International Journal of Mechanical Sciences 53 (2011) 505512

Contents lists available at ScienceDirect

International Journal of Mechanical Sciences

journal homepage: www.elsevier.com/locate/ijmecsci

Modied liquid column damper for vibration control of structures

K.A. Al-Saif a,b,n, K.A. Aldakkan b, M.A. Foda a
a
b

Mechanical Engineering Department, College of Engineering, King Saud University, P. O. Box 800, Riyadh 11421, Saudi Arabia
King Abdulaziz City for Science and Technology, National Robotics and Intelligent Systems Center, P. O. Box 6080, Riyadh 11422, Saudi Arabia

a r t i c l e i n f o

abstract

Article history:
Received 13 May 2010
Received in revised form
13 April 2011
Accepted 18 April 2011
Available online 28 April 2011

A modied version of the traditional tuned liquid column damper (TLCD) absorber is proposed as a
passive vibration control device for structures vibrating at low frequencies. This new version, denoted
as tuned liquid column ball damper (TLCBD), is equipped with a coated steel ball, in place of the orice
in TLCD, immersed inside the horizontal column of the damper. The current study examines the
performance of TLCBD for a harmonic excitation which is a simplied model for the vortex shedding
forces on structures in the cross wind direction. A parametric study to investigate the effect of the ball
size and absorber mass on the suppression capacity is carried out. The absorber damping characteristics
is identied experimentally using a single point laser vibrometer system and the measured damping
factor is used in the mathematical model. Intensive numerical simulations were conduced and the
results are compared with the traditional TLCD with optimum parameters. The results revealed an
improvement of the vibration suppression capability of the proposed version that exceeds around 66%
reduction.
& 2011 Elsevier Ltd. All rights reserved.

Keywords:
Liquid column ball damper
Vibration control
Passive control

1. Introduction
The use of light-weight, high strength materials and advanced
construction techniques in the erection of high rise buildings and
towers have led to increasingly exible and lightly damped
structures. These structures are very sensitive to environmental
excitations such as wind and earthquakes, which cause unwanted
vibrations resulting in occupant discomfort and possible structural
failure. Hence it has become important to search for effective and
practical devices for suppression of these vibrations. Among these
devices is the tuned liquid column damper (TLCD), which initially
was proposed by Sakai and his co-authors [1]. It consists of a
U-shaped tube, lled with liquid, preferably water. At the center of
the horizontal section of the tube an orice exists, which causes
energy dissipation in the vibration of the liquid. Since the TLCD
was introduced, many analytical and experimental works are
conducted aiming to assess its effectiveness and to nd its optimal
design parameters. In addition, many improvement ideas for the
TLCDs have been proposed. Felixa et al. [2] motivated by the works
of Yalla and Kareem [3,4], presented a numerical analysis of a
vibration control liquid column damper that was mounted on a
structural frame under excitation of an unbalanced DC motor.
Ghosh and Basu [5] proposed a modied version of the LCD by
n
Corresponding author at: Mechanical Engineering Department, College of
Engineering, King Saud University, P. O. Box 800, Riyadh 11421, Saudi Arabia.
Tel.: 966504488828; fax: 96614676652.
E-mail address: alsaif@ksu.edu.sa (K.A. Al-Saif).

0020-7403/$ - see front matter & 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijmecsci.2011.04.007

