Slide For Vibration

C. R. Yang, NTNU MT C. R.
Yang, NTNU MT
Chapter 3
Harmonically Excited Vibration Learning Objectives
 Find the response of undamped and viscously damped single-DOF
systems subjected to different types of harmonic force, including
base excitation and rotating unbalance
3
 Distinguish between transient, steady-state, and total solutions
 Understand the variations of magnification factor and phase angles
with the frequency of excitation and the phenomena of resonance
and beats
 Find the response of systems involving Coulomb, hysteresis, and
other types of damping
 Identify self-excited problems and investigate their stability aspects
 Derive transfer functions of systems governed by linear differential
equations with constant coefficients
台灣師範大學機電科技學系 -1- 台灣師範大學機電科技學系 -2-
C. R. Yang, NTNU MT C. R. Yang, NTNU MT
Chapter Outline
Learning Objectives
 Solve harmonically excited single-DOF vibration problems using
3.1 Introduction
Laplace transforms
3.2 Equation of Motion
 Derive frequency transfer function from the general transfer function 3.3 Response of an Undamped System Under Harmonic Force
and represent frequency-response characteristics using Bode
3.4 Response of a Damped System Under Harmonic Force
diagrams
3.5 Response of a Damped System Under F (t )  F0eit
 Solve harmonically excited vibration response using MATLAB 3.6 Response of a Damped System Under the Harmonic Motion of
the Base
3.7 Response of a Damped System Under Rotating Unbalance
3.8 Forced Vibration with Coulomb Damping
3.9 Forced Vibration with Hysteresis Damping
3.10 Forced Motion with Other Types of Damping
3.11 Self-Excitation and Stability Analysis

Chapter Outline 3.1

Introduction
3.12 Transfer-Function Approach
3.13 Solutions Using Laplace Transforms
3.14 Frequency Transfer Functions
台灣師範大學機電科技學系 -5-
3.1 台灣師範大學機電科技學系 -6-
3.1 Introduction 3.1 Introduction

• Transient response is defined as the response of a dynamic
• Forcedvibrations occurs whenever external energy is always system to suddenly applied nonperiodic excitations.
supplied to the system during vibration.
• Undera harmonic excitation, the response of the system will also
• Theexternal force can be supplied through either an applied force
or an imposed displacement excitation, which may be harmonic, be harmonic.
nonharmonic but periodic, nonperiodic, or random in nature.
• If
the frequency of excitation coincides with the natural frequency of
• Theresponse of a system to a harmonic excitation is called the system, the response will be very large, which is known as
harmonic response. The harmonic excitations is the form:
resonance and should be avoided.
F (t )  F0e i (t  ) or F (t )  F0 cos(t   ) or F (t )  F0 sin(t   )
• The vibration produced by a unbalanced rotating machine, the
where F0 is the amplitude,  is the frequency, and  is the phase oscillation of a tall chimney (煙囪) due to vortex (渦流) shedding
angle of the harmonic excitation. The value of  depends on the (旋渦脫落) in a steady wind, and the vertical motion of an
value of F(t) at t=0 and is usually taken to be zero.
automobile on a sinusoidal road surface are examples of
harmonically excited vibration.
3.2 3.2 Equation of Motion

Equation of Motion
• Fromthe figure below, the equation
of motion using Newton’s Second
Law of Motion states that
mx  cx  kx  F (t ) (3.1)
3.2
To find the homogeneous solution xh
and particular solution xp, and A spring-mass-damper system
general solution as x(t)=xh+xp
• Thehomogeneous solution is
obtained from the homogeneous As seen in Section 2.6.2, this free
equation (the free vibration) vibration dies out with time under each
of the three possible conditions of
damping (underdamping, critical
mx  cx  kx  0 (3.2) damping, and overdamping) and under
all possible initial conditions
3.2 Equation of Motion 3.3

Response of an Undamped System Under Harmonic
• Thevariations of homogeneous and general solutions with time for
Force
a typical case are shown in the figure below.
(the free-vibration part)
Homogeneous solution: Transient vibration
The rate at which the
transient motion decays
depends on the values of
the system parameters k,
c, and m (the force-vibration part)
3.3
Particular solution: Steady-state vibration
Xh(t) dies out and x(t)

General solution
becomes xp(t) after  time
Homogenous and general solutions of Eq. (3.1) for an underdamped case

3.3 Response of an Undamped System Under 3.3 Response of an Undamped System Under
Harmonic Force Harmonic Force
• Because the exciting force and particular solution is harmonic and
• Consider
an undamped system subjected to a harmonic force. If a
has the same frequency, we can assume a solution in the form:
harmonic force F (t )  F0 cos t acts on the mass m of the system,
代入(3.3)式，等號兩邊做係數比較
x p (t )  X cos t (3.5) 求出X (待定係數法)
mx  kx  F0 cos t (3.3) where X is the Max. amplitude of xp(t)
F0 F0 / k  st
• The homogeneous solution is given by: (mx  cx  kx  0 (3.2)) X   (3.6)
k  m 2 1  m  2  
2
k 1   
 n 
xh (t )  C1 cos nt  C2 sin nt (3.4)
where  st  F0 / k denotes the static deflection
where n  (k / m)1/ 2 is the natural frequency Thus, the total solution of Eq. (3.3) is
F0
x( t )  xh ( t )  x p (t )  C1 cos n t  C 2 sin n t  cos  t (3.7)
k  m 2
• The quantity X /  st is called the magnification factor, amplification
• Using initial conditions x(t  0)  x0 and x(t  0)  x0 factor, or amplitude ratio.
F0 x0
C1  x0  , C2  (3.8)
k  m 2 n • The variation of the amplitude ratio with
Hence the frequency ratio is shown in the figure. Case 1
 F0   x 
x(t )   x0   cos nt   0  sin nt
 k  m 2   n  • Theresponse of the system can be
 F0  identified to be of three types.
 2 
cos t (3.9) Case 3
 k  m  
1. 0  1
The Max. amplitude in Eq. (3.6) can be expressed as X 1 n
Case 2
 
X 1 The ratio X /  st of the dynamic to the static  st  
2
2. 1
 (3.10) amplitude of the motion is called magnification 1   n
 st  n 
2
 
1   factor, amplification factor, or amplitude ratio 
 n  3. 1
n Magnification factor of an undamped system
• Case 1: • Case 2:
When 0 <  / n < 1, the denominator in Eq.(3.10) is positive and When  / n > 1, the denominator in Eq.(3.10)
the response is given by Eq.(3.5) without change. The harmonic is negative and the steady-state solution can
response of the system is in phase with external force, shown in be expressed as
figure.
x p (t )   X cos t (3.11)
x p (t )  X cos t (3.5)
where the amplitude is redefined as
 st  X cos(t   )
X

1
(3.10) X 2
(3.12)
 st   
2
 
1     1 ( n  , X  0)

 n  相角差=0o  n 
The response of the system to a
Harmonic response when 0   / n  1 harmonic force of very high frequency 相角差=180o
is close to zero Harmonic response when  / n  1
3.3 Response of an Undamped System Under

Harmonic Force
• Case
3:
When  / n  1 , the amplitude X given by Eq.(3.10) or (3.12)
becomes infinite. The forcing frequency  is equal to the natural
frequency n of the system is called resonance.
The total response if the system at resonance is rewrite Eq. (3.9) as
 F0  x   F0 
x ( t )   x0  2 
cos  n t   0  sin  n t   2 
cos  t (3.9)
 k  m   n   k  m  與時間成正比關係
x   F0 
 x0 cos  n t   0  sin  n t   2 
(cos  t  cos  n t )
 n   k  m 
x   cos  t  cos  n t 
 x0 cos  n t   0  sin  n t   st   (3.13)
 n   1  (  n ) 
2

x0  stn t 3.3 Response of an Undamped System Under

x( t )  x0 cos n t  sin n t  sin n t (3.15) Harmonic Force
n 2
F0
• Total Response x( t )  C cos  t  C sin  t  cos  t (3.7)
1 n 2 n
k  m 2
The total response of the system, Eq.(3.7) or Eq.(3.9), can also be
expressed as
 st 
x( t )  A cos( n t   )  cos  t ; for 1 (3.16)
 
2
n
1  
• Or  n 
 st
x( t )  A cos( n t   )  X cos  t  A cos( n t   )  2
cos  t
 
  1
And  n 
 st 
x( t )  A cos( n t   )  cos  t ; for 1 (3.17)
 
2
n
Harmonic response when  / n  1 (i.e.,  n ) 1   
台灣師範大學機電科技學系 -21-  n  台灣師範大學機電科技學系 -22-

 st 
Harmonic Force x( t )  A cos( n t   )  cos  t ; for 1 (3.17)
 
2
n
1   
 n 
 st 
x( t )  A cos( n t   )  cos  t ; for 1 (3.16)
 
2
n
1  
 n 
The complete motion can be expressed as the sum of two cosine

curves of different frequencies.


