CHAPTER 3

Forced Vibration of Single Degree

of Freedom Systems

Delbante Alebachew
3. Forced Vibration of Single Degree of
Freedom Systems
Introduction
• Free Vibration: Occur due to an energy source which
is removed while vibration occur.
• Forced Vibration:- Occurs when work is being done
on a system while vibration occur.
For example
– Elastic structure vibrating due to earthquake
– rotating components constantly excited by harmonic
force of the driving motor
– reciprocating piston of an engine transmits a
sinusoidally varying force to adjacent components
• Periodic input/excitation results in a periodic response
• It could be sinusoidal driving functions at a single frequency
• f(t) = f(t+T) for all values of t.
• Example f (t )  Asin t

f(t) T f(t) is a general periodic

function of period T

t
 review
• 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
an   f (t ) cos nT tdt
T
average of the 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,....
 review

Simplification:
• Orthogonality:
– Integral of the products of two functions is zero.
0 m  n
0 sin nT t sin mT t dt  T m  n
T

 2
0 m  n
0 cos nT t cos mT t dt  T m  n
T
and
 2


T
cos nT t sin mT t dt  0
0

Where m & n areintegers.

 review

• 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

 3.1. Harmonic Excitation
k
m F(t)=Focosωdrt

c
Frictionless

 mx  cx  kx  F (t )  Fo cos dr t
where
• Fo is the maximum amplitude of excitation and
• ωdr is the driving (forcing) frequency.

• For linear forced systems the total response is obtained by

superposition, i.e.

Response of Response of External

Total response of = Forces
homogenous +
a system
system (Non-homogenous)
3.1.1 Undamped System (C = 0)
mx(t )  kx(t )  F (t )  F0 cos dr t
• F(t) has a form of a sine or cosine (here we take cosine)
x(t )  n2 x(t )  fo cos dr t , fo  F0 ..............(1)
m
• This is a linear non-homogenous ODE (using method of
undetermined coefficient)

 xh(t) homogeneous solution

 xp(t) is the particular solution

 xp(t) = A0 cosωdrt if F(t) which has cosine form

 xp(t) = A0 sinωdrt if F(t) which has sine form.
Substituting x p (t ) and x p (t ) in the differential equation
f0
and solving: Ao  2

n   dr2 
f0
 x p (t )  2 cos dr t

n   dr
2

provided that the two frequencies are not equal n  dr
x(t )  xh (t )  x p (t )
xh (t )  A1 sin nt  A2 cos nt
Having the initial conditions:
x(t  0)  xo and x(t  0)  vo
 v0 f0
A1  and A2  xo  2
n 
n   dr2 
v0  f0 
 xh (t )  sin nt   xo  2  cos nt
n     2
 n dr 
The total response will then be:
v0  f0  f0
x(t )  sin nt   xo  2  cos nt  2 cos dr t
n     2
n   dr 2
 n dr 
............................(*)
x(t)

t
• Two very important phenomenon occur when the driving frequency
(ωdr) becomes close to the systems natural frequency (ωn)

Case i. when ωn - ωdr becomes very small and x0=0 & v0=0

f0
2 
x(t )  2 cos dr t  cos nt 
n   dr

• Using a simple trigonometric identities

2 f0  n  dr   n  dr 
x(t )  2 sin  t  sin  t
n  dr
2
 2   2 
 but n  dr  n  dr
 n  dr 
 sin  t  oscillates with a much longer period,
 2 
4  n  dr 
T than the term sin  t
n  dr  2  sin  n  dr t 
x(t)  
 2 

   dr
t

sin  n t
 2 
x(t)

The resulting motion is a rapid oscillation with slowly varying amplitude and
is called a BEAT.
 Case ii. When ωn = ωdr
• In this case the choice of the function,
xp=A0cosωdrt, fails because it is also a solution of
the homogenous equation.

 xp= tA0sinωdrt
• Inserting in equation (1) and solving for A0, then:

fo
Ao 
2dr
fo
 x p (t )  t sin dr t
2dr
• The total response will then be:
x(t )  xh (t )  x p (t )
fo
 x(t )  A1 sin nt  A2 cos nt  t sin dr t
2dr
Using the initial conditions:
x(t  0)  xo and x(t  0)  vo
vo
 A1  and A2  xo
n
vo fo
 x(t )  sin nt  xo cos nt  t sin dr t
n 2dr
This is a harmonic function with amplitude which grows
unboundedly.
• This defines the phenomenon of Resonance.
• Resonance causes our system to fail & break.

x(t)