connecting the LCD to the primary structure through springdamper system as a passive vibration control of stiff structures
subjected to earthquake excitations. Haroun and Pires [6] have
introduced a hybrid liquid column damper which works by
maintaining an optimal damping condition using a variable orice
in the tuned liquid column damper. Yalla et al. [7] utilized a semiactive suppression system that consists of an electro-pneumatic
actuator driving a control valve to change the cross section of a
tuned liquid damper, thus adjusting the properties of the damper.
Ying et al. [8] developed a semi-active optimal control method for
non-linear multi-degree-of-freedom systems based on the dynamical programming principle, statistical linearization method and
variational principle. This method was applied to a tall building
structure with magneto-rheological-tuned liquid column damper
(MR-TLCD) for random wind response reduction. Wang et al. [9]
devised a semi-active tuned liquid column damper with the use of
magneto-rheological uid (MR) to mitigate vibration of a tall
building. The MR uid can reversibly change from a free-owing
linear viscous uid to a semi-solid when exposed to a magnetic
eld. Therefore the TLCD has alterable uid viscosity which results
in adjustable and control damping forces. Lee et al. [10] investigated the vibration mitigation of a tension leg offshore platform
system when incorporated with the TLCD device and subjected to
surge wave motion. Chaiviriyawong et al. [11] simulated the
induced velocity distribution of the uid inside a liquid column
damper using the numerical potential-ow method aiming to
estimate the effective length of liquid dampers, hence the natural
frequency. The experimental and theoretical works conducted by

506

K.A. Al-Saif et al. / International Journal of Mechanical Sciences 53 (2011) 505512

Nomenclature
A
A0
Ab
b
d
deq
cs
ct
fext
F
g
h
Jc
ks
L
mb

cross-sectional area of the U-tube

orice area of TLCD
main cross sectional of the ball
length of the horizontal section
diameter of the all
equivalent damping coefcient
viscous damping coefcient of the main structure
equivalent damping coefcient of the LCBD
external excitation
amplitude of the excitation of the force
gravitational acceleration
undisturbed length of the liquid in the vertical
column
mass moment inertia of the ball about center of mass
fundamental mode stiffness of the structure
total length of the liquid in the tube
mass of the ball

Xu et al. [12], and Haroun et al. [13] studied the use of the TLCD in
reducing lateral vibrations of a structure. Xue et al. [14] and Wu
et al. [15] showed that the TLCD could reduce the torsional
vibration of a structure effectively under harmonic excitation.
Multiple tuned liquid column dampers (MTLCD), which consists of
a series of TLCDs has been studied by Gao et al. [16] for reducing
lateral vibration and by Shum and Xu [17] and Xu and Shum
[18]for reducing torsional vibration of a structure.
Another variation of TLCD called a liquid column variable
absorber (LCVA) proposed by Watikns [19]. The major difference
between TLCD and LCVA is that the cross-section of the later is
not uniform. Many studies on the performance of LCVA were
carried out by many researchers. Among them were Hitchcock
et al. [20,21], Watkins and Hitchcock [22] and Chang and Hus
[23]. Gao and Kwok [24] found that the increase of the cross
section of horizontal tube might reduce the liquid column in a
vertical tube and the optimal parameters could be obtained to
reduce the amplication factor when the structure subjected to a
harmonic vibration. A recent study by Shum [25] proposed the
optimal parameters of a TLCD to suppress the vibration caused by
harmonic excitations. In his paper, a closed form solution for the
case of undamped primary structure was developed and numerical technique was used for the case of damped primary structure.
In the current investigation, a modied version of TLCD, which
includes a rolling ball inside the horizontal section of the damper
is investigated. It is believed that the ball will act as a moving
orice and will disturb the ow in a manner as to improve the
absorber attenuation performance. A damped single degree of
freedom structural model is used to simulate the vibrating
structure and the LCBD is attached to the top plate of the model.
The governing equations of motion are derived using Lagranges
equations. The viscosity of the uid is included in the model which
contributes to the damping term associated with ball movement
inside the horizontal section of the tube. The governing equations
are solved numerically and the performance of LCBD is compared
with that of TLCD. The optimum design parameters for LCD
provided by Wu et al. [26] are utilized in this comparison. In
addition, other optimal design parameters which are taken from
Shum [25] are used to evaluate the performance of the proposed
system and to conrm its effectiveness. Excellent results are
obtained using the proposed absorber with suppression magnitude exceeding 66% as opposed to the current optimum designs of
the TLCD reported in the literature for harmonic excitations.

