Journal of Constructional Steel Research 65 (2009) 260268

Contents lists available at ScienceDirect

Journal of Constructional Steel Research

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

Evaluation of yielding shear panel device for passive energy dissipation

Ricky W.K. Chan a,b , Faris Albermani b, , Martin S. Williams c
a City University of Hong Kong, Hong Kong
b University of Queensland, Australia
c University of Oxford, UK

article

info

Article history:
Received 17 October 2007
Accepted 25 March 2008
Keywords:
Yielding shear panel
Energy dissipation
Seismic retrofitting

a b s t r a c t
The paper describes an experimental investigation of a new earthquake damper, the yielding shear panel
device (YSPD), for civil structures. It utilizes energy dissipation through plastic shear deformation of a thin
diaphragm steel plate welded inside a square hollow section (SHS). Its performance is verified by nineteen
monotonic and cyclic tests. Experiments showed that certain specimens exhibited stable behavior and
were capable of dissipating a significant amount of energy. The performance is influenced by the
diaphragm plate slenderness and by the in-plane rigidity of the surrounding SHS. Slender plates undergo
elastoplastic shear buckling and exhibit stable though slightly pinched hysteresis response. Stocky plates
impose high deformation demand on the surrounding SHS that hinders their cyclic performance. The
equivalent viscous damping offered by the test specimens, on their own, and the cumulative energy
dissipation are quantified. Fabrication, implementation and replacement of the damper proved to be easy
and inexpensive. The YSPD offers a potentially viable alternative for seismic retrofitting of existing frame
structures.
2008 Elsevier Ltd. All rights reserved.

1. Introduction
Interest in the development of passive energy dissipation in
earthquake risk mitigation of civil structures has greatly increased
in the last two decades. [1,2] During an earthquake, a large
amount of energy is imparted to a structure. The traditional
design approach relies on the energy dissipation as a consequence
of inelastic deformation of particular structural zones. The
permanent damage of the post-disaster structure is often so
serious that it would be expensive to repair, if at all possible. The
concept of passive energy dissipation, however, attempts to reduce
such permanent damage to the structure. With designated energy
dissipative devices installed within a structure, a portion of the
input seismic energy could be diverted into these devices; as a
result damage of the parent structure can be effectively reduced.
The inclusion of dissipative devices in a structure is expected to
alter its stiffness and damping and hence influence its structural
response [3]. In addition, by strategically locating these devices,
repair and/or replacement of the devices following an earthquake
can be carried out with minimal interruption to occupancy, a
crucial benefit to building owners and occupants.
A number of dissipative devices utilizing plastic deformation of
metals have been proposed. Devices which make use of flexural

Corresponding author. Tel.: +61 7 3365 4126; fax: +61 7 3365 4599.
E-mail address: f.albermani@uq.edu.au (F. Albermani).
0143-974X/$ see front matter 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jcsr.2008.03.017

deformation of metals include the patented ADAS [4], its variants

TADAS [5] and Cu-ADAS [6] and the Steel Slit Damper (SSD) [7]. The
Buckling-restrained brace (BRB) [8], on the other hand, makes use
of the axial deformation of steel.
Devices such as the ADAS, SSD and the proposed YSPD are
usually envisaged to be connected between the top of an inverted
V-brace (chevron brace) system and a floor beam in a structural
panel (Fig. 1). This results in the device being connected in series
to the bracing system. The resultant in-plane lateral stiffness of
the brace-device assembly kbd , can be obtained from the individual
stiffness of the brace, kb , and the device, kd ,
kbd =

1
1
kb

1
kd

kb kd
kb + kd

(1)

Eq. (1) indicates that the brace stiffness is compromised by the

insertion of a flexible damper. If the brace stiffness is required to
withstand in-service lateral loads, a relatively high device-to-brace
stiffness is then necessary. Dampers which rely on plastic flexural
deformation are generally flexible; hence multiple plates are used
to build up the required stiffness.
On the other hand, the in-plane shear strength and stiffness of
steel plates has been used by designers as a primary lateral load
resistant system. The steel plate shear wall (SPSW) is one such
application that has been used in Europe, North America and Japan.
The SPSW provides high in-plane stiffness to resist lateral loads
and represents an alternative to conventional reinforced concrete
walls. Thin steel plates buckle elastically at a rather low level

