414 CHAPTER 6

FIGURE 6.15 10
Frequency response curves: (a) M1,1
versus r1 for r2  0.5 and ␮  0.5. 9
(b) M2,1 versus r1 for r2  0.5 and 8
␮  0.5.
7
6

M11
5
4
3
2
1
0
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
r1
(a)

9
8
7
6
M21

5
4
3
2
1
0
0.2 0.4 0.6 0.8 1 1.2 1.4
r1
(b)

or in nondimensional form as
r1 2 r1 4 r1 2
1 + a b (1 + m)  a b (1 + m)2 + 2a b (m - 1) + 1
1
r1 = (6.89)
22 B r2 A r2 r2

6.11 DYNAMIC VIBRATION ABSORBERS

When the machine of Figure 6.17 is subject to a harmonic excitation at a frequency near its
natural frequency, large amplitude steady-state vibrations are a result. One remedy is to
change the properties of the system such that the natural frequency is away from the excita-
tion frequency. An alternate remedy is to add an auxiliary mass-spring system such that the
system has two natural frequencies both of which are away from the excitation frequency.

Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 Two Degree-of-Freedom Systems 415

10 FIGURE 6.16
Frequency response curves when
9 a force is applied to mass m1:
8 (a) M1,2 versus r1 for r2  0.5 and
␮  0.75. (b) M2,2 versus r1 for
7
r2  0.5 and ␮  0.75.
6
M11

5
4
3
2
F0 sinωt
1
0
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 k1
r1 m1 ω≈ m1
(a)

35 k1

30

25 FIGURE 6.17
Large amplitude steady-state
M22

20 vibrations occur when the

excitation frequency is close
15
to the natural frequency of
10 the machine.

0 F0 sinωt
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
r1
(b)
m1
A vibration absorber is the auxiliary system. The original machine is termed the pri-
mary system. The resulting two degree-of-freedom system is illustrated in Figure 6.18. This
is the configuration that was analyzed in Section 6.10, and its frequency response is k2
k1 k1
k 1X1 1 - r 22
= ` `
2 m2 2
(6.90)
F0 r 21r 22 - r 22 - (1 + m)r 21 + 1
The parameter ␻1,1 is the natural frequency of the primary system, and the parameter ␻2,2
is the natural frequency of the absorber if it were grounded (that is, directly connected to FIGURE 6.18
A vibration absorber is an aux-
the ground). The system composed of the primary system attached to the auxiliary system
iliary mass-spring system
is a two degree-of-freedom system with natural frequencies given by Equation (6.88). which is added to the primary
The steady-state amplitude of the absorber is given by system (the machine) to add
k 1X2 one degree of freedom to the
= ` 2 2 `
1
(6.91) system and change its natural
F0 r 1r 2 - r 22 - (1 + m)r 21 + 1 frequencies.

Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
416 CHAPTER 6

F0 sinωt The steady-state amplitude of the primary system is zero when the absorber is tuned
such that r2  1 or that
k 2 = m 2v2 (6.92)

When r2  1, the steady-state vibrations of the primary system are zero. Thus, the excitation
force is transmitted directly to the absorber system. Using the FBD of Figure 6.19, the
steady-state behavior of the auxiliary system is
F0 sinωt
F0
x 2(t) = - sin vt (6.93)
k2
kx2(t)
Hence, the steady-state amplitude of the absorber mass when it is tuned such that
k2  m22 is

F0
X2 = (6.94)
k2
FIGURE 6.19 The frequency response for the primary system as a function of r2 for ␻2,2  ␻ is
FBD of the primary system illustrated in Figure 6.20. Note that one of the system’s two natural frequencies is less
and the auxiliary system than the tuned frequency while the other is greater.
when the absorber is tuned
to the excitation frequency.
If the excitation speed varies slightly from the tuned speed, the larger the separation in
natural frequencies the smaller the steady-state amplitude of the primary system. Defining
v22
q = (6.95)
v11
the separation in natural frequencies is a function of ␮, as shown in Figure 6.21, and by
the equation
v22 - v21 = v21,1 2q 4(1 + m)2 + 2(m - 1)q 2 + 1 (6.96)