mf
ms
Qi
R
Rb
t
xb
xs
y

mass of the uid

fundamental mass of the main structure
ith generalized forces
ball-tube diameter ratio
radius of the ball
time
displacement of the ball
displacement of the main structure
displacement of the surface of the uid
coefcient of absolute viscosity
ratio of the horizontal length to total length of the
liquid in the tube
uid density
natural frequency of the TLCD
natural frequency of the main structure
external excitation
mass ratio
optimal frequency tuning
head loss coefcient

n
a
r
o0
os
oex
m
l
d

The rest of the paper is organized as follows: In Section 2, the

equations of motions are derived. Section 3 presents the experimental work to estimate the equivalent damping parameter of
the proposed absorber. In Section 4, numerical simulations are
presented to address the effects of varying systems parameters
and to assess the effectiveness of the absorber. In addition, a
design guide lines for the implementation of the present absorber
is also given. Some general conclusions are given in Section 5.

2. Mathematical modeling
The system shown in Fig. 1 illustrates a structure with
attached U-tube that has a cross sectional area A and is lled
with a uid of density r and absolute viscosity u. The horizontal
section of the U-tube is equipped with a ball of radius Rb. Fig. 2
shows the same structure with attached traditional TLCD. The
following formulation will cover both systems. The kinetic energy
of the entire system in Fig. 1 can be expressed as
T Tfluid Tball Tstructure ,

where
Tfluid

1
2

Z
V1

ry_ 2 x_ 2s dV

1
2

_ 2 dV
rx_ s y

V2

1
2

Z
V3

ry_ 2 x_ 2s dV,
2

where V1 is the volume of the uid in the left column, V2 is the

volume of the uid in the horizontal section, while V3 is volume of

y
Xb

ks
Ms

F(t)
X

Cs
Fig. 1. Structure with a proposed liquid column ball damper (LCBD).

K.A. Al-Saif et al. / International Journal of Mechanical Sciences 53 (2011) 505512

b
Q2

y
y

orifice

ks
Ms

F(t)
X

Cs
Fig. 2. Structure with a conventional liquid column damper (TLCD).

1
_ 2,
rbAx_ s y
2

where h is undistributed length of the liquid in the vertical

column, and b is the length of the horizontal section of the U-tube.
With the assumption of no slipping, the kinetic energy of the
ball can be expressed as
Tball

1
1 Jc
m x_ 2
x_ x_ s 2 ,
2 b b 2 Rb 2 b

where mb is the mass of the ball, Rb its radius, xb is displacement

of its center of mass, and Jc is its mass moment of inertia about its
center of mass (Jc 2=5mb R2b ).
The kinetic energy of the structure is
1
ms x_ 2s ,
2

Ts

where ms is the fundamental modal mass of the structure.

The substitution of Eqs. (3)(5) into Eq. (1) leads to the kinetic
energy of the entire system
T

1
1
1 Jc
1
_ 2:
ms x_ 2s mb x_ 2b
x_ x_ s 2 rAhy_ 2 x_ 2s rbAx_ s y
2
2
2 Rb 2 b
2

The potential energy of the uid can be expressed as

Z
Z
gzrdV
gzrdV:
Uf
V1

for the case of TLCD,

: ct y,
_

for the case of LCBD,

for the case of LCBD,

11

12

where fext is the external excitation force acting on the structure,

which is assumed to be harmonic of amplitude F and excitation
frequency oex, cs is viscous damping coefcient of the structural
system, d is head loss coefcient due to the orice which can be
determined experimentally, Ao is orice area for TLCD, Ab is main
cross-sectional area of the ball, ct the equivalent damping of the
LCBD (obtained experimentally) and deq is the equivalent damping coefcient which can be represented as (White [27], p. 482)
deq 3pdn,

the uid in the right column of the U-tube, xs is the displacement

of the structure, and y is the displacement of surface of the uid.
Performing the integration leads to,
Tfluid rhAy_ 2 x_ 2s

