k
 x+ x =0
m
 x + n2 x = 0
k
where n = (Undamped Natural Frequency)
m
Let x = Ae st  x = As 2 e st
 s 2 Ae st + n2 Ae st = 0
 s = in 28
• General Solution:
x(t ) = A1e s1t + A2e s2t
= A1eint + A2e − int
Note: ei = cos  + isin & e − i = cos  − i sin 
 x(t ) = A1  cos nt + i sin nt  + A2  cos nt − i sin nt 
= ( A1 + A2 ) cos nt + i ( A1 − A2 ) sin nt

• But, A1 and A2 are complex conjugates and hence:

A1 + A2 = C (Real) and i( A1 − A2 ) = D (Real)
D
  x(t ) = C cos nt + D sin nt
C
= C 2 + D 2 ( cos  cos nt + sin  sin nt )
 x(t ) = A cos(nt −  ).........(**) 29
  x(t ) = A cos(nt −  ).........(**)
where A (Amplitude) and  (Phase angle) are the constants
of integration.
• As can be seen from (**), the system executes a simple
harmonic motion with a natural frequency ωn , an
amplitude A and a phase lag  .
• The period for one complete revolution is given by:
2
T=
T
n
A x(t)

 t
( / 2 +  ) n
n
30
• Find the constants of integration A and  in terms of
initial conditions.
Let x(t = 0) = xo and x(t = 0) = vo
x(t ) = A cos(nt −  )
x(t ) = −n A sin(nt −  )
 xo = A cos(− ) and vo = −n A sin(− )
2
 vo   vo 
  = tan  −1
 and A = xo +  
2

 n xo   n 

31
 2.4. Free Vibration of Damped 2nd Order System

mx + cx + kx = F (t ) F (t ) = 0
c k
 x(t ) + x(t ) + x (t ) = 0
m m
x + 2n x + n2 x = 0
• where ζ is the non dimensional parameter called
viscous damping and is given by:
c c
= =
2mn cr
• and ωn is the natural frequency of the undamped
Oscillation and is given by:
k
n = 32
m
• Let x(t) = Aest
 s 2 + 2n s + n2 = 0
−2n  4 2n2 − 4n2
s1,2 =
2

=
( )
− 2   2  2 − 1 n
2

(
s1,2 = −   2 − 1 n )
• This shows that s1 & s2 depends on the viscous damping factor ζ
which it self depends on the elements of the physical system i.e.
k, m & c.
33
 Complex Plane (Root Locus):
Im
ζ=0 iω S Plane
n
in 1 −  2
When  = 0 ωn
 s1,2 =  jn ζ>1 Complex
-ωn -ω ζ Conjugates
 harmonic Oscillation, stable s2 ζ=1
n
s1 Re

−in 1 −  2
When 0    1 -iωn
ζ=0
  2 − 1 is less than 0.

( ) ( )
 s1,2 = −   2 − 1 n = −  i 1 −  2 n
 under damped, marginally stable

When  = 1 When   1
 s1,2 = −n
 a case of double root
(
 s1,2 = −   2 − 1 n )
 As  → , s1 → 0 & s2 → -
 critically damped 34
 Over damped
 x(t ) = A1e s1t + A2e s2t
( − +  −1 ) t
2
( − −  −1 ) t 2

= Ae +Ae
n n

1 2

=e −n t Ae  2 −1 n t
+ A2e −  2 −1 n t 
 1 
Case 1:
– When ζ ≥ 1,
– Aperiodic and exponentially decaying response
– An overdamped case

x(t) ωn increasing x(t) ζ=1

Vo increasing x(t) ζ increasing
(ζ & Vo constant)
(ωn & ζ constant) (ωn & Vo constant)
Very Stiff System

t t t
35
• The case ζ = 1, has a double root, s1= s2= -ωn
 x(t ) = ( A1 + tA2 )e−nt An exponentially
decaying response

• It is also called a critically damped system

Ccr
  =1= , since C = Ccr
2mn
k
 Ccr = 2mn = 2m = 2 km
m
• For a given initial excitation a critically
dampened system tends to approach the
equilibrium position the fastest. (A critically
damped system returns to its equilibrium quicker
than an overdamped system)
36
Case 2:
– When 0< ζ<1 (Under Damped case)

x(t ) = e −n t Ae i 1− 2 n t

+ A2e −i 1− 2 n t 
 1 
= e −nt  A1eid t + A2e − id t 

Where d = n 1 −  2
“Frequency of the damped free vibration”
or “Damped Natural Frequency”

 x(t ) = Ae−nt sin (d t + d )