n  
Harmonic Force
(F m)   n     n    2 (3.19)
x( t )  2 0 2 (2 sin t  sin n t) (3.18)
n   2 2
• Beating Phenomenon  n    2 (3.20)
F /m 
If the forcing frequency is close to, but not exactly equal to, the  0 sin  t  sin  t ( 3.22) (3.19)  (3.20)
 2 
natural frequency of the system, beating may occur.   n2   2  4
數值小、週期大數值大、週期小
The phenomenon of beating can be expressed as Eq. (3.18) from Eq.
(3.9) under the initial conditions of x0  x0  0 : 波的振幅
 F0  x   F0 
x ( t )   x0  cos  n t   0  sin  n t   cos  t
k  m 2  2 
(3.9)
  n   k  m  波的振幅
(F m)
 2 0 2 (cos  t  cos  n t )
n  
( F0 m )   n n  
 (2sin t  sin t) (3.18)
 n2   2 2 2 the period of beating
Example 3.1 Plate Supporting a Pump
• Beating Phenomenon
A reciprocating pump, having a mass of 68kg, is mounted at the
middle of a steel plate of thickness 1cm, width 50cm, and length
The time between the points of zero amplitude or the points of
250cm, clamped along two edges as shown in Figure. During operation
maximum amplitude is called the period of beating and is given of the pump, the plate is subjected to a harmonic force, F(t) = 220 cos
by 62.832t N. Find the amplitude of vibration of the plate.
2 2 1 2 2 n  
b      (3.23)
 n    2
b  2 2 n   (3.19)
The frequency of beating defined as
b  2  n  

Example 3.1 Plate Supporting a Pump
Example 3.1 Plate Supporting a Pump Solution
Solution
The plate can be modeled as a fixed-fixed beam having Young’s
F0  220 N
Thus,
modulus (E) = 200GPa, length = 250cm, and area moment of inertia, F0 m  68 kg
X (3.6) k  102400.82 N / m
1 k  m 2
I (50 102 )(1102 )3  41.667 109 m4 plate b=1 cm
  62.832 rad / s
12
a=50 cm
F0 220
1 3 X 
The bending stiffness of the beam is given by: I ab k  m 2 102400.82  68(62.832)2
12
 0.00132487 m  1.32487 mm (E.2)
192 EI (192)(200 109 )(41.667 109 )
k   102400.82 N/m (E.1)
l3 (250 102 )3 The negative sign indicates that the response x(t) of the plate is
out of phase with the excitation F(t)
3.4 3.4 Response of Damped System Under Harmonic

Response of a Damped System Under Harmonic Force
Force
•If the forcing function is given by F (t )  F0 cos t
•The equation of motion becomes mx  cx  kx  F0 cos t (3.24)

•The particular solution of Eq. (3.24) is x p (t )  X cos(t   ) (3.25)
3.4
•By substituting Eq.(3.25) into Eq.(3.24)
  
X  k  m 2 cos t     c sin t     F0 cos t

(3.26)
•Using trigonometric relations,
cos( t   )  cos  t cos   sin  t sin 
sin( t   )  sin  t cos   cos  t sin 

• Equating the coefficients of cost and sint on both sides of the 3.4 Response of Damped System Under Harmonic
resulting equation, we obtain Force
• The figure shows typical plots of the forcing function and steady-
 
X  k  m 2 cos   c sin    F0 The particular solution
=Steady-state response state response.
X  k  m  sin   c cos    0
 2
(3.27) x p ( t )  X cos( t   ) (3.25)
just find X and  t   t  
sin  c c
 k  m  sin   c cos   0  tan   cos
2

 k  m 2
   tan 1 (
k  m 2
)
t  
(3.29)
 
2
if A  [ k  m 2  (c )2 ]1/ 2
F0 F0 (b) Vectorial representation
X  (a) Graphical representation
 k  m  cos   c sin 
2
 k  m   c  c 
2
for Eq.(3.26)
 k  m 2 
F F F0
A A
 
X  k  m 2 cos  t     c sin  t     F0 cos  t

(3.26)
 0  0  (3.28)
  kX cos  t     m X cos  t     c X sin  t     F0 cos  t
2
A A [ k  m 2  (c )2 ]1/ 2 2
A
3.4 Response of Damped System Under Harmonic

Force
• Dividing both the numerator and denominator of Eq. (3.28) by k
Undamped natural
and making the following substitutions frequency:
Undamped natural  c   c k 
X
F0
(3.28) 
F0 k   tan 1  2 
(3.29)  tan 1   k
frequency:  k  m   1  m k 
2
n  ;
   
1/ 2 1/ 2
 k  m 2 2
 c 2 2   k  m 2 2
 c 2 2  /k
    m
 m 

 st

 st
n 
k
;  ( 2 mk ) k   2  c c c c
 tan 1
 tan 1
 k       2n ;
 1  m 2 k 
1/ 2 1/ 2
 m 2  c 2 2  2
  2  4 2 mk 2 2
m ;
 1    2   1  2       1   2
m k  cc 2mn 2 mk m

k  k 
  n  k2  c c c c  
    ;  2n ;  
 st  st cc 2mn 2 mk m Deflection under the static
     force:
   2  
1/ 2 1/ 2
  2 2 4 2 2   1 r 2 2
 ( 2 r )  2
2 r 
 1  2  
 n  n2 
   Deflection under the static  tan 1  n
  tan 1  2 
(3.31) F0
 force:  1   n 
2 2
 1 r   st  ;
  k
F0   
X

1

1
(3.30)  st  ; Frequency ratio: r
 st    4  
1/ 2

 1 r
  ( 2 r ) 
1/ 2 k n

2 2
2 2 2 2 2
 1  2      Frequency ratio: r
 n  n2 
n
3.4 Response of Damped System Under Harmonic 3.4 Response of Damped System Under Harmonic
Force Force
• The following characteristics of the magnification factor (M=X/st)
can be noted from the figure as follows: 1. For an undamped system (  0) , Eq.(3.30) reduces to Eq.(3.10),
X 1 and M   as r  1 .

 st   2 2 4 2 2 
1/ 2
 1  2    2. Any amount of damping (  0) reduces the magnification factor

n  n2 
  (M) for all values of the forcing frequency.
1
 (3.30)
 
1/ 2
 1 r 2 2
 ( 2 r )2  3. For any specified value of r, a higher value of damping reduces the
 
value of M.
if   0
4. In the degenerate (退化的) case of a constant force (when r = 0),
X 1
 (3.10) the value of M = 1.
 st   
2
1  
 n 
Force Force
5. The reduction in M in the presence of damping is very significant at 8. The maximum value of X (when r  1  2 2 ) is given by:
or near resonance.
X 1  X  1
 (3.30)     (3.33)
The amplitude of forced vibration becomes smaller with increasing  st 
 
1/ 2
6.  1 r  (2 r )   st max 2 1  
2 2
2 2
values of the forcing frequency (that is, M  0 as r   ).  
(可實驗求得X與，代入求得，反之亦然 )
1 and the value of X at   n by
7. For 0     0.71 , the maximum value of M occurs when
2
X 1  X  1
  (3.30)     (3.34) ( i .e ., r   n  1)
 st   st  n 2
 
1/ 2
r  1  2 2  or   n 1  2 2 (3.32)  1  r 2 2  (2 r )2 
 
n (see Problem 3.32)
1 dM 1
which can be seen to be lower than the undamped natural 9. For   ,  0 when r = 0. For    0.71 , the graph
2 dr 2
frequency n and the damped frequency d  n 1   2 of M monotonically decreases with increasing values of r.

Force Force
• The following characteristics of the phase angle can be observed
from the figure and Eq.(3.31) as follows: 1. For an undamped system (  0) , Eq.(3.31) shows that the phase
angle is 0 for 0 < r < 1 and 180° for r > 1. This implies that the
excitation and response are in phase for 0 < r < 1 and out of phase
   for r > 1 when   0 .
 2  
  tan 1  n  2. For   0 and 0 < r < 1, the phase angle is given by 0 < Φ < 90°,
 1   n 
2 2
  implying that the response lags the excitation.