8


< rgA djy_ jy_  1 rAo d9y9
_ y,
_
2
2

_ 2rAb yg,
Q3 deq x_ b y

507

13

where n is the coefcient of absolute viscosity and d is the

diameter of the ball.
It should be noted that the damping term in the case of LCBD is
considered to be linear as suggested by the experimental observation as discussed in Section 3.
The equations of motion of the system can be derived from the
following Lagranges equations


d @TU
@
TU Qi , i 1,2,3
14

dt
@q_ i
@qi
with q1 xs, q2 y and q3 xb and t is the time
Invoking Eq. (14), the equations of motion of the structure
with LCBD can be expressed as
!
Jc
Jc

ms mf 2 x s cs x_ s ks xs amf y
15
x b fext t,
Rb
Rb 2
mf y ct y_ 2rgAy amf x s 0,
mb

Jc
Rb 2

!
x b deq x_ b

Jc
Rb 2

x s deq y_ 2yrgAb :

16

17

The equations of motion of the structure with LCD are given by

ms mf x s cs x_ s ks xs amf y fext t,

18

1
_ y_ 2rgAy amf x s 0,
rAo d9y9
2

19

mf y

where mf rAL is mass of the uid with L2hb is the total

length of the liquid in the tube and a b/L (length ratio).

3. Estimation of the equivalent damping coefcient for the

LCBD

V3

Performing the integration one obtains

Uf

1
1
rgAhy2 rgAh y2 ,
2
2

where g is gravitational acceleration. Including the potential

energy of the structure and knowing that the potential energy
of the ball is zero, the total potential energy of the entire system is
U

1
ks x2s rAgh2 y2 ,
2

where ks is the fundamental modal stiffness of the structure.

To use Lagranges equations, one needs to obtain the generalized forces, which are those not already contained in the potential
energy U (i.e. damping and external forces). Therefore, the
generalized forces associated with the coordinates xs, y and xb;
respectively, can be expressed as;
Q1 fext tcs x_ s ,

10

The experimental setup is shown in Fig. 3. The length of the

tube L equals 0.5 m, h0.12 m and its inside diameter is 0.044 m
is lled with water of assumed viscosity n 10  3 N s/m. Six balls
with different ball-tube diameter ratio R, which is dened as the
ball diameter divided by the inside diameter of the horizontal
tube, are used. The free vibration response of the system is
measured. The TLCBD is given an initial pulse using electrodynamic shaker and the response of the water level in the vertical
section of the U-tube is measured using HeNe laser. The time
history of the U-tube liquid level response resembles that of the
linear single degree of freedom system with an exponential decay.
This explains the linear assumption used for the damping term in
Eq. (11) for the TLCBD case. From the time history signal the
natural frequency can be obtained using the FFT algorithm. The
damping coefcient is obtained according to the logarithmic
decrement technique for each ball diameter. Fig. 4 shows the
natural frequency of the TLCBD versus the ball-tube diameter

508

K.A. Al-Saif et al. / International Journal of Mechanical Sciences 53 (2011) 505512

Data acquisition
system
16-bit PCMCIA
A/D converter
Laptop computer
Time and frequency
signal processing

He-Ne laser
transducer

Signal processing
unitVelocity and
displacement of
the water level
calculations

Liquid damper
column with moving
ball

Pulse input

Fig. 3. Experimental set up to measure the damping ratio and natural frequency of LCBD.

1.5

Table 1
Damping ratio for several ball-tube diameter ratios

1.4
1.3
1.2
1.1
1

Ball-tube diameter ratio R

Damping ratio (xt)

0.90
0.79
0.68
0.56
0.45
0.34

0.081
0.046
0.044
0.042
0.039
0.039

0.9
0.8
Table 2
System parameters for two optimum designs of LCD.