• This is an oscillatory motion with a constant frequency ωd

and phase angle d but with exponentially decaying
amplitude, Ae −n t where A & d depend on the initial 37
conditions.
  V0 + n x0 
2
 x0d 
A = x0 + 
2
 and d = tan −1

  d   0
V +  n 0 
x

• The curve  Ae −nt provides the enveloping or bounding

curves which envelops the decaying oscillatory
response.
Vo increasing ζ increasing ωn increasing
(ωn & ζ constant) (ωn & Vo constant) (ζ & Vo constant)
x(t) x(t) x(t)
Ae −nt

t t t

− Ae −nt
Practical Applications
1. Recoil mechanisms of guns are designed with
critical damping to allow rapid firing.
2. Automobile suspension systems are often subject to
both free & forced vibrations. Free vibrations occur
when the vehicle is subject to a sudden change in
road contour. In this case the shock absorbers
should have a damping ratio near one. However, if the
vehicle is traveling on a bumpy road, the vehicle is
subject to a possible random excitation. In this case
the system should be under-damped. For these
reasons, vehicle shock absorbers are self-adaptive.

39
2.6. Stability

• In the preceding sections, the physical parameters

m, c, and k are all considered to be positive.
• This allows the treatment of the solutions of the
equation of motion to be classified into three groups:
• Over damped,
• Under damped, or m, c, and k are positive

• Critically damped.
Undamped (with c=0 provides a fourth class)
• These three solutions are all well behaved in the
sense that they do not grow with time and their
amplitudes are finite.
55
• There are many situations, however, in which the
coefficients are not positive, and in these cases the
motion is not well behaved.
• This situation refers to the stability of solutions of a
system.
• Undamped case, consider that the value of k is -ve
and m is +ve.
mx + kx = 0
• The solution will then be of the form:
x(t ) = A sinht + B cosht
which increases with out bound as t does.
• Such solutions are called divergent or unstable.

56
• In the damped case, if c or k is -ve (m positive), the
motion grows with out bound and becomes unstable
as in the undamped case.
• Similar to overdamped solutions and underdamped
solutions, the motion may grow without bound and
not oscillate, or it may grow without bound and
oscillate.
• the motion may be unstable in one or two ways.
• The non oscillatory case is called divergent instability
and the oscillatory case is called flutter instability or
just flutt.
• These types of instability occur in a variety of
situations, called self-excited vibrations and requires
some source of energy.

57
 x(t) x(t)

divergent instability flutter instability or flutt.

Example 1:
• Consider the inverted pendulum connected to two
equal springs as shown.

kl ml
k k 2 sin 
2 θ mg
l/2 fp
p

58
Solution:
• Assume that the springs are undeflected when in
the vertical position and that the mass m of the ball
at the end of the pendulum rod is substantially larger
than the mass of the rod itself, so that the rod is
considered to be massless.
• The equation of motion is then:
 M p = I
 kl 2 
 ml  + 
2
sin   cos  − mgl sin  = 0
 2 
 
• For values of θ less than about π/12 , sinθ and cosθ
can be approximated by sinθ  θ and cosθ  1.

59
  kl 2 
 ml  + 
2
− mgl   = 0
 2 
 
 2ml + ( kl − 2mg ) = 0
Conclusion:
• If k, l, and m are all such that the effective stiffness
is negative, that is, if:
kl - 2mg < 0
the pendulum motion will be unstable by
divergence.

60
Example 2:
• The vibration of an aircraft wing can be crudely
modeled as:
mx + cx + kx =  x
• where:
– m, c and k are the mass, damping, and stiffness
values of the wing modeled as a single-degree-of-
freedom system and
–  x is an approximate model of the aerodynamic
forces on the wing ( >0).
• Rearranging this expression yields:
mx + ( c −  ) x + kx = 0
61
• If γ and c are such that c-γ > 0, the system is
asymptotically stable.
• However, if  is such that c- < 0, then:

 = (c −  ) / 2mn  0
 x(t ) = Ae−nt sin(d t +  )
where − nt  0 for all t  0.
• Such solutions increase exponentially with time.
• It is an example of flutter instability and self-
excited oscillation.
End of chapter two
62