• This phenomenon of Resonance also makes the

system to vibrate with large amplitudes that it will
no more be treated with linear relationships.
• The system will behave non-linear and finally
fails.
3.1.2 Damped System
mx(t )  cx(t )  kx(t )  F0 cos dr t
x(t )  2n x(t )  n2 x(t )  f 0 cos dr t.................(3)
• By the method of undetermined coefficients and
expecting a phase shift of the response because
of the effect of the damping force:
x p (t )  Ao cos(dr t   )
This can be written as:
x p (t )  As cos dr t  Bs sin dr t
 Bs 
From which: Ao  A  B 2
s
2
s and  =tan  -1

 As 
 x p (t )   Asdr sin dr t  Bsdr cos dr t
x p (t )  dr2  As cos dr t  Bs sin dr t 
Substituting x p (t ), x p (t ) and x p (t ) in equation (3) and
solving for As and Bs :

As 
 2
n  dr2 f0  and
    2  
2
  dr2
2 2
n n dr

2ndr f0
Bs 
    2  
2
  dr
2 2 2
n n dr

f0  1 2ndr

 x p (t )  cos  dr t  tan 
    
    2  
2 2 2
2
n   dr 2
n dr
2
 n dr 
 x(t )  xh (t )  x p (t )
x(t )  Aent sin d t     Ao cos dr t   
 x(t )  xh (t )  x p (t )
x(t )  Aent sin d t     Ao cos dr t   

Note:
• For large values of t, the xh(t) term approaches
zero, hence the x(t) approaches xp(t).
• Thus xp(t) is called the steady-state response and
the xh(t) is called the transient response.
• Let us observe how the system is responding
(specially the particular solution) for a varying
values of
– r  dr (frequency ratio) and
n
– ζ (damping ratio).
 f0 2ndr
i.e. Ao  ,   tan 1

n2   dr2
      2ndr 
2 2
2 2
n dr

dr 2 r
r  Ao 
f0
,  tan 1
n 1 r2
 2
n 1  r    2 r 
2 2 2

Hence the normalized amplitude of the steady-state response

will then be:
Ao k
Aon 2
1 m A
   o
f0
 2 2

1  r   2 r 
2 F0
m
F0
k
Ao 1
M  ............(4)
 st

1  r 2   2 r  
2 2

where  st is the deflection of the spring under the static force F0 .

deflection
The plot of the normalized amplitude and the phase lag are given in
the following two figures
• Note that as ωdr→ ωn (r → 1)
– the normalized amplitude approaches a maximum
value for a system with very small (light) damping
(ζ< 0.1) and
– the phase shift crosses through 900. This defines
Resonance for the undamped case.
fo Aon2
• As ωdr→0, A0→ 2
n fo
1 and
• As ωdr becomes very large the amplitude
approaches zero asymptotically.
• Note: Resonance does not exactly correspond
with the value of ωdr at which the peak value of
the steady-state response occurs.
• To find the value of r (frequency ratio) at which
the normalized amplitude (magnification factor)
M, attain maximum value:
 
dM d  A0k  d  1 
     0
dr dr  F0  dr 

 1  r 2 2
  2 r 
2 


 rpeak  1  2 2  1
Substituting this in (4):
1
 M max  ...........................(5)
2 1   2
• Note: The rpeak value illustrates that the
maximum value of M occurs at:
1
i) r  1  2  1 if 0   
2
and
2
1
ii) r  0 if  
2  1 

• For the small damping case,  
2 , the value of the
driving frequency corresponding to the maximum
value of M is called the peak frequency, ωp.
1
 p  n 1  2  1 for 0   
2
2
Note: When   0   p  n , i.e. the usual undamped
resonance condition.
• The undamped resonance condition (ωdr= ωn) also works for the lightly
damped system.
3.2. Base Excitation

Examples:
– Automobile excited by rough road surface
– Engine excited by the frame work supporting it
– Machinery excited by the surface on which the
machinery is mounted
– Airplane engine excited by wing or tail supporting
engine
 x(t)
Model: m

k c Suspension system (Rubber mountings)

y(t)
Base (Foundation)
x(t)
FBD: m

k(x-y) c( x  y)

• The equation of motion will then be:

mx  c(x  y)  k (x  y)  0........................(6)
• If we assume the base is excited harmonically,
- That is: y(t )  Y sin(b t )  y(t )  Yb cos(b t )
where b is the frequency of base oscillation
• Two important transmissibility ratios to be studied
– Displacement transmitted to the mass
– Force transmitted to the mass

a) Displacement transmitted to the mass

mx  cx  kx  cy  ky
 mx  cx  kx  cY b cos(b t )  kY sin(b t )
• It is an equation of motion of mass-spring-damper
system with two harmonic excitations.
• Hence, the equation will have one homogenous
(xh(t)) and two particular (xp1(t) and xp2(t) ) solutions,
one for each component of the excitation.
 x  2n x  n2 x  2nb Y cos(b t )  n2Y sin(b t )