0.7
0.6

Parameter

TLCD case I
Ref. [26]

TLCD case II
Ref. [25]

Mass ratio m mf /ms

Length ratio a b/(b 2h)
Primary structure fundamental frequency,
os (Hz)
Optimum frequency tuning ratio,
l ot /os
Primary structure damping ratio, zs
Optimum head loss coefcient, d

0.01
0.8
0.93

0.05
0.7
0.93

0.986

0.946

0.03
21.9

0.05
140

0.5
0.4

0.5

0.6

0.7 0.8 0.9

1
Ball-Tube Ratio R

1.1

1.2

1.3

Fig. 4. Normalized LCBD free oscillations frequency versus ball-tube diameter

ratio R.

ratio. It can be seen from the gure that the oscillation frequency
of the uid inside the TLCBD is almost independent of the ball size
until the ball diameter approaches the internal tube diameter;
R 1, where it starts to decrease slightly. However, one can
assume that the natural frequency of the TLCBD is close to the
corresponding TLCD so that
the natural frequency of TLCBD can
p
be calculated from ot 2g=L o0 ; where o0 is the TLCD
natural frequency.
The damping ratio of the TLCBD is measured for the six
different cases (6 balls) and the results are listed in Table 1. As
the ball diameter increases the damping also increases, since the
opening between the ball and the internal cross-sectional area of
the tube will be smaller and the ow would experience a greater
resistance. In the numerical simulation of the TLCBD system, the
damping ratio zt corresponding to each ball diameter will be
utilized. The value of the damping coefcient ct can be calculated
from the knowledge of the damping ratio zt ct =cc , where the

critical damping coefcient cc 2mf ot . Therefore

ct 2r2h bAot zt :

20

4. Numerical results and discussion

Equations of motion are solved numerically using the Rung
Kutta method. In order to compare the performance of the
proposed LCBD, two optimum cases of TLCD are considered. The
optimum parameters of both cases are listed in Table 2. In the
rst case the data is taken from Wu et al. [26]. For the second
case, the optimal design parameters are taken from Shum [25].

K.A. Al-Saif et al. / International Journal of Mechanical Sciences 53 (2011) 505512

These parameters are used in the numeral integration of the

equations of motion.
The corresponding design parameters (frequency ratio l, mass
ratio m and length ratio a) used for TLCBD absorber are the same
as the values used for the TLCD. Fig. 5 depicts the normalized
frequency responses of the structure with attached TLCBD and
with attached TLCD for case I. The ball-tube diameter ratio R,
which is used in the simulation, equals to 0.822. It is observed
that that the performance of LCBD is very favorable within the
resonance region; however, both responses are almost similar
outside this region. The time history of the structure response at
excitation frequency ratio oex/os 1 is shown in Fig. 6 when the
TLCD is attached. While for the same system with TLCBD, the
structural time history is illustrated in Fig. 7.
Fig. 8 exhibits the normalized frequency responses of the
structure with LCBD and with TLCD for case II, where the
optimum parameters as suggested by Shum [25] are utilized.
The ball-tube diameter ratio R is taken as same as in case I. The
gure reveals the good vibration reduction performance of the
TLCBD in the resonance region. It should be noted that the
performance of the TLCBD can be improved further by adjusting

1
0.8
0.6
0.4
0.2
0
-0.2
-0.4
-0.6
-0.8
-1
0

10

15

20
25
Time, sec

30

35

40

Fig. 7. Time history of the response of the structure with LCBD when oex/os 1
and R 0.822; for case I.

1
1

509

LCD
LCBD

Xs without Ball
Xs with Ball

0.8
Xs / Xmax

0.8

0.6

0.6
0.4

0.4
0.2
0.2

0
0.2

0.4

0.6

0
0.2

0.4

0.6

0.8

1.2

0.8
ex / s

1.2

1.4

1.4
Fig. 8. Normalized frequency responses of the structure with LCBD and with TLCD
when os 0.93 Hz, R 0.822 and zs 0.05; for case II.

Fig. 5. Normalized frequency responses of the structure with LCBD and with TLCD
when os 0.93 Hz, R 0.822 and zs 0.03; for case I.

1
0.8
0.6

1
0.8
0.6
0.4
0.2
0
-0.2
-0.4
-0.6
-0.8
-1

0.4
0.2
0
-0.2
-0.4

-0.6

10

15

20

25

30

35

40

Time, sec

-0.8

Fig. 9. Time history of the response of the structure with attached LCD when
oex/os 0.96; for case II.