 
●
 2 r  3. For   0 and r > 1, the phase angle is given by 90° < Φ < 180°,
 tan 1  2 
(3.31) implying that the response leads the excitation.
 1 r 
Force Force
• Total response: The complete solution is x(t )  xh (t )  x p (t )
4. For   0 and r = 1, the phase angle is given by Φ = 90°,
For an underdamped system,
implying that the phase difference between the excitation and the
response is 90°. x(t )  xh  x p  Eq.(2.70)  Eq.(3.25)
x(t )  X 0e nt cos(d t  0 )  X cos(t   ) (3.35) ( where d  1   2 n )

5. For   0 and large values of r, the phase angle approaches 180°,
implying that the response and the excitation are out of phase. F  st  st
X  20 1   12/  (3.28)  (3.36)  (3.30)
   
2 n 1/ 2 1/ 2
 k  m
d
2
c  
2 2   2  4 2 2 
2
 1 r 2 2
 ( 2 r )2 
   1  2     
 n  n2 

  
 c   2   2 r 
  tan 1  2 
( 3.29 ) = tan 1
 n
  tan 1  2 
(3.31)
 k  m   1   n   1 r 
2 2
 
 

X0 and 0 can be determined from the initial conditions,
Force
x(t  0)  x0 and x(t  0)  x0 , Eq.(3.35) yields
Example 3.3
x(t )  X 0e nt cos(d t  0 )  X cos(t   ) (3.35) Total Response of a System
Find the total response of a single degree of freedom system with m
解聯立
x( t  0)  x0  X 0 cos 0  X cos  =10kg, c = 20 N-s/m, k = 4000 N/m, x0 = 0.01 m, x0  0 under the
方程式 x( t  0)  x0  n X 0 cos 0  d X 0 sin 0   X sin  (3.36) following conditions:
a. An external force F (t )  F0 cos t acts on the system with

The solution of Eq.(3.36) gives X0 and 0 as
  10 rad/s and F0  100 N .
a. Free vibration with F(t) = 0.
(Practice by yourself)
Force Force
Example 3.3 Total Response of a System
Solution Example 3.3
a. From the given data, Total Response of a System
Solution
x(t )  X 0e nt cos(d t  0 )  X cos( t   ) (3.35)
 2r   2  0.05  0.5 
We have   tan 1  2 
 tan 1    3.814075 (E.2)
n    20 rad/s 1 r   1  0.5 
k 4000 2
m 10
 st  F0
 100
 0.025 m
k 4000
Using initial conditions x0  0.01 and x0  0 , Eq.(3.36) yield:
 c
cc
 c
2 km
 2
20
 4000 10 
 0.05
0.01  X 0 cos 0  (0.03326)(0.997785)
d  1   2 n  1   0.05   20   19.974984 rad/s
2
 X 0 cos 0  0.023186 (E.3)

r  n  10
20
 0.5
 st 0.025
X   0.03326 m (E.1)
0  (0.05)(20) X 0 cos 0  X 0 (19.974984) sin 0
1  r    2 r  1  0.05    2  0.5  0.5  
2 2 2 2
2 2
(0.03326)(10) sin(3.814075) (E.4)

Force Force
Example 3.3 Total Response of a System Example 3.3 Total Response of a System
Solution Solution
Substituting Eq.(E.3) into (E.4), b. For free vibration, the total response is
X 0 sin 0  0.002268 (E.5) x(t )  X 0e nt cos(d t  0 ) (E.8)

Hence, Using the initial conditions, x0  0.01 and x0  0 (See Eqs.(2.73) and (2.75)))

X 0  ( X 0 cos 0 )  ( X 0 sin 0 )
2

2 1/ 2
 0.023297 (E.6)
1/ 2
   x  
1/ 2
  0.05  20  0.01  
2 2
X 0   x02   n 0    0.012      0.010012 (E.9)

X 0 sin 0   d     19.974984  
tan 0   0.0978176
X 0 cos 0
 x0  n x0  1  0.05  20 
0  5.586765 (E.7) 0  tan1    tan    2.865984 (E.10)
 d x0   19.974984 

Force  X 
  
1
(3.33)
  st max 2 1   2 1
Q (3.38)
2
• Quality factor and bandwidth:
For small values of damping (  0.05,)
 X   X  1
      Q (3.38)
  st  max   st  n 2 2 2  (1  r 2 )2  (2 r )2 
The value of the amplitude ratio at resonance

is called as quality factor or Q factor of the
system
From figure, R1 and R2, where the amplification factor falls to Q 2 , are called ( 2  0)
half-power points because the power absorbed (W) by the damper, responding
harmonically at a given frequency, is proportional to the square of the
amplitude (See Eq.(2.94))
W  cX 2 (3.39)
12  R12n2 R12 1  2
and 3.5
22  R22n2 R22 1  2 Response of Damped System Under F (t )  F0eit
From Eq .(3.42)
代入得 
1
1  2 1
n
2
1  2 1
n
3.5
 X   X  1
     Q (3.38)
  st max   st  n 2
3.5 Response of Damped System Under F (t )  F0eit 3.5 Response of Damped System Under F (t )  F0eit
• The equation of motion becomes
y
mx  cx  kx  F0eit (3.47) • Using the relation x  iy  Ae i  A  x 2  y 2 and tan  
x
• Assuming the particular solution F
X 0
ei
(k  m ) 
(3.51)
 c 2 2
2 2 1/ 2
it
x p (t )  Xe (3.48)
• Substituting Eq. (3.48) into Eq. (3.47),  c 

  tan 1  2 
(3.52)
F0  k  m 
X  (3.49)
(k  m 2 )  ic
• Hence, the steady-state solution, Eq.(3.48), becomes
Multiplying the numerator and denominator by [( k  m 2
)  ic ] x p ( t )  Xe i t ( 3.48)
and separating the real and imaginary parts, we obtain
F0 F0
 e  i e i t  e i ( t  ) (3.53)
 k  m 2 c 
1/2 1/2
( k  m 2 )2  c 2 2  ( k  m 2 )2  (c )2 
X  F0  i
(k  m 2 )2  c 2 2 
(3.50)
 (k  m )  c 
2 2 2 2

it
3.5 Response of Damped System Under F (t )  F0e kX 1 kX 1
H ( i )   (3.54) H ( i )   ( 3.55)
F0 1  r 2  i 2 r F0 [(1  r 2 )2  (2 r )2 ]1/ 2
 Frequency Response
 2r 
The complex frequency response H(i) is given from Eq. (3.49): H (i)  H (i) ei where   tan1  2 
(3.56) (3.57)
1 r 
F0 F0 F 1 The complex frequency-response function, H(i), contains
X = (3.49)= 0
( k  m 2 )  ic k[(1   2  n2 )  i 2 /  n ] k 1  r 2  i 2 r both the magnitude and phase of the steady-state
response
kX X 1 F
 H ( i )    (3.54)  X  0 H ( i )   st H ( i )
F0  st 1  r 2  i 2 r k Thus, the steady-state solution Eq. (3.53) becomes,
1 a  ib a b F0 F0
H =  i 2 x p ( t )  Xe i t  e i ( t  ) (3.53)  e i ( t  )
The absolute value of H(i) given by a  ib a 2  b 2 a 2  b 2 a  b2 1/ 2
k[(1   2  n2 )2  (2 /  n )2 ]1/ 2
( k  m 2 )2  (c )2 
a b 1
 H  ( )2  (  2 )2  c c c
a 2  b2 a  b2 a 2  b2 (    )
cc 2mn 2 mk
F0 1 F0
kX 1 (the same with Eq. (3.30) of the = e i (  t  )  H ( i ) e i (  t  )
(3.58)
H ( i )   ( 3.55) k [(1  r 2 )2  (2 r )2 ]1/ 2 k
F0 [(1  r 2 )2  (2 r )2 ]1/ 2 magnification factor X  st )
3.5 Response of Damped System Under F (t )  F0eit 3.5 Response of Damped System Under F (t )  F0eit
• Frequency Response • Frequency Response
If F (t )  F0 cos  t , the corresponding steady-state solution is given If F (t )  F0 sin  t , the corresponding steady-state solution is given
by the real part of Eq.(3.53) by the imaginary part of Eq.(3.53)
F0 The same with Eq. (3.25) F0

x p (t )  cos( t   ) x p (t )  sin( t   )
( k  m )  (c )  2 1/ 2 x p (t )  X cos(t   ) 1/ 2
2 2
( k  m 2 )2  (c )2 
F  F  F 
 Re  0 H ( i )e i t   Re  0 H ( i ) e i ( t  )  (3.59)  Im  0 H ( i )e i ( t  )  (3.60)
k  k  k 

it
3.5 Response of Damped System Under F (t )  F0e
The various terms of the equation of motion (3.47)
can be represented in the complex plane.
• Complex Vector Representation of Harmonic Motion
Differentiating Eq.(3.58) with respect to time,

the response x(t) lags
F the excitation F(t) with 
x p ( t )  0 H ( i ) e i ( t  ) (3.58)
k phase angle
 F0e it
 = /2
F F i
Velocity  x p ( t )  i 0 H ( i ) e i ( t  )  i x p ( t )   ( 0 H ( i ) e i ( t  ) )e 2
k k
2 F0 i (  t  ) F
Acceleration  x p ( t )  ( i ) H ( i ) e   x p ( t )   2 ( 0 H ( i ) e i ( t  ) )e i
2
k k
(3.61)
 phase angle
  i The velocity leads the displacement by the phase =
i  cos  i sin e 2
 angle /2, and that it is multiplied by 
2 2
mx  cx  kx  F0e it (3.47)
i The acceleration leads the displacement by the
1  cos   i sin   e 
phase angle , and that it is multiplied by 2
3.6 3.6 Response of Damped System Under the Harmonic

Response of Damped System Under the Harmonic Motion of the Base
Motion of the Base
• From the figure, the equation of motion is
mx  c( x  y )  k ( x  y )  0 (3.64)
3.6
The net elongation of the spring
is x-y and the relative velocity
between the two ends of the
damper is x  y
The base undergoes harmonic motion