R.W.K. Chan et al. / Journal of Constructional Steel Research 65 (2009) 260268

261

Fig. 1. Frame-brace-device assembly.

of shear stress; however plate buckling is not synonymous with

failure if the plate is adequately supported along its boundaries.
Basler [9] demonstrated that in the presence of tension field action,
steel plates offer substantially higher strength and ductility than
concrete walls. The cyclic behavior of steel plate shear wall panels
has been investigated by a number of researchers [1013].
Generally, a good metallic device for seismic applications
must exhibit: (1) adequate elastic stiffness to withstand inservice lateral load (e.g. wind); (2) a yield strength of the
damper exceeding the expected in-service lateral loads; (3)
large energy dissipative capability; and (4) a stable hysteretic
forcedisplacement response which can be modeled numerically.
To utilize the high in-plane stiffness and energy dissipative
capability of shearing actions in steel plates, a new metallic
damper, the Yielding Shear Panel Device (YSPD) has been proposed
by Schmidt et al. [14] and further developed by Williams and
Albermani [15]. They performed quasi-static tests on a half-scaled
moment resisting frame equipped with a YSPD and showed that
a considerable amount of energy was dissipated through the
damper [15].
This paper concentrates on the performance of the YSPD
itself (isolated from the parent frame structure). A series of subassemblage tests on half-scaled YSPD specimens is conducted. A
specially designed test setup was developed for this purpose. A
total of 19 tests were conducted using various plate slenderness
and device configurations. These tests are presented and discussed
in this paper. In addition a simple analytical design method of the
device is presented together with a hysteretic model that can be
calibrated using the obtained experimental results.
2. Yielding shear panel device (YSPD)
Figs. 2 and Fig. 8 show the yielding shear panel device tested in
this research. It was fabricated using a short segment of a square
hollow section (SHS, dimension D DxT ) with a steel diaphragm
plate (thickness t) welded inside it. The square section was chosen
in this study since it was expected to be more effective than a
rectangular section in developing 45 degree tension field.
The length of the SHS section was chosen to equal its width
(i.e. D). The relative horizontal displacement between the top and
bottom connections causes the diaphragm plate to deform in shear
(Fig. 2(c)). When the displacement is sufficiently large the plate
deforms plastically, and as a result input energy is dissipated.
Fabrication and installation of the device is simple and inexpensive.

Fig. 2. Yielding shear panel device (YSPD) (a) Elevation (b) Top view (c) Deformed
shape.

where G is shear modulus and t is the thickness of the diaphragm

plate. For a compact diaphragm plate the yield strength can be
taken as the shear yield strength (assuming a von Mises yield
criterion);
fy
Fy = td

(3)

where d is the width of steel plate and fy is its tensile yield stress.
Consequently, the yield displacement of the device is,
Fy

fy d

.
(4)
3G
For a device with a slender diaphragm plate, elastic shear buckling
will take place. The critical shear stress for a simply supported plate
is given by,
 2
2 E
t
cr = ks
(5)
2
12(1 ) d
where ks depends on the aspect ratio of the plate, and equals 9.35
for square plates. E and are Youngs modulus and Poissons ratio
respectively. Taking = 0.3 and E = 205 GPa, the limiting plate
slenderness ratio at which buckling occurs is
q
d/t = 1732/ fy
(6)
uy =

kd

where fy is the tensile yield strength of the diaphragm plate (MPa).

4. Test program
Performance of passive energy devices is often influenced
by factors such as connection details, the surrounding SHS and
possible fabrication flaws. Therefore, it is essential to conduct
a testing program to verify the cyclic performance and energy
dissipation capacity of the proposed YSPD.