-1
0

10

15

20
25
Time, sec

30

35

40

Fig. 6. Time history of the response of the structure with TLCD when oex/os 1;
for case I.

the optimum ball-tube diameter ratio which will be discussed in

details in Section 4.3.
The time response of the structure at the frequency ratio
equals 0.96 is shown in Fig. 9 with TLCD and in Fig.10 with TLCBD.

510

K.A. Al-Saif et al. / International Journal of Mechanical Sciences 53 (2011) 505512

120

1
0.8

100

0.6

LCD

0.4
80

0.2
0

60

-0.2
-0.4

40

-0.6

LCBD

-0.8
20

-1
0

10

15

20

25

30

35

40

Time, sec

Fig. 10. Time history of the response of the structure with LCBD when
oex/os 0.96 and R 0.822; for case II.

Table 3
Percent improvement of the LCBD with R 0.822 over the TLCD.
Frequency
ratio

0.90
0.93
0.95
0.97
0.98
0.99
1.00
1.01
1.02
1.03
1.04

Structure
response
ratio with TLCD;
Shum [25]

Structure
response
ratio with
TLCBD

Percent (%)
improvement

0.59
0.73
0.89
1.00
1.00
0.94
0.87
0.78
0.69
0.62
0.55

0.46
0.38
0.32
0.32
0.33
0.35
0.37
0.40
0.43
0.46
0.49

22.15
48.36
63.57
68.33
67.09
63.32
57.06
48.47
37.71
24.84
11.51

0.5

1.5

2.5
x 10-4

Normalized Input Force

Fig. 11. Effect of the input force on the response of the structure with LCBD and
with LCD; for case II.

To conrm this behavior, intensive numerical experiments were

conducted for various system parameters such as the structure
frequency, damping ratio and ball diameters. The simulation
results showed similar trends.
The qualitative measure of the performance of the LCBD is
listed in Table 3 for case II. It can be readily seen that there is an
additional reduction of 22% when oex/os 0.9 and up to 67% when
oex/os 0.98 with extended bandwidth over the optimized TLCD.
The physical explanation of the LCBD model works better than
TLCD is simply because part of the input energy to the system is
transformed to the translational and angular kinetic energies of
the ball in addition to the part of energy that dissipated as heat
due to friction between ball and uid (included in the damping
term in the equations of motion). Referring to Den Hartogs
dynamic vibration absorber theory, the excitation force that acts
on the primary mass (main structure) results in vibrations of the
main mass and the two auxiliary masses that represented by
oscillating mass of the uid and the mass of the oscillating ball.
Therefore, effectively, we have two dynamic vibrations absorbers
added to the main system in the case of TLCBD.
4.1. Effect of excitation amplitude ratio
Fig. 11 depicts the effect of the applied force magnitude on the
response of the structure with attached LCBD and with attached
TLCD for case II, with parameters listed in Table 2. A signicant
difference in the performance of the two absorbers is clearly

LCD

5
LCBD

3
0

0.01

0.02

0.03 0.04 0.05

Mass Ratio

0.06

0.07

0.08

Fig.12. Responses of the structure with LCBD and with TLCD versus mass ratio for
Case II.

manifested as the applied load is increased. However for a small

input force ratio, the difference in performance gets smaller. It is
also observed that the trend between the applied force amplitude
and the response amplitude is linear as clearly shown in the
gure. It should be mentioned that the input force is taken as a
fraction of the structure weight (i.e. 0.0005) and it is normalized
by the weight of the structure ms g.
4.2. Effect of mass ratio on the steady state response of the structure
Fig. 12 shows the effect of the mass ratio on the absorber
performance. It can be readily seen that the greater the mass ratio

K.A. Al-Saif et al. / International Journal of Mechanical Sciences 53 (2011) 505512

the better the performance, which is similar to the behavior of

standard dynamic vibration absorber. It is also noticed that, the
difference between the two curves widens as the mass ratio
assumes greater values and the response reduction with TLCBD is
much greater than that with TLCD. However increasing the mass
ratio would entail heavy absorber and high cost. Therefore a
practical mass ratio should be selected depending on the application and the weight of the structure.