3.6 Response of Damped System Under the Harmonic 3.6 Response of Damped System Under the Harmonic
Motion of the Base Motion of the Base
• If y(t )  Y sin t , Eq. (3.64) becomes • Equation (3.66) can also be written in a more convenient form as
mx  cx  kx  ky  cy  kY sin  t  cY cos  t  A sin( t   ) (3.65)
Similar with F ( t )  F0 sin  t
1  c 
where A  Y k  (c ) and   tan 
2 2 x p (t )  X sin(t   ) (3.67)    1  

 k  where
This shows that giving excitation to the base is equivalent to apply a
harmonic force of magnitude A to the mass
Y k 2  (c )2
x p (t )  1/ 2
sin( t  1   ) (3.66)  X sin( t   ) (3.67)
( k  m 2 )2  (c )2 
• By Eq. (3.60), the steady-state response of the mass can be expressed as
1/ 2 1/ 2
If F ( t )  F0 sin  t  Eq.(3.60) Y k 2  (c )2 Y k 2  (c )2 X  k 2  (c )2   1  (2 r )2 
x p (t )  sin( t  1   ) (3.66) X     2
( 3.68)
2 1/ 2 ( k  m 2 )2  (c )2 
1/ 2
Y  ( k  m 2 )  (c )2   (1  r )  ( 2 r ) 
2 2
x p (t ) 
F0
sin( t   ) ( k  m )  (c ) 
2 2
2 1/ 2
( k  m 2 )2  (c ) 
 c 
where The ratio of the amplitude of the response xp(t) to that of the base
where   tan 1  2 
(3.52) motion y(t), X/Y, is called the displacement transmissibility.
 k  m   c 
1  tan  1
2 
 k  m  -65- -66-
台灣師範大學機電科技學系台灣師範大學機電科技學系
 c   c 
1  tan 
1
2 
,   tan 1   
 k  m   k  The variations of
X
 Td and  given by Eqs. (3.68) and (3.69) are
  1   Y
shown for different values of r and 
tan 1  tan 
 tan   tan(1   ) 
1  tan 1 tan 
c c
 ( )
mc 3
 k  m
2
k 
c c k ( k  m 2 )  ( c )2
1 ( )(  )
k  m 2 k
 mc 3   2 r 3 
So   tan 1  2
 tan 1  2
( 3.69)
 k ( k  m )  ( c )   1  (4  1)r 
2 2
1/ 2
Y k  (c )
2 2
 1  (2 r ) 2

x p (t )  X sin( t   ) (3.67) X Y  2
 (1  r )  (2 r ) 
1/ 2 2 2
( k  m )  (c ) 
2 2 2

• The following aspects of displacement transmissibility can be noted 5. For r <√2, smaller damping ratios lead to larger values of Td. On
from the figure: the other hand, for r >√2, smaller values of damping ratio lead to
smaller values of Td.
1. The value of Td is unity at r = 0 and close to unity for small values
of r. 6. The displacement transmissibility, Td, attains a maximum for 0 < ζ
< 1 at the frequency ratio r = rm < 1 given by:
2. For an undamped system (ζ = 0), Td →∞ at resonance (r = 1).
3. The value of Td is less than unity (Td < 1) for values of r >√2 (for 1  1/ 2
rm  1  8 2  1 (See problem 3.60)
any amount of damping ζ ). 2  
4. The value of Td = 1 for all values of ζ at r =√2. For
  0.5  rm  0.73205
  0.2  rm  0.96473
• Force transmitted to the base or support, like in Fig. 3.14
• Note that if the harmonic excitation of the base is expressed in
complex form as y(t )  Re(Ye i t ) , the response of the system can This force can be determined as
be expressed, using the analysis of Section 3.5 , as F  k ( x  y )  c( x  y )  mx (3.72)
 1  i 2 r  i t  mx  c( x  y )  k ( x  y )  0 (3.64) From x p (t )  X sin( t   ) (3.67)
x p ( t )  Xe i t  Re   Ye  (3.70)
 1-r  i 2 r 
2
 mx  cx  kx  ky  cy
F  m 2 X sin( t   )  FT sin( t   ) (3.73)
 kYe i t  cYe i t  Ae i t
• The displacement transmissibility as.
where A  kY  cY
Where FT is the amplitude or Max. value of the force transmitted to
i t
X 1/ 2 類似3.5節之 F (t )  F0e the base. (F(t) is in phase with the motion of mass x(t))
 Td  1  (2 r )2  H ( i ) (3.71)
Y 1/ 2 The force transmissibility is
FT m 2 X m 2 X  2  1  (2 r )2 
1      FT
Where H ( i ) is given by Eq. (3.55) ( H ( i )  ) kY kY k Y  n2  (1  r 2 )2  ( 2 r )2 
known as the ratio of
[(1  r 2 )2  (2 r )2 ]1/ 2 where r    n Eq. (3.68) kY

3.6 Response of Damped System Under the Harmonic

The variations of the force transmitted to the base with the Motion of the Base
frequency ratio r is shown for different values of 
• Relative Motion
If z=x-y denotes the motion of the mass relative to the base, the
equation of motion, Eq. (3.64), can be rewritten as
mx  c( x  y )  k ( x  y )  0 (3.64) ( y(t )  Y sin  t )
mz  cz  kz  my  m2Y sin t (3.75)
The steady-state solution of Eq. (3.75) is given by:
● m 2Y sin(t  1 )
z (t )   Z sin(t  1 )
(k  m ) 
(3.76)
 (c ) 2
2 2 1/ 2
Similar with
Section 3.4
where Z, the amplitude of z(t), can be expressed as
(3.24)-(3.31)
See Section 3.7 m 2Y r2
Z Y (3.77) ,
(k  m 2 )2  (c )2 (1  r 2 )2  (2r )2
 c   2r 
1  tan 1  2 
 tan 1  2 
台灣師範大學機電科技學系 -73-  k  m  1 r  台灣師範大學機電科技學系 -74-
3.6 Response of Damped System Under the Harmonic

Motion of the Base
Example 3.4
Vehicle Moving On a Rough Road
The figure below shows a simple model of a motor vehicle that can
vibrate in the vertical direction while traveling over a rough road. The
vehicle has a mass of 1200kg. The suspension system has a spring
constant of 400 kN/m and a damping ratio of ζ = 0.5. If the vehicle
speed is 20 km/hr, determine the displacement amplitude of the
See Section 3.7 vehicle. The road surface varies sinusoidally with an amplitude of Y =
0.05m and a wavelength of 6m.
Variation of (Z/Y) or (MX/me) with frequency ratio r=/n

Example 3.4 Example 3.4

Vehicle Moving On a Rough Road Vehicle Moving On a Rough Road
Solution
The frequency can be found by   2 f  2  v 1000  1  0.290889v rad/s

 3600  6
For v = 20 km/hr, ω = 5.81778 rad/s. The natural frequency is given
by, 1/ 2
k  400 103 
n      18.2574 rad/s
m  1200 
The frequency ratio is

 5.81778
v( km / h) 1 v 1000 r   0.318653
f (cycles / s)   ( ) n 18.2574
 (m )  60  60
3.6 Response of Damped System Under the Harmonic 3.7

Motion of the Base Response of a Damped System Under Rotating
Unbalance
Example 3.4 Vehicle Moving On a Rough Road
Solution
The amplitude ratio can be found from Eq.(3.68):

1/ 2 1/ 2
X  1  (2r )2   1  (2  0.5  0.318653)2 
   2
 1.100964
3.7
Y  (1  r 2 )2  (2r )2   (1  0.318653)  (2  0.5  0.318653) 
2
The displacement amplitude of the vehicle is given by
X  1.100964Y  1.4100964(0.05)  0.055048 m

A 5 cm bump in the road is transmitted as a 5.5 cm bump to the
chassis (機殼) and the passengers of the car, and the passengers feel
an amplified motion.