3. Preliminary design of YSPD

4.1. Specimens

Assuming a minor contribution from SHS, the theoretical elastic

in-plane lateral stiffness of the device kd is given by,

Twelve specimens similar to Fig. 2 were fabricated at the

structures laboratory of the City University of Hong Kong. Two
different SHS sections (100 100 4 and 120 120 5) and three
diaphragm plate thicknesses (2, 3, and 4 mm) were used, resulting

kd = Gtd/d = Gt

(2)

R.W.K. Chan et al. / Journal of Constructional Steel Research 65 (2009) 260268

262
Table 1
Material properties for test specimens
Elements

Measured thickness (mm)

Tensile yield strength (N/mm2 )

2 mm PL
3 mm PL
4 mm PL
100 100 4
120 120 5

1.86
2.83
3.78
3.76
4.91

211.3
321.3
351.2
414.9
333.3

Table 2
Test specimen details
Test no.

Specimen

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

100-0M
100-2M
100-3M
100-4M
100-0C
100-2C
100-3C
100-4C
100-2CS
100-3CS
100-4CS
120-0M
120-2M
120-3M
120-4M
120-0C
120-2C
120-3C
120-4C

SHS

Test regime

Monotonic
100 100 4

Cyclic

Monotonic
120 120 5
Cyclic

Plate slenderness

49.5
32.5
24.3

49.5
32.5
24.3
49.5
32.5
24.3

59.1
38.9
29.1

59.1
38.9
29.1

in six different diaphragm plate-to-SHS combinations. Slenderness

ratios of all diaphragm plates are lower than the elastic buckling
limit of Eq. (6). Four bolt holes spaced at s = 50 mm were drilled on
each of the two opposite SHS flanges for connecting the specimen
to the test setup (Fig. 2). The square diaphragm plates (d d) were
cut to slightly smaller than the internal dimensions of the SHS. The
four corners were coped, and then the plate was positioned inside
the SHS, spot-welded at four corners and subsequently welded
along the plates edges from both sides. The fillet welds leg size
was approximately 3 mm.
Experimental details are listed in Tables 1 and 2. The measured
plate and wall thicknesses are listed in Table 1 together with the
measured yield stress from coupon tests. The measured diaphragm
plate thicknesses are within 86%96% of the nominal thickness. The
following notation is adopted to identify different specimens used
in the test; D tM or D tC , where D indicates the size of the SHS
section (either 100 or 120 mm), t is the thickness of the diaphragm
plate used (either 2, 3 or 4 mm). A zero plate thickness indicates
no diaphragm plate is used inside the SHS. The letter M indicates a
monotonic test while C is a cyclic test. Three specimens are shown
as 100-2CS, 100-3CS and 100-4CS; here the S indicates a stiffened
section (Fig. 3).
Four control tests of SHS without a diaphragm plate were
conducted to identify the contribution of the SHS. Further, in order
to investigate the effect of stiffening of the SHS connecting flanges,
three additional specimens with stiffened flanges were fabricated.
In these specimens, two pieces of 19.5 mm thick mild steel plates
were welded to the two connected flanges of the SHS as shown in
Fig. 3.

Fig. 3. YSPD with stiffened flanges.

was applied by an MTS 100 kN capacity computer-controlled

actuator quasi-statically to the specimen via the L-beam. To ensure
the verticality of the applied load, a pantograph system was welded
to the right hand side of the L-beam. To prevent the L-beam
from deflecting out-of-plane, lateral supports (with rollers) were
initially provided (not shown for clarity). However, these supports
were later removed as it was noticed that the pantograph system
was adequate to prevent the L-girder from deflecting out-ofplane. The complete test setup rested on a reaction frame which
was significantly stiffer. The centerline of the actuator implied an
eccentricity, e, to the specimen. A free-run of the setup (i.e. without
the specimen installed) was performed, and the result showed that
effect of friction and gravity was negligible.
Following the ECCS recommendation for simulating ground
motion [16], a quasi-static loading history comprised of 3 repeated
cycles at 0.5, 1.0, 3.0, 5.0, 10.0 and 20.0 mm amplitudes (a
total of 18 cycles) as shown in Fig. 6 was adopted. A positive
amplitude indicates a downward movement of the actuator.
Based on previous test results conducted on a half-scaled braced
frame [15], a maximum amplitude of 20 mm of device deformation
corresponds to 2.67% drift factor which is more than the
permissible drift limit used in practice [14].
Displacements of the specimens were measured independently
by a set of LVDTs, marked as 1 through 3 in Fig. 4. While LVDT
1 measures the elastic deformation of the support, the difference
across LVDT 1 and 2 measured the absolute deformation of the test
specimen. With LVDT 3 and the distance between LVDT 2 and 3
measured, in-plane rotation of the L-beam could be monitored. For
monotonic tests, strain rosette gages were positioned at the center
of the diaphragm plates to monitor the actual strain behavior.
4.3. Test observations
The test setup was robust and repeatable, and no visible
damage occurred after all tests were carried out. The pantograph
system effectively restrained the force delivery vertically. In-plane
rotation of the L-beam were measured (LVDT 2 and 3) and it
was found to be less than one degree. Its effect is considered
negligible. The ground support deflection (LVDT 1) was elastic.
Average shear strain of specimens is defined as = /D, where
is the displacement difference across LVDT 1 and 2.