4.3. Effect of ball-tube diameter ratio on the response

The effect of the ball diameter is investigated for different
system parameters such as the primary structure damping ratio,
fundamental natural frequency and input excitation amplitude.
The optimum design of the conventional TLCD based on Shum
parameters [25], (case II) is used as a case study. The steady state
response of the structure for several ball-tube diameter ratios is
calculated and the results are plotted in dimensionless form as
shown in Fig. 13. It should be mentioned that similar behavior as
depicted in Fig. 13 is obtained for different primary structure
natural frequency, damping ratio, and input excitations. The
gure reveals that there is an optimum ball diameter ratio,
R0.8, at which the response of the structure becomes minimum.
This nding is in line with other studies for the orice type
absorber where the obstruction ratio of the orice (orice area/
pipe cross-sectional area) has a specic optimum value at which
the response of the structure reaches a minimum value.
For the structural engineer to utilize the proposed concept the
following procedure is suggested:
(1) Select the appropriate mass ratio (0.0050.05).
(2) Set the frequency ratio based on the length ratio of the LCBD
and the mass ratio m (mass of the liquid in LCBD /fundamental
modal mass of the structure) from Refs. [25,26].
(3) Find the LCBD frequency based on the structure fundamental
frequency.
p
(4) Obtain the liquid column length from the relation ot 2g=L.
(5) Find the cross sectional area of the absorber and obtain the
internal diameter of the absorber tube.
(6) From the optimum ball-tube diameter ratio R 0.8, nd the
required ball diameter. If the calculated diameter is large,
then more than one absorber can be implemented.

0.34
Fitted curve
0.32
0.3
0.28
0.26
0.24
0.22
0.4

0.5
0.6
0.7
0.8
Ball-Tube Diameter Ratio

0.9

Fig. 13. Steady sate response of the structure for different values of R, zs, F.

511

Before closing this section, it should be noted that the

calculation made in this study is based on harmonic loading
which is a simplied model for the vortex shedding forces on
structures in the cross wind direction.

5. Conclusions
In this study, a new concept of the tuned liquid column
damper is proposed. The mathematical model of the absorber
attached to a single degree of freedom structure is developed. The
viscosity of the uid used is included in the model. A comparative
study between the responses of a structure with attached TLCBD
and with attached TLCD is conducted. Two sets of optimum
design parameters for TLCD that are available in the literature
are used in the comparison. It is shown that the performance of
the proposed absorber is excellent with up to 67% vibration
reduction, within the resonance region, of the primary structure
as opposed to the traditional TLCD with optimal design parameters. The main parameters that inuence the performance of
the proposed absorber are the ball-tube diameter ratio R and the
mass ratio m. It was shown that the performance of the proposed
LCBD can be improved by selecting the optimum ball-tube
diameter ratio R and by increasing the mass ratio to an acceptable
and practical limit.