3.7 Response of a Damped System Under Rotating 3.7 Response of a Damped System Under Rotating
Unbalance Unbalance
mv 2
Centrifugal force   mr 2
r
• The equation of motion can be derived by the usual procedure:
Mx  cx  kx  me 2 sin t (3.78)
• The solution of this equation will be identical to Eq.(3.60) if we

replace m and F0 by M and me 2 . The solution can be expressed as
x p (t ) 
F0
sin( t   ) x p ( t )  X sin( t   )
1/ 2
( k  m 2 )2  (c )2 
 me   2 
i (  t  )
 Im    H ( i ) e  (3.79)
F 
 M  n  
 Im  0 H ( i ) e i ( t  )  (3.60)
k 
where n  k M  k  Mn2
3.7 Response of a Damped System Under Rotating

Unbalance
n  k M  k  Mn2
• The amplitude X and phase angle  is given by
2
me 2 me    me 2
X    H ( i ) 
1/ 2
( k  M  2 )2  (c )2  M  n   2  2
1/ 2
k (1  2 )2  ( 2
n 
)
 n
      
     2 
 c  cc 1  2 M n  1  n 
  tan 1  2 
( 3.80 )  tan 1
   tan  tan
 k  M  M
 k (1   2 )    
2
  
2
 k   k (1   2 )   (1   2 ) 
 n   n 
• By defining   c / cc and cc  2Mn , Eq. (3.80) can be rewritten as

Fig. 3.17
The graph of  versus r
 2r 
2
MX r
  r 2 H ( i ) ;   tan1  2 
(3.81) remains as in Fig. 3.11(b)
me (1  r 2 )2  ( 2 r )2 1/ 2
  1 r 
Unbalance Unbalance
1
• The following observations can be made from Eq.(3.81) and the The solution of Eq.(3.82) gives: r  1
1  2 2
figure above:
1. All the curves begin at zero amplitude. The amplitude near  MX  1

With corresponding maximum value:   
resonance is markedly affected by damping. Thus if the machine is  me  max 2 1   2
to be run near resonance, damping should be introduced
Thus the peaks occur to the right of the resonance value of r = 1
purposefully (故意) to avoid dangerous amplitudes.
2. At very high speeds (ω large), MX/me is almost unity, and the

For   1 2 ,  
MX
effect of damping is negligible. 4. me  does not attain a maximum. Its value grows
from 0 at r = 0 to 1 at r → ∞ .
3. For 0 < ζ < 1/√2 (=0.71) , the maximum of MX/me occurs when
d  MX 
 0 (3.82)
dr  me 
Solution of
5.The force transmitted to the foundation due to the rotating Problem 3.73
unbalance force (F) can be found as F (t )  kx(t )  cx(t ) .
The magnitude (or maximum value) of F can be derived as
1/ 2
 1  4 2 r 2  (See Problem 3.73)
F  me 2   (3.84) (Practice yourself)
 1  r 2   4 2 r 2 
2

Unbalance Unbalance
Example 3.7 Francis Water Turbine
The schematic diagram of a Francis Example 3.7
water turbine is shown in the figure Francis Water Turbine
in which water flows from A into the Solution
blades B and down into the tail race
C. The rotor has a mass of 250 kg
and an unbalance (me) of 5kg-mm.
The max amplitude can be obtained from Eq.(3.80) by setting c = 0 as
The radial clearance between the
rotor and the stator is 5mm. The me 2 me 2
X  (E.1)
turbine operates in the speed range (k  M 2 ) k (1  r 2 )
600 to 6000rpm. The steel shaft The value of ω ranges from:
carrying the rotor can be assumed to
be clamped at the bearings. 2 2
600rpm  600   20 rad/s to 6000rpm  6000   200 rad/s
Determine the diameter of the shaft 60 60
so that the rotor is always clear of While the natural frequency is given by
the stator at all the operating speeds k k
of the turbine. Assume damping to n    0.625 k rad/s (E.2)
M 250
be negligible.
UnBalance UnBalance

From Fig. 3.17, we find that the amplitude of vibration of
Francis Water Turbine Francis Water Turbine the rotating shaft can be minimized by making r=/n
Solution Solution very large. This means that   k m must be made small
n
compare to that is, k must be made small.
For ω = 20π, Eq.(E.1) gives The stiffness of the cantilever beam is given by
(5.0  103 )  (20 )2 2 2
0.005    k  10.04  104  2 N/m (E.3) 3EI 3E   d 4 
 (20 )2  k  105 2 k  3   (E.5)
k 1  l3 l  64 

 0.004k 
兩者該如何選擇呢？
For ω = 200π, Eq.(E.1) gives Diameter of the beam is
64kl 3 (64)(10.04 104  2 )( 23 )
(5.0 10 )  ( 200 )
3 2
200 2 d4    2.6005 104 m 4
0.005    k  10.04 106  2 N/m (E.4) 3 E 3 ( 2.07 1011 )
 ( 200 )2  k  107  2
k 1 
 0.004k  d  0.1270 m  127 mm (E.6)
3.8 3.8 Force Damping with Coulomb Damping

Force Vibration with Coulomb Damping • From Fig. 3.22, the equation of motion is given by
mx  kx   N  F (t )  F0 sin t (3.85)
If the dry-friction force is large, the motion of the mass will be discontinuous.
On the other hand, if the dry-friction is small compared to the amplitude of
the applied force F0, the steady-state solution is expected to be nearly
harmonic.
3.8 台灣師範大學機電科技學系 -93-

Single-degree-of-freedom system with Coulomb damping
3.8 Force Damping with Coulomb Damping 3.8 Force Damping with Coulomb Damping
• The equation of motion is given by • The amplitude can be found from Eq.(3.60):
mx  kx  N  F (t )  F0 sin t (3.85) X

F0

( F0 / k )
(k  m ) 
(3.90)
 (ceq ) 2
1/ 2 1/ 2
2 2
  2  2   
2
• The energy dissipated by dry friction damping during a full cycle of motion 1  2    2 eq  
is
 n   n  

W  4NX (3.86) (X is the amplitude) where
ceq ceq 4 N 2 N
If the equivalent viscous damping constant is denoted as c eq,  eq     (3.91)
•
cc 2mn 2mn X  mn X
W  ceqX 2 (3.87) (See Eq.(2.94))
• Substitute Eq.(3.91) into (3.90) gives:
(The energy dissipated due to dry n 
k
 mn2  k
4 N
ceq  (3.88) friction= The energy dissipated due m
 X to an equivalent viscous damper) ( F0 / k ) ( F0 / k ) ( F0 / k )
• Thus the steady-state response is: X 1/2
 1/2
 1/2
(3.92)
  2  2     2  2 2 N  2    2  2  4  N  2 
1  2   (2 eq )2  1  2   (2 )   1  2     
x p (t )  X sin(t   ) (3.89)  n  n 
  n   mn X n 
  n    kX  
必須抽出
• The phase angle  appearing in Eq(3.89) can be found using

The solution of this equation gives the amplitude X as Eq(3.52):
• 4 N
ceq  2 eq mn  (3.88)
1/ 2  X
  4 N  2 
1     Eq.(3.93) shows that friction serves to     4 N 
F   F0    ceq   2 eq   
X  0  n
limit the amplitude of forced vibration
(3.93) for    1 . However, at resonance   tan 1   1    tan 1   kX  (3.94)
2 
k   2 2 
(3.52) tan
n
 k  m   2   2 
 1    (  n  1 ), the amplitude become  1 2   1 2 
  n2   infinite.  n   n 
X=Eq.(3.93)
   
 4 N   4 N 
• To avoid imaginary values of X, we need to have     
1

 F0  1

 F0  (3.96)
= tan (3.95)  tan
 4 N 
2
  2 1/2 

    2 1/2 
 4 N  
F0 4 
 1   4  N 
1   0 or   F0  4 N     1     
    F0        F0   
  F0  N       
Eq.(3.95) shows that tan is a constant for a given value of N/F0.  is discontinuous
at /n=1 (resonance), since it takes a positive value for /n<1 and a negative
value for /n>1. Thus Eq.(3.95) can also be expressed as Eq.(3.96).
3.8 Force Damping with Coulomb Damping

•The energy directed into the system over one cycle when it
is excited harmonically at resonance is
2 / 
 dx   2 / 
(3.97)  W    F0 X  sin t cos(t   )dt   F0 X sin  (3.99)
W    F dx   F dt   F0 sin t [ X cos(t   )]dt (3.97) 0
cycle 0 dt 0
( F (t )  F0 sin t; x p (t )  X sin(t   ))

Due to the presence of sin  in Eq.(3.99), the input energy curve
Since Eq. (3.94) gives  =90o at resonance, Eq. (3.97) becomes in Fig. 3.23 is made to coincide with the dissipated energy curve,
so the amplitude is limited. Thus the phase of the motion  can be
2 /  seen to limit the amplitude of the motion.
W   F0 X   sin 2 t dt   F0 X (3.98)
0
Fig. 3.23 Energy input and energy

dissipated with Coulomb damping

Example 3.8
Example 3.8
Spring Mass System with Coulomb Damping
Solution
A spring-mass system, having a mass of 10kg and a spring of stiffness
of 4000 N/m, vibrates on a horizontal surface. The coefficient of
The vertical force (weight) of the mass is N = mg = 10 X 9.81 =
friction is 0.12. When subjected to a harmonic force of frequency 2 Hz,
98.1N. The natural frequency is
the mass is found to vibrate with an amplitude of 40 mm. Find the
k 4000
amplitude of the harmonic force applied to the mass. n    20 rad/s
m 10
Frequency ratio is
 2  2
  0.6283
n 20
3.8 Force Damping with Coulomb Damping 3.9

Forced Vibration with Hysteresis Damping
Example 3.8
Solution 1/ 2
  4 N  2 
1    
F   F0  
X  0 
3.9
2  (From
(From)
Eq. (3.93))
 1    
k 2
 2 
  n  
  4(0.12)(98.1)  2 
1    
F0   F0    F  97.9874 N
0.04 
4000  (1  0.6283 ) 
2 2 0
 