(1)
Y (t ) Y (2) (t )

 fo(1)  2nb Y for Y (1) (t )

2nb Y
 x(1)
p = cos(bt  1 )
 n2 2 2
 b    2nb 
2

 2n 
where 1 = tan 1  2 b 
 n   2 
 b 
 and fo(2)  n2Y for Y (2) (t )
n2Y
 x(2)
p = sin(bt  2 )
 n2  b 
2 2
  2nb 
2

where 2  1
i.e. independent of the amplitude of excitation
 x p =x(1)
p +x (2)
p
1/ 2
 
n   2b 
2 2
 
 x p (t )  nY  cos(bt  1  3 )
2
 2 2

 n  b   2nb  

2

1  n 
where 3  tan  
 2b 
• The magnitude of the particular solution, xp(t) will
then be:
1/ 2
 1  2 r 
 
2
b
X Y   where r 
 
 1  r 2   2 r 2  n
2

 
1   2 r 
2
X
 
    2 r 
2
1 r2
2
Y

• This is the ratio of the maximum response

magnitude to the input displacement magnitude and
is known as the Displacement Transmissibility.
• It shows how motion is transmitted from the base to
the mass as a function of the frequency ratio, r and
the damping ratio ζ.
 X
ζ increasing
Y
ζ decreasing
1

1 2 r
Note:
X
i. For r  2,  1 , i.e.
Y
 the motion of the mass is an amplification of the
motion of the base
 Large ζ yields smaller transmissibility ratios.
X
ii. For r  2, 1 , i.e.
Y
 the motion of the mass is smaller in amplitude than
that of the base for all ζ
 but it increases for increasing ζ.
 b) Force transmitted to the mass
• Force is transmitted to the mass through the
spring and damper
i.e. F (t)  k (x  y)  c(x  y)  mx(t )............(7)
• For the steady-state case (the transient solution
does not stay long and hence is ignored)
1/ 2
 
n   2b 
2 2
 
x(t )  nY  cos(bt  1  3 )
2
 
2 2
 n  b   2nb  

2


• Differentiating x(t) twice and substituting in equation (7)

 FT ζ=0.01
kY ζ increasing
ζ=0.1
ζ increasing
ζ=0.2 ζ=1
2

1 2 r

• Unlike the displacement transmissibility, the force

transmitted does not necessarily fall off for r  2 . In
fact, as the damping increases, the force transmitted
increases dramatically for r  2 .
3.2 Rotating Unbalanced Mass
• It is due to small irregularities in the distribution
of a rotating mass
• Let the frequency of rotation of machine be ωr.
mo
e x x(t) mo
o
m y(t) e
θ=ωr t
k c Rubber floor mountings
as a spring and damper

• The x- component of motion of the mass mo is

given by:

xr  e sin r t, xr  er cos rt, xr  er2 sin rt

• And the reaction force, Fr , generated by the
rotating mass mo has a component in the x-
direction
Fr = mo xr  emor2 sin r t
• This force acts on the machine mass, m.
– Note: The force in the horizontal direction are
canceled by the guides.
• Considering the machine and the unbalanced
mass as separate bodies the equation of motion
of the entire system will be:
d2
(m  mo ) x  mo 2  x  e sin r t   kx  cx
dt
 d2
(m  mo ) x  mo 2  x  e sin r t   kx  cx
dt
 mx  mo x  mo x  moer2 sin r t  kx  cx
 mx  cx  kx  moer2 sin r t
mo 2
 x  2n x  n x 
2
er sin r t
m
 x p (t )  X sin(r t   )
moe r2 moe
where X    (r ,  )
m (1  r 2 )2  (2 r )2 m

1 2 r 
  tan  2
 1 r 
 mX
mo e ζ=0.1

ζ increasing
Normalized

ζ=0.25
amplitude

1
ζ=0.707

ζ=1

1 r
1 1
rpeak  for 0   
(1  2 2 ) 2
1
max 
2 (1   2 )
Note:
• The maximum deflection is less than or equal to 1
1
for any system with   . This indicates that the
2
increase in amplification of the amplitude caused by
the unbalance can be eliminated by increasing the
damping, which is not always practical
• The magnitude of the dimensionless displacement
approaches unity if r is large. Hence if the running
frequency ωr is such that r >> 1, the effect of the
unbalance is limited. For large values of r, all the
magnitude curves for each value of ζ approaches
unity, so that the choice of damping coefficient (ζ) for
large r is not important.