References
[1] Sakai F, Takaeda S, Tamaki T. Tuned liquid column damper-new type device
for suppression of building vibration. In: Proceedings of the International
Conference on High Rise Buildings, Nanjing, China; 1989:92631.
[2] Felixa Jorge LP, Balthazar Jose M, Brasilc Reyolando MLRF. On tuned liquid
column dampers mounted on a structural frame under a non-ideal excitation.
Journal of Sound and Vibration 2000;282:128592.
[3] Yalla SK, Kareem A. Optimum absorbers parameters for tuned liquid column
dampers. Journal of Structural Engineering 2000;125:90615.
[4] Yalla SK, Kareem A. Beat phenomenon in combined structureliquid damper
systems. Engineering Structure 2001;125:62230.
[5] Ghosh A, Basu B. Seismic vibration control of short period structures using
the liquid column damper. Engineering Structures 2004;26:190513.
[6] Haroun. MA, Pires JA. Active dampers, Proc orice control in hybrid liquid
column, 1st World Conf. on Structural Control Vol. In: Los Angeles I, Lewis FL,
Syrmos VL, editors. Optimal Control. New York: Wiley; 1995.
[7] Yalla S, Kareem A, Kantor JC. Semi-active tuned liquid column dampers for
vibration control of structures. Engineering Structures 2001;23:146979.
[8] Ying ZG, Ni YQ, Ko JM. Semi-active optimal control of linearized systems with
multi-degree of freedom and application. Journal of Sound and Vibration
2005;279:37388.
[9] Wang JY, Ni YQ, Ko JM, Spencer Jr. BF. Magneto-rheological tuned liquid
column dampers (MR-TLCDs) for vibration mitigation of tall buildings:
modeling and analysis of open-loop control. Computer and Structures
2005;83:202334.
[10] Lee HH, Wong S-H, Lee R-S. Response mitigation on the offshore oating
platform system with tuned liquid column damper. Ocean Engineering
2006;33:111842.
[11] Chaiviriyawong P, Webster WC, Pinkaew T, Lukkunaprasit P. Simulation of
characteristics of tuned liquid column damper using a potential-ow
method. Engineering Structures 2007;29:13244.
[12] Xu YL, Samali B, Kwo KCS. Control of along-wind response of structures by
mass and liquid dampers. Journal of Engineering Mechanics ASCE118
1992:2039.
[13] Haroun MA, Pires JA, Won AYJ. Suppression of environmental induced
vibrations in tall buildings by hybrid liquid column dampers. The Structural
Design of Tall Buildings 1996:4554.
[14] Xue SD, Ko JM, Xu YL. Tuned liquid column damper for suppressing pitching
motion of structures. Engineering Structures 2000;22:153851.
[15] Wu JC, Wang YP, Lee CL, Liao H, Chen YH. Wind-induced interaction of a nonuniform tuned liquid column damper and a structure in pitching motion.
Engineering Structures 2008;30:355565.
[16] Gao H, Kwok KSC, Samali B. Characteristics of multiple tuned liquid column
dampers in suppressing structural vibration. Engineering Structures
1999;21:31631.
[17] Shum KM, Xu YL. Multiple tuned liquid column dampers for torsional
vibration control of structures: experimental investigation. Earthquake
Engineering and Structural Dynamics 2002;31:97791.

512

K.A. Al-Saif et al. / International Journal of Mechanical Sciences 53 (2011) 505512

[18] Xu YL, Shum KM. Multiple tuned liquid column dampers for torsional
vibration control of structures: theoretical investigation. Earthquake Engineering and Structural Dynamics 2003;32:30928.
[19] Watkins RD. Tests on a liquid column vibration absorber for tall structures.
In: Lee SL, Shanmugam NE, editors. Proceedings of the International Conference on Steel and Aluminum Structures; 1991. p. 81826.
[20] Hitchcock PA, Kwok KCS, Watkins RD, Samali B. Characteristics of liquid
column vibration absorbers (LCVA)-I. Engineering Structures 1997;19:12634.
[21] Hitchcock PA, Kwok KCS, Watkins RD, Samali B. Characteristics of liquid column
vibration absorbers (LCVA)-II. Engineering Structures 1997;19:13544.
[22] Watkins RD, Hitchcock PA. Tests on various liquid column vibration absorbers. In: Proceedings of International Conference on Motion and Vibration
Control, Yokohama; 1992:11304.

[23] Chang CC, Hsu CT. Control performance of liquid column vibration absorbers.
Engineering Structures 1998;20:5806.
[24] Gao H, Kwok KCS, Samali B. Optimization of tuned liquid column dampers.
Engineering Structures 1997;19:47686.
[25] Shum KM. Closed form optimal solution of a tuned liquid column damper for
suppressing harmonic vibration of structures. Engineering Structures
2009;31:8492.
[26] Wu J-C, Shih M-H, Lin Y-Yi, Shen Y-C. Design guidelines for tuned liquid
column damper for structures responding to wind. Engineering Structures
2005;27:1893905.
[27] White FM, editor. New York: McGraw-Hill; 2003. p. 482.