 
3.9 Forced Vibration with Hysteresis Damping 3.9 Forced Vibration with Hysteresis Damping
• From the figure below, the equation of motion can be derived
k • Substituting Eq.(3.101) into
mx  x  kx  F0 sin t (3.100) • Eq.(3.100), we obtained

F0
P 195  k h X (3.102)
Eq.(2.155)-(2.157) ceq  cc   eq  2 mk    mk   (2.157)   2  2 
1/ 2
2   k 1  2    2 
k k  n  
(   mk  )
m 
• Steady-state solution can be assumed as x p (t )  X sin(t   ) (3.101)  

k  
( p193  p195) c 
h
,
h
c   
    tan 1
(3.103)
k   2  
 1  2  
  n  
 : the hysteretic damping constant Fig. 3.25 Steady-state response

System with hysteresis damping (Compared with Fig. 3.11 of next page)
3.9 Forced Vibration with Hysteresis Damping
X
• The amplitude ratio is
( F0 / k )
attains its max. value of F0/k at the resonant frequency (n) in
the case of hysteresis damping, while it occurs at a frequency
below resonance (<n) in the case of viscous damping.
• The phase angle  has a value of tan-1(β) at ω = 0 in the case of

hysteresis damping, while it has a value of zero at ω = 0 in the
case of viscous damping. This indicates that the response can never
be in phase with the forcing function in the case of hysteresis
damping.
Fig. 3.11 variation of X and  with frequency ratio r

3.9 Forced Vibration with Hysteresis Damping 3.10

• If the harmonic excitation is assumed to be F (t )  F0eit , the equation Forced Motion with Other Types of Damping
of motion becomes
k k
mx  x  kx  F0eit (3.104)  mx  (i x)  kx  F0eit
 
The response x(t )  Xeit (is also harmonic function)  x(t )  i x(t )
mx  k (1  i ) x  F0eit (3.105)
3.10
where the quantity k (1  i ) is called the complex stiffness or
complex damping.
• The steady-state solution is given

F0e i t
x( t )  (3.106)
   2 
k 1     i 
   n  
3.10 Force Motion with Other Types of Damping 3.10 Force Motion with Other Types of Damping
Example 3.9
Quadratic Damping
Solution
The damping force is assumed to be
Fd   a( x )2 (E.1) where a is a constant and x(t )  X sin  t
The energy dissipated per cycle is

Example 3.9 Quadratic (二次) Damping  /2  /2
dx
W  2 a( x )2 dx   )d ( t )  2a 2 X 3  cos 3  td ( t )
x
a ( x )2 (
Find the equivalent viscous damping coefficient corresponding to x  / 2 d t   /2
quadratic or velocity squared damping that is present when a body  / 2 cos 3 t  3cos  t 4 8
moves in a turbulent fluid flow.  2a 2 X 3  d ( t )  2a 2 X 3 ( )   2aX 3 (E.2)
 / 2 4 3 3
W   ceq X 2 (E.3) (See Eq. (2.94))

3.10 Force Motion with Other Types of Damping 3.11

Self-Excitation and Stability Analysis
Example 3.9 Quadratic Damping Ceq is not a constant but
Solution varies with  and X
( E .2)  ( E .3)
The equivalent viscous damping coefficient is 8
 ceq  a X (E.4)
3
The amplitude of the steady-state response is
3.11
X 1  ceq ceq
 (E.5) where r  and  eq   (E.6)
 st (1  r 2 ) 2  (2 eq r ) 2 n cc 2mn
(From Eq. (3.30))
Using Eqs. (E.4) and (E.6), Eq.(E.5) can be solved to obtain

1/ 2
 2
3m  (1  r 2 ) 2 (1  r 2 ) 4  8ar 2 st  
X      (E.7)
8ar   3m  
2
2 4

3.11 3.11 Self-Excitation and Stability Analysis

Self-Excitation and Stability Analysis
• The force acting on a vibrating system is usually external to the • Dynamic Stability Analysis
system and independent of the motion. However, there are
Consider the equation of motion of a single degree of freedom system:
systems for which the exciting force is a function of the motion
parameters of the system, such as displacement, velocity, or
acceleration. Such systems are called self-excited vibrating
mx  cx  kx  0 (3.107)
systems, since the motion itself produces the exciting force.
If a solution of the form x(t )  Ce st is assumed, this leads to a
• The instability of rotating shafts, the flutter (振顫) of turbine blades,
characteristic equation
the flow-induced vibration of pipes, and the automobile wheel
shimmy(汽車的異常振動), and the aerodynamically (空氣動力)- c k
s2  s 0 (3.108)
induced motion of bridges are typical examples of self-excited m m
vibrations.
The roots of the equation are: The motion will be diverging
and aperiodic if the roots s1
• The motion diverges and the system becomes unstable if energy is 1/ 2 and s2 are real and positive.
fed into the system through self-excitation. c 1  c   k 
2
This situation can be avoided
s1, 2       4  (3.109) if c/m and k/m are positive.
2m 2  m   m 
3.11 Self-Excitation and Stability Analysis 3.11 Self-Excitation and Stability Analysis
• Dynamic Stability Analysis Example 3.10
Instability of Spring-Supported Mass on Moving Belt
The motion will also diverge if the roots
Let the roots be expressed as s1 and s2 are complex conjugates with Consider a spring-supported mass on a moving belt as shown in the
positive real parts.
Figure (a). The kinetic coefficient of friction between the mass and the
s1  p  iq, s2  p  iq (3.110) belt varies with the relative (rubbing) velocity as shown in Figure (b).
As rubbing velocity increases, the coefficient of friction first decreases
where p and q are real numbers so that
from its static value linearly and then starts to increase. Assuming that
c k the rubbing velocity, v, is less than the transition value, vQ, the
( s  s1 )(s  s2 )  s 2  ( s1  s2 ) s  s1s2  s 2  s 0 (3.111)
m m coefficient of friction can be expressed as
Hence,
a a
c
 ( s1  s2 )  2 p,
k
 s1s2  p 2  q 2 (3.112)   0  v  0  (V  x)
m m W W
Where a is a constant and W=mg is the weight of the mass. Determine
Eq.(3.112) show that for negative p, c/m must be positive and for positive p 2+q2,
k/m must be positive. Thus the system will be dynamically stable if c and k are the nature of the free vibration about the equilibrium position of the
positive (assuming that m is positive). mass.

Instability of Spring-Supported Mass on Moving Belt Instability of Spring-Supported Mass on Moving Belt
Solution
Let the equilibrium position of mass m correspond to an extension of

x0 of the spring. Then,
W  kx0
W 0W aV
v V  x V
x0    x  0 at the equilibrium position
k k k
The rubbing velocity v is given by:
Motion of a spring-supported mass due to belt friction v  V  x

Example 3.10
Example 3.10 Instability of Spring-Supported Mass on Moving Belt
Instability of Spring-Supported Mass on Moving Belt Solution
Solution The solution is given by
The equation of motion for free vibration is
 a 
mx   k ( x0  x )  W   k ( x0  x )  W  0  (V  x ) 

x(t )  e ( a / 2m ) t C1e r1t  C2e r2t  (E.2)
 W 
Since the coefficient of x
is negative, where C1 and C2 are constants
 mx  ax  kx  0 (E.1)
the motion given by Eq.(E.1) will be
unstable
Thus, 1/ 2
1  a   k 
2
As can be seen from Eq. (E.2), the value
r1     4  of x increases with time. It increases
The characteristic roots 2  m   m 
from V  x  0 to V  x  vQ . After this,
a 1 a 1 a k a 1/ 2 the  will have a positive slope, and hence
s1,2   a 2  4mk   [( )2  4( )]1/ 2   r1,2 1  a   k 
2
r2      4  the nature of the motion will be different.

2m 2m 2m 2 m m 2m
2  m   m 
3.11 Self-Excitation and Stability Analysis • Dynamic Instability Caused by Fluid Flow
• Dynamic Instability Caused by Fluid Flow

疾馳
• The vibration caused by a fluid flowing around a body is known as 顫動機翼

flow-induced vibration, in which the vibration of the system
continuously extracts energy from the source, leading to larger and
(旋渦脫落、旋渦分離、渦流溢放 )
larger amplitude of vibration.
• For example, tall chimneys (煙囪), submarine periscopes (潛望鏡),

electric transmission lines, and nuclear fuel rods are found to
vibrate violently under certain conditions of fluid flow around them.
Similarly, water and oil pipelines and tubes in air compressors
under severe vibration under certain conditions of fluid flow through
them.
(Induce the instability)

• Dynamic Instability Caused by Fluid Flow • Dynamic Instability Caused by Fluid Flow
The figure illustrates the phenomenon of galloping of wires: The figure illustrates the phenomenon of singing of wires:
Smooth cylinder
Under certain conditions, alternating vortices (旋渦)
in a regular pattern are formed downstream, which
are called Karman vortices. They are alternately
clockwise and counterclockwise and thus cause
harmonically varying lift forces on the cylinder
perpendicular to the velocity of the fluid.
Fluid flow past a cylinder
Experimental data show that regular vortex shedding occurs strongly in the
range of Reynolds number (Re) from about 60 to 5000. In this case,
where d is the diameter of the cylinder, 