In situations where absorbers are employed, q ≈ 1. Setting q  1 in Equation (6.96) leads to

v22 - v21 = v21,1 2m(4 + m) (6.97)

2 2 2
The separation in natural frequencies is larger for larger ␮. For m = 0.25, v2 - v1 L v1,1.
The denominator in Equation (6.90) is positive for ␻ 6 ␻1 and ␻ 7 ␻2. It is nega-
tive in the range ␻1
␻
␻2. The numerator is positive for ␻
␻ 2,2 and negative other-
wise. When the ratio of the numerator to denominator is negative, the response of the
primary system is 180° out of phase with the excitation. When the denominator is negative,
the response of the auxiliary system is 180° with the excitation.
A dynamic vibration absorber is used to eliminate steady-state vibrations of a particle
where the absorber is attached if the natural frequency of the absorber is tuned to the
excitation frequency. The absorber has many applications in industrial processes. When
the absorber is used on a SDOF system, it converts the system to two degrees of freedom.
The following must be kept in mind when using an absorber:
• The steady-state amplitude of the primary system is zero when the auxiliary system
(the absorber) is tuned such that ␻2, 2  ␻.
• One of the natural frequencies of the resulting two degree-of-freedom system is less
than the tuned frequency, and one is higher than the tuned frequency. The lower

Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 Two Degree-of-Freedom Systems 417

7 FIGURE 6.20
(a) Frequency response curve for
6 primary system with absorber
tuned to frequency of excitation
5 and ␮  0.25. (b) Frequency
response of auxiliary system
under same conditions.
4
␮  0.3
M11

q  1.1
3

0
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
r2
(a)
15

10
M21

0
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
r2
(b)

2.00
1.75
1.50
1.25
ω /ω11

1.00
0.75
0.50 q = 1.0
q = 1.2
0.25 FIGURE 6.21
q = 0.8 Natural frequencies of
0.00 two degree-of-freedom
0 0.2 0.4 0.6 0.8 1 system as a function of
µ the mass ratio ␮.

Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
418 CHAPTER 6

natural frequency must be passed during start-up and stopping, leading to large-
amplitude vibrations during these transient periods.
• The steady-state vibrations of the primary system are eliminated only at a single
frequency. If the system operates over a wide range of frequencies, the steady-state
amplitudes at frequencies away from the tuned frequency may be large. An effective
operating range should be defined for each application by limiting the amplitude of
vibrations to an acceptable maximum.
• If the absorber is tuned to the excitation frequency and a given mass ratio ␮ is not to
be exceeded, the maximum value of the absorber stiffness is
k 2 max = mm 1v2 (6.98)
and the minimum steady-state amplitude of the absorber mass is
F0
X 2 min = (6.99)
mm 1v2

• The analysis is valid only for undamped systems. If damping is present either in the
primary system or in the absorber, it is not possible to eliminate steady-state vibrations
of the primary system.

EXAMPLE 6.15
A machine of mass 150 kg with a rotating unbalance of 0.5 kgm is paced at the midspan of
a 2-m-long simply supported beam. The machine operates at a speed of 1200 rpm. The
beam has an elastic modulus of 210
109 N/m2 and a cross-sectional moment of inertia
of 2.1
106 m4.
(a) What is the steady-state amplitude of the primary system without an absorber?
(b) Design the dynamic vibration absorber of minimum mass such that, when attached to
the midspan of the beam, the vibrations of the beam will cease and the steady-state
amplitude of the absorber will be less than 20 mm.
(c) What are the system’s natural frequencies when the absorber is in place?
(d) What is the effective operating range such that the midspan deflection does not exceed
5 mm when the absorber is in place?