4.2. Test setup and instrumentation

Fig. 4 shows a schematic view and Fig. 5 an overview of the
experimental platform used in this study. The test specimens were
installed between a ground beam and an L-beam, securely fastened
by four M16 bolts (snug tight) on each side. Forced displacement

4.3.1. Monotonic tests

Monotonic tests were terminated when the actuator reached
an absolute displacement of 20 mm. Due to elastic deformation
of the experimental setup, the deformation of the test specimens
(difference across LVDT 1 and 2) is less than the imposed

R.W.K. Chan et al. / Journal of Constructional Steel Research 65 (2009) 260268

263

Fig. 4. Experimental setup.

Table 3
Summary of monotonic test results
Specimen

100-0M
100-2M
100-3M
100-4M
120-0M
120-2M
120-3M
120-4M

Experimental

Predicted

Initial stiffness (kN/mm)

Yield (kN)

Yield disp. (mm)

10

Initial stiffness (kN/mm)

(Eq. (2))

Yield (kN)
(Eq. (3))

Elastic buckling (kN)

(Eq. (5))

1.3
14.6
26.3
29.4
2.4
44.9
18.5
58.9

6.3
22.1
27.4
30.8
6.3
15.7
37.1
17.7

5.2
1.8
1.2
1.2
3.5
0.7
2.6
0.7

1.34
1.59
1.77
2.32
1.40
2.51
1.64
4.03

130.2
198.1
264.6

132.3
198.1
256.2

22.7
52.5
76.6

26.1
63.0
89.1

111.5
392.6
935.7

97.5
327.2
707.8

Fig. 5. Overview of test setup.

Fig. 6. Displacement history for cyclic tests.

displacement. Control tests of the SHS without diaphragm plates

indicated low stiffness and yield strengths (Table 3), suggesting
that the contribution of the SHS is minimal.
Normalized force-shear strain curves of monotonic tests are
shown in Fig. 7. Distinct yield points were not noticeable; instead
there was a smooth transition from the elastic to inelastic regime.
Yielding is taken at 0.2% strain offset and is shown in Table 3. An
over strength factor 10 , defined as the force sustained at a shear
strain = 10%, divided by the experimental yield strength is also
shown in Table 3. All the specimens offer an over strength factor much higher than 1. The nominated 10% strain corresponds to
1.5% inter-story drift factor which is consistent with practical limits [15].

The 2 mm specimens (100-2M and 120-2M) performed

similarly, with strength approaching 1.8Fy towards the end of
the tests. These two specimens, unlike other specimens with
thicker diaphragm plates, showed inelastic shear buckling (Fig. 8a).
Specimens with 3 and 4 mm diaphragm plates were not able
to reach a force higher than the theoretical yield strength (Eq.
(3)). A closer inspection of these specimens after test revealed
localized distortion of the SHS. Due to the high in-plane stiffness
of the thicker plates, the SHS sustained more deformation. The
deformation was mainly in the vicinity of bolt holes and at midlength of the SHS where the diaphragm plate is welded. The poor
performance of specimens with thicker diaphragm plates can be

R