Vd
Re  (3.113) is the density, V is the velocity, and  is the
Fig. 3.28 Galloping of a wire  absolute viscosity of the fluid
• Vortex shedding (旋渦脫落、旋渦分離、渦流溢放 )(影片欣賞) 3.11 Self-Excitation and Stability Analysis
• Dynamic Instability Caused by Fluid Flow
For Re > 1000, the dimensionless frequency of vortex shedding, expressed

as a Strouhal number (St), is approximately equal to 0.21.
fd
St   0.21 (3.114) where f is the frequency of vortex
V shedding
In fluid dynamics, vortex shedding is an unsteady oscillating flow that takes place when a fluid such
as air or water flows past a blunt cylindrical body at certain velocities, depending to the size and
shape of the body. In this flow, vortices are created at the back of the body and detach periodically The harmonically varying lift force (F) is given by
from either side of the body. The fluid flow past the object creates alternating low-pressure vortices
1
on the downstream side of the object. The object will tend to move toward the low-pressure zone. If
the cylindrical structure is not mounted rigidly and the frequency of vortex shedding matches the
F (t )  cV 2 A sin t (3.115)
2
resonance frequency of the structure, the structure can begin to resonate, vibrating with harmonic
oscillations driven by the energy of the flow. This vibration is the cause of the "singing" of overhead where
power line wires in a wind, and the fluttering of automobile whip radio antennas at some speeds. Tall c = constant (c  1 for a cylinder)
chimneys constructed of thin-walled steel tube can be sufficiently flexible that, in air flow with a A = projected area of the cylinder perpendicular to the direction of V
speed in the critical range, vortex shedding can drive the chimney into violent oscillations that can ω = circular frequency and t is the time.
damage or destroy the chimney.

3.11 Self-Excitation and Stability Analysis
From a design point of view, we have to ensure the following:
1.The magnitude of the force exerted on the cylinder, given by Eq.

(3.115), is less than the static-failure load.
2.Even if the magnitude of force F is small, the frequency of

oscillation (f) should not cause fatigue during the expected
lifetime of the structure (or cylinder).
3.The frequency of vortex shedding (f) does not coincide with the
natural frequency of the structure or cylinder to avoid 擾流板邊條
resonance.
駕駛
擾流板
Fig. 3.30 Stockbridge damper
Fig. 3.31 Helical spoilers

Tuned Liquid Damper (TLD) Tuned Liquid Damper (TLD)

The Tuned Liquid Damper (TLD) relies upon fluid to provide a cost-effective method for Several compartments are connected to form a “Chamber”. The damper will have a
damping. The fluid is contained in an annular box, which is provided near the top of the minimum of two chambers, but could have more depending upon the size. These multiple
stack. This box contains an array of baffles which separate the damper into many chambers provide a factor of safety in the design as well as redundancy . Depending upon
compartments. The baffles and compartments are designed to allow fluid to move the size of the damper, it may be supplied in one or more pieces. If the damper is designed
internally and create the required damping force. more than one piece, the assembly will be joined with bolted connections.
http://www.mecaenterprises.com/damper_liquid.htm
http://www.mecaenterprises.com/damper_liquid.htm
Tuned Mass Damper (TMD)

The Tuned Mass Dampers (TMD's) utilize mass to provide the damping. The dampers are
"tuned" to the natural frequency of the stack so that they are most effective at or near the
natural frequency of the stack. The Tuned Mass Damper consists of a cylindrical ring of
steel mass, which is suspended by cables. The ring is larger than the diameter of the stack,
and is separated from the stack with coils of cables which serve as springs.
Helical Strakes
Dampers for Bridge Cables and Guyed Towers

http://www.mecaenterprises.com/damper_mass.htm
http://mecaenterprises.com/stack_vibration
Example 3.12
Example 3.12 Flow-Induced Vibration of a Chimney
Flow-Induced Vibration of a Chimney
Model the chimney as a cantilever beam and equate the natural

A steel chimney has a height of 20 m, an inner diameter 0.75m, and frequency of the transverse vibration of the chimney to the frequency
an outer diameter 0.80m. Find the velocity of the wind flowing around of vortex shedding.
the chimney which will induce transverse vibration of the chimney in
the direction of airflow. The natural frequency of transverse vibration of a cantilever beam is
EI
1  ( 1l ) 2 (E.1) (See Section 8.7)
Al 4
where 1l  1.875104 (E.2)

Instability of typical vibrating structures
Example 3.12
Flow-Induced Vibration of a Chimney Example 3.12
Flow-Induced Vibration of a Chimney
For the chimney, E = 207X109 Pa, ρg = 76.5X103 N/m3, l = 20m, d = The frequency of vortex shedding is given by Strouhal number:
0.75m, D = 0.80m,
  fd
A ( D 2  d 2 )  (0.802  0.752 )  0.0608685 m 2 St   0.21
4 4 V
 
I  ( D  d )  (0.804  0.754 )  0.004574648 m 4
4 4
64 64 The velocity wind (V) which causes resonance can be determined as

1/ 2
 
 
(207 10 )(0.004574648) 
Thus, 1  (1.875104) 2 
9
f1d 1.975970(0.80)
  12.415417 rad/s  1.975970 Hz V   7.527505 m/s
  76.5 10 (0.0608685)(20) 4 
3
0.21 0.21
  9.81  

3.12 3.12 Transfer-Function Approach

Transfer-Function Approach
• The transfer-function approach, based on Laplace transforms, is
used for the formulation and solution of dynamic problems in the
controls literature. Also used for solving forced-vibration problems.
• Transfer function relates a system’s output to its input. This

function permits separation of the input, system, and output into
3.12
three separate and distinct parts (unlike the differential equation, in
which the three aspects cannot be separated easily).
• The transfer function of a linear, time-invariant differential equation

is defined as the ratio of the Laplace transform of the output or
response function x(t) to the Laplace transform of the input or
forcing function f(t), assuming zero initial conditions.
L[ x( t )] X ( s )
T ( s)  
L[ f ( t )] F ( s )
3.12 Transfer-Function Approach 3.12 Transfer-Function Approach

Transfer Function Corresponding to a Differential Equation Transfer Function Corresponding to a Differential Equation
Solution L[ f ]  s 2 F ( s )  s1 f (0)  f (0)
Consider the following nth-order linear, time-invariant differential
equation that governs the behavior of a dynamic system: L[ f ]  s 3 F ( s )  s 2 f (0)  sf (0)  f (0)
By taking Laplace transforms of both sides of Eq. (E.1), we obtain

d n xt  d n 1 xt 
an  an 1   a0 xt  an s n X  s   an 1s n 1 X  s   a0 X  s   initial consitions involving x  t 
dt n
dt n 1
0
d f t 
m
d f t 
m 1
 bm s m F  s   bm1s m1F  s   b0 F  s   initial consitions involving f t   E.2 
 bm  bm 1    b0 f t  E.1
dt m dt m 1 0
If all initial conditions are assumed to be zero, Eq. (E.2) reduces to the
Where x(t) is the output, f(t) is the input, t is the time, and ai’s and bi’s following form:
are constants. Find the transfer function of the system and show the
input, system, and output in a block diagram. a s
n
n
 
 an1s n1  a0 X s   bm s m  bm1s m1  b0 F s   E.3
3.12 Transfer-Function Approach Notes for transfer function:
Example 3.13
Transfer Function Corresponding to a Differential Equation
Solution
Transfer function of the system evaluated at zero initial conditions is
L[output ] X  s
T  s  ︳Zero initial conditions   X s  T s F s  E.5 
L[input ] F  s
By taking the inverse Laplace transform of Eq. (E.5), we can find the
output of the system in the time domain for any known input.