SOLUTION
Modeling the vibrations of the machine on the beam using a SDOF system model and
ignoring the mass of the beam, the stiffness and natural frequency of the primary system
are calculated as
48EI 48(210 * 109 N/m2)(2.1 * 10-6 m4)
k1 = = = 2.65 * 106 N/m (a)
L3 (2 m)3
and
k1 2.65 * 106 N/m
v11 = = = 132.9 rad/s (b)
A m1 A 150 kg
The operating speed is

ba b = 125.7 rad/s
rad 1 min
v = (1200 rpm)a2p (c)
rev 60 s

Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 Two Degree-of-Freedom Systems 419

(a) Since the excitation speed is near the natural frequency of the primary system, it will
have large amplitude vibrations without an absorber. The frequency ratio is
v 125.7 rad/s
r = = = 0.945 (d)
v11 132.9 rad/s
Steady-state amplitude of the machine is
m 0e 0.5 kg # m (0.945)2
X1 = ¶(0.945, 0) = a b = 0.285 m (e)
m 150 kg 1 - (0.945)2

(b) Steady-state vibrations of the primary system are eliminated when the absorber is tuned
to the excitation frequency using
k2
v22 = = 125.7 rad/s (f)
A m2
Since the ratio of the absorber stiffness to absorber mass is fixed, the absorber with the
minimum mass is also the absorber with the minimum stiffness. The amplitude of the
absorber is to be limited to 20 mm, which from Equation (6.94) leads to
F0 F0 (0.5 kg # m)(125.7 rad/s)2
X2 = Q k2 Ú = = 3.95 * 105 N/m (g)
k2 X2 0.002 m
The minimum absorber stiffness is 3.95
105 N/m, leading to an absorber mass of
k2 3.95 * 105 N/m
m2 = = = 25 kg (h)
v222 (125.7 rad/s)2

(c) The natural frequencies of the two degree-of-freedom system are calculated from
25 kg
Equation (6.88) using m = 150 kg = 0.167 as

v1 = 105.8 rad/s v2 = 157.6 rad/s (i)

(d) The effective operating range is obtained by setting F0  0.5␻2 and using Equation (6.90).
The denominator is negative between the two natural frequencies, and the numerator is
positive for r2
1. Take away the absolute value symbol and set X1  0.005 m in this
case. Rearrange the equation to

v4 - 7.63 * 104v2 + 8.28 * 108 = 0 (j)

which (when solved for ␻) leads to a lower bound on the operating range of 114.8 rad/s.
For r2  1, set X1  0.005 m, leading to

v4 - 2.79 * 104v2 + 1.67 * 108 = 0 (k)

and a upper bound on the operating range of 138.5 rad/s. Thus, the effective operating
range is
114.8 rad/s 6 v 6 138.5 rad/s (l)

Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
420 CHAPTER 6

6.12 DAMPED VIBRATION ABSORBERS

Two problems exist when a vibration absorber is used. The lowest natural frequency of the
two degree-of-freedom system must be passed through in order to build up to the operating
speed. If the absorber is slightly mistuned, the vibration amplitude of the primary system
can be large. Perhaps the addition of damping to the absorber can help with these issues.
Consider the configuration of the system of Figure 6.22 in which viscous damping is
added in parallel with the stiffness in the auxiliary system. This is known as a damped
m1
vibration absorber. The steady-state amplitude of the primary system is given by
k 1X1
k2 c
= M 1d (r1, q, m, z)
F0
k1 k1
(2zr1q)2 + (r 21 - q 2)2
2 m2 2 = (6.100)
A {r 41 - [1 + (1 + m)q 2r 21] + q 2 }2 + (2zr1q)2[1 - r 21(1 + m)]2

The steady-state amplitude of the auxiliary system is

FIGURE 6.22
The auxiliary system of a k 1X2
= M 2d (r1, q, m, z)
damped vibration absorber F0
consists of a mass attached
to a spring in parallel with a q 4 + (2zq)2
= (6.101)
viscous damper. A {r 41 - [1 + (1 + m)q 2r 21] + q 2}2 + (2zr1q)2[1 - r 21(1 + m)]2

where
c
z = (6.102)
22m 2k 2
is the damping ratio of the auxiliary system if it were grounded. The nondimensional
steady-state amplitude of the primary system, given by Equation (6.100), is illustrated in
Figure 6.23 for ␮  0.25 and q  1 for several values of ␨. The steady-state amplitude of
the primary system is not zero for any value of r1. A minimum amplitude is reached for r1
near one between the peaks. The absorber was successful in significantly reducing the peak
near the second natural frequency, but not very successful in reducing the peak amplitude
near the first natural frequency. An investigation of the parameters affecting the damped
vibration absorber is necessary. It is noted that each curve, for different ␨, passes through
the same two points.
M1d is plotted in Figure 6.24 for ␮  0.25 and q  0.8. The peak at the lower reso-
nant frequency is smaller than the peak at the higher resonant frequency. However, the
higher peak occurs near r1  1, which is the region where an absorber is usually needed.
Also, the effective operating range is still small. It is noted again that there are two fixed
points through which each curve passes. These fixed points are different than those in
Figure 6.23.
Since it is not possible to eliminate steady-state motion of the original system when
damping is present, a damped vibration absorber must be designed to reduce the peak at the
lower resonant frequency and to widen the effective operating range. Absorbers using the
parameters used to generate Figure 6.23 and Figure 6.24 are not suitable for these purposes.

Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 Two Degree-of-Freedom Systems 421

10 FIGURE 6.23
ζ = 0.1 Response of primary system
ζ = 0.2 when a damped vibration
8 absorber is used with
ζ = 0.15
␮  0.25 and q  1 for
6 several values of ␨.
k1X1 /F0

0
0 0.5 1 1.5 2
r1

Widening the operating range requires that the two peaks have approximately the
same magnitude. Since the locations of the fixed points are dependent on q, it should be
possible to tune the absorber such that the values of M1d at the fixed points are the same.
Since curves for all values of ␨ pass through the fixed points, it should be possible to find
a value of ␨ such that the fixed points are near the peaks.
For fixed values of ␮ and q, there are two values of r1 which yield a value of M1d ,
independent of ␨. The value of M1d at these points is written as
A(m, q)z2 + B(m, q)
M 1d = (6.103)
A C(m, q)z2 + D(m, q)
Since Equation (6.103) holds for all ␨ and powers of ␨ are linearly independent,
A B
= (6.104)
C D
Using Equation (6.100) to determine the forms of A, B, C, and D, substituting into
Equation (6.104), and rearranging leads to

r 41 a1 + b - 31 + q 2(1 + m)4r 21 + q 2 = 0
m
(6.105)
2

ζ = 0.1
6 ζ = 0.2
ζ = 0.15

4
k1X1/F0

FIGURE 6.24
Response of primary system
0 when an optimum damped
0 0.5 1 1.5 2 vibration absorber is used
r1 with ␮  0.25 and q  0.8.

Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
422 CHAPTER 6

The solution of Equation (6.105) places the fixed points at

1 + (1 + m)q 2  21 - 2q 2 + (1 + m)2q 4
r1 = (6.106)
D 2 + m
Since Equation (6.103) yields the same value of M1d , independent of ␨ for r1 given by
Equation (6.106), letting z : q gives
1
A 31 - r 21(1 + m)42
M 1d = (6.107)

Requiring M1d to be the same at both fixed points leads to

1
q = (6.108)
1 + m
An optimum absorber could be designed with an appropriate value of ␨ such that the
smaller r1 given by Equation (6.106) corresponds to both a fixed point and a peak on the
frequency response curve. The appropriate value of ␨ is obtained by setting dM1d /d␨  0,
using q from Equation (6.108). The same procedure can be followed to yield the value of
␨ such that the larger value of r1 given by Equation (6.106) corresponds to both a fixed
point and a peak. Since the values of ␨ are not equal, their average is usually used to define
the optimum damping ratio
3m
zopt = (6.109)
A 8(1 + m)
In summary, the optimum design of a damped vibration absorber requires that the
absorber be tuned to the frequency calculated from Equation (6.108) with the damping
ratio of Equation (6.109). For ␮  0.25, Equation (6.109) gives an optimum damping
ratio of ␨  0.2379 and an optimum q  0.80. Figure 6.25 shows M1d for these values as
a function of r1. This figure also shows M1d for the same ␮ and ␨ but with values of q, one
on each side of the optimum. The curve corresponding to the optimum value of q has
smaller resonant peaks and the value of M1d does not vary much between the peaks.

q = 0.80
4 q = 0.90
q = 0.70
k1X1 /F0

FIGURE 6.25
Steady-state amplitude
of primary system for 0
0 0.5 1 1.5 2
␮  0.25, ␨opt  0.2739,
and qopt  0.80. r

Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 Two Degree-of-Freedom Systems 423

EXAMPLE 6.16
A milling machine has a mass of 250 kg and a natural frequency of 120 rad/s and is subject
to a harmonic excitation of magnitude 10,000 N at speeds between 95 rad/s and 120 rad/s.
Design a damped vibration absorber of mass 50 kg such that the steady-state amplitude is
no greater than 15 mm at all operating speeds.

SOLUTION
The mass ratio is
50 kg
m = = 0.2 (a)
250 kg
Since a wide operating range is required, the optimum absorber design is tried. From
Equations (6.108) and (6.109),
1 3(0.2)
q = = 0.833 z = = 0.25 (b)
1.2 A 8(1.2)
The steady-state amplitude at any operating speed for this absorber design is calculated by
Equations (6.100) and (6.101). The results are used to generate the frequency response
curve of Figure 6.26.
The fixed-points are calculated from Equation (6.106) as

1 + (1 + 0.2)(0.833)2  31 - 2(0.833)2 + (1 + 0.2)2(0.833)4

r1 =
D 2 + 0.2

= 0.7629, 1.0414 (c)

which leads to v = 91.5 rad/s, 125.0 rad/s.

Since the extremes of the operating range lie between the peaks and the steady-state
amplitudes at the extremes are
X (v = 95 rad/s) = 10.1 mm X(v = 120 rad/s) = 12.7 mm (d)

5
Absorber
No absorber
4

3
M

FIGURE 6.26
1
Frequency response for
primary system of
Example 6.16 with
optimum damped absorber
0.5 1 1.5 2 with ␮  0.25 attached.

Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
424 CHAPTER 6

and both are less than 15 mm, the optimum design is acceptable. The absorber stiffness
and damping ratio are calculated as
k 2 = m 2v222 = mq 2k 1 = (0.2)(0.833)2(3.6 * 106 N/m) = 5.08 * 105 N/m (e)

c = 2z 2k 2m 2 = 2500 N # s/m (f)

6.13 VIBRATION DAMPERS

A vibration damper is an auxiliary system composed of an inertia element and a viscous
damper that is connected to a primary system as a means of vibration control. Vibration
dampers are used in situations where vibration control is required over a range of frequencies.
The Houdaille damper of Figure 6.27 is an example of a vibration damper that is used
for vibration control of rotating devices such as engine crankshafts. The damper is inside a
casing that is attached to the end of the shaft. The casing contains a viscous fluid and a
mass that is free to rotate in the casing. The differential equations governing the motion of
the two degree-of-freedom torsional system are
$ #
c 1 dc$ d + c dc # d + c dc d = c 0 d
J 0 u1 c - c u1 k 0 u1 M sin vt
(6.110)
0 J2 u2 -c c u2 0 k u2 0
The steady-state amplitude of the primary system is obtained by the methods of
Section 6.10 as
M0 4z2 + r 2
®1 = (6.111)
k A 4z2(r 2 + mr 2 - 1)2 + (r 2 - 1)2r 2
v c J2
where r = z = m = (6.112)
k k J1
2J2
A J1 A J1
The optimum damping ratio is defined as the damping ratio for which the maximum value
of ® 1 is smallest. The peak amplitude, ® 1p(z) is the value of ® 1(rm) where rm is the value
of r that yields d ® 1>dr = 0. The optimum damping ratio is the value of ␨ such that
d ® 1p >dz = 0. Extensive algebra leads to
1
zopt = (6.113)
22(m + 1)(m + 2)

θ2
Inertia element
θ1 rotates in damper.
Damping provided
J2 by fluid.

FIGURE 6.27 J1, kt

Houdaille damper. ct

Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.