X ( s )  T ( s )F ( s )

3.13 3.13 Solutions Using Laplace Transforms

Solutions Using Laplace Transforms
The computation of responses of single-DOF systems using Laplace
transforms.
Example 3.15 Response of a Damped System Using Laplace

Transforms
Derive an expression for the complete response of a damped single-
degree-of-freedom system subjected to a general force, f(t), as shown
3.13
in Fig. 3.1 using Laplace transforms.
3.13 Solutions Using Laplace Transforms 3.13 Solutions Using Laplace Transforms
Example 3.15
Example 3.15
Response of a Damped System Using Laplace Transforms
Solution
Solution
The inverse Laplace transform of Fi s  will be equal to the known forcing
The Laplace transform of Eq. (3.1) leads to the relation (see Eq. (E.4) of Ex. 3.14) function
(ms 2  cs  k ) X ( s )  F ( s )  (ms  s ) x(0)  mx(0) 
f i t   F0 cos wt and f s t  
1 wnt
e sin wd t where wd  1   2 wn
F s  s  2wn mwd
X s   x0  2 x 0 E.1
1

 
m s 2  2wn s  wn2 s 2  2wn s  wn2 s  2wn s  wn2 (See Appendix D)
Inverse Laplace transform of the first term on the right-hand side of Eq. (E.1)
For convenience, the subscripts I and s denoting the input and system,
can be expressed by Convolution integral
Fi  s   F  s  ( E .2) t t
1
L1[ Fi  s  FS  s ]   f   f  t    d  mw  f  e sin wd  t    d  E.7 
 wn  t  

1
Fs  s    E.3  i s i

m s 2  2 wn s  wn2  r 0 d r 0

3.13 Solutions Using Laplace Transforms 3.13 Solutions Using Laplace Transforms
Example 3.15
Example 3.15
Solution
Solution
The inverse Laplace transform of the coefficient of x(0) can be obtained by
multiplying fs(t) by m so that
The inverse Laplace transform of the coefficient of x(0) in Eq. (E.1)
yields 1 1  wn
s  2 wn w L1  e sin wd t (E.10)
L1 2  n e  wn t cos  wd t  1   E.8  s 2  2 wn s  wn2 wd
s  2 wn s  wn2 wd
The complete response of the system can also be expressed as

 wn 
where 1  tan -1  tan 1  E.9 
wd 1 2 1
t
wn  wn
x t   f  e sin wd  t    d  cos  wd t  1  x(0)
 wn  t  
i e
mwd r 0
wd
1  wn
 e sin wd t x(0)  E.11
wd
Example 3.16 (自行推導一次)

Noting that the inverse Laplace transform of the product function in Eq.
(E.7) can also be expressed as
t t
1
L1[ Fi  s  FS  s ]   f  t    f  t  d  mw  f  t   e
i s

i
 wn
sin wd d  E.12 
r 0 d r 0
Convolution integral
t t
L1[G  s  R  s ]   g  t    r   d   g   r  t    d
r 0 r 0
The complete response of the system can also be expressed as
t
1 wn  wn
x t   f  e sin wd  t    d  cos  wd t  1  x(0)
  wn  t  
i e
mwd r 0
wd
1  wn
 e sin wd t x(0)  E.13 
wd (部份分式展開)

Problem 3.99 Derive Eqs. (E.4)-(E.7) in Ex. 3.16

3.14 3.14 Frequency Transfer Functions

Frequency Transfer Functions
• The steady-state response of a linear system subjected to a
sinusoidal (or harmonic) input will also be sinusoidal (or harmonic)
of the same frequency.
• Although the response is at the same frequency as the input, it

differs in amplitude and phase angle from the input. These
difference are functions of the frequency (see Fig. 3.11)
3.14
• Any sinusoid (正弦曲線) can be represented as a complex number
(called phasor). The magnitude of the complex number is the
amplitude of the sinusoid, and the angle of the complex number is
the phase angle of the sinusoid. Thus, the input phasor, M i sin( t  i )
i
, can be denoted in polar form as M i e i , where the frequency,, is
considered to be implicit (不明確的).
M i ( )e i i ( )


• Because a system causes both the amplitude and phase angle of
the input to be changed, we can think of representing the system
itself by a complex number or function defined such that the output
phasor is given by the product of the system function and the input
phasor.
• See Fig. 3.36, the input-output relation can be shown in the form of
a block diagram. Thus, the steady-state output or response sinusoid
of the system can be denoted as
Mo ( )e i0 ( )  M s ( )e i s ( ) M i ( )e i i ( )  M s ( ) M i ( )e i { s ( ) i ( )} (3.120)
Where M’s denote the amplitudes and ’s indicate the phase angles,
Fig. 3.11 variation of X and  with frequency ratio r respectively, of the sinusoids.

i s ( )
The system function, M s ( )e ,
is defined by its magnitude
i s ( )
• The system function, M s ( )e , is called the frequency-
M ( ) response function, with Ms() being called the amplitude or
M s ( )  o (3.121)
M i ( ) magnitude frequency response and s() the phase-response
function of the system.
And the phase by • The frequency-response function is sometimes called the
frequency transfer function, denote T(i), so that
s ( )  o ( )  i ( ) (3.122)
T (i )  M s ( )e i s ( ) (3.123)
The frequency-response magnitude
is given by the ratio of the output
sinusoid’s magnitude to the input • Relation between the general transfer function T(s) and the
sinusoid’s magnitude. The phase frequency transfer function T(i):
response is given by the difference
in phase angles between the T ( s)  T (i ) if substituting s  i
output and the input sinusoids.

Example 3.17
1

M i ( )
 i
M i ( )  (k  m 2 )2  ( c )2
Representation of frequency-response characteristics
(As shown like in Fig. 3.37)

Example 3.18
1
增加一積分器( s )，相位減90o
1
故雙積分器( s 2 )，相位為減180o
1 3.15 Examples using MATLAB

增加一積分器( s )，相位會減90o
1
故雙積分器( s 2 )，相位將減180o
• To practice by yourself from Ex. 3.19 to Ex.3.22

• The source codes of all MATLAB programs are given
at the companion website
Homework:
The Problems 3.103, 3.105, 3.110 are assigned as

homework, the due date of hand over is 03/Dec./2012
(Mon.).
1. Paper report (must add annotation)

Fig. 3.37 Bode diagram 2. Program

Slide For Vibration

Uploaded by

Copyright:

Available Formats

Slide For Vibration

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Slide For Vibration

Uploaded by

Copyright:

Available Formats

C. R. Yang, NTNU MT C. R.

台灣師範大學機電科技學系 -1- 台灣師範大學機電科技學系 -2-

C. R. Yang, NTNU MT C. R. Yang, NTNU MT

台灣師範大學機電科技學系 -3- 台灣師範大學機電科技學系 -4-

Chapter Outline 3.1

C. R. Yang, NTNU MT C. R. Yang, NTNU MT

3.1 Introduction 3.1 Introduction

3.2 3.2 Equation of Motion

C. R. Yang, NTNU MT C. R. Yang, NTNU MT

3.2 Equation of Motion 3.3

Xh(t) dies out and x(t)

Homogenous and general solutions of Eq. (3.1) for an underdamped case

C. R. Yang, NTNU MT C. R. Yang, NTNU MT

C. R. Yang, NTNU MT C. R. Yang, NTNU MT

3.3 Response of an Undamped System Under

The total response if the system at resonance is rewrite Eq. (3.9) as

台灣師範大學機電科技學系 -19- 台灣師範大學機電科技學系 -20-

x0  stn t 3.3 Response of an Undamped System Under

C. R. Yang, NTNU MT C. R. Yang, NTNU MT

3.3 Response of an Undamped System Under

The complete motion can be expressed as the sum of two cosine

台灣師範大學機電科技學系 -23- 台灣師範大學機電科技學系 -24-

3.3 Response of an Undamped System Under

台灣師範大學機電科技學系 -25- 台灣師範大學機電科技學系 -26-

C. R. Yang, NTNU MT C. R. Yang, NTNU MT

The frequency of beating defined as

台灣師範大學機電科技學系 -27- 台灣師範大學機電科技學系 -28-

台灣師範大學機電科技學系 -29- 台灣師範大學機電科技學系 -30-

C. R. Yang, NTNU MT C. R. Yang, NTNU MT

3.4 3.4 Response of Damped System Under Harmonic

•The equation of motion becomes mx  cx  kx  F0 cos t (3.24)

台灣師範大學機電科技學系 -31- 台灣師範大學機電科技學系 -32-

C. R. Yang, NTNU MT C. R. Yang, NTNU MT

3.4 Response of Damped System Under Harmonic

 1  2    2. Any amount of damping (  0) reduces the magnification factor

台灣師範大學機電科技學系 -37- 台灣師範大學機電科技學系 -38-

C. R. Yang, NTNU MT C. R. Yang, NTNU MT

台灣師範大學機電科技學系 -39- 台灣師範大學機電科技學系 -40-

  implying that the response lags the excitation.

台灣師範大學機電科技學系 -41- 台灣師範大學機電科技學系 -42-

C. R. Yang, NTNU MT C. R. Yang, NTNU MT

x(t )  X 0e nt cos(d t  0 )  X cos(t   ) (3.35) ( where d  1   2 n )

3.4 Response of Damped System Under Harmonic

a. An external force F (t )  F0 cos t acts on the system with

台灣師範大學機電科技學系 -45- 台灣師範大學機電科技學系 -46-

C. R. Yang, NTNU MT C. R. Yang, NTNU MT

 X 0 cos 0  0.023186 (E.3)

(0.03326)(10) sin(3.814075) (E.4)

X 0 sin 0  0.002268 (E.5) x(t )  X 0e nt cos(d t  0 ) (E.8)

X 0   x02   n 0    0.012      0.010012 (E.9)

台灣師範大學機電科技學系 -49- 台灣師範大學機電科技學系 -50-

C. R. Yang, NTNU MT C. R. Yang, NTNU MT

3.4 Response of Damped System Under Harmonic

The value of the amplitude ratio at resonance

台灣師範大學機電科技學系 -53- 台灣師範大學機電科技學系 -54-

C. R. Yang, NTNU MT C. R. Yang, NTNU MT

• Substituting Eq. (3.48) into Eq. (3.47),  c 

台灣師範大學機電科技學系 -55- 台灣師範大學機電科技學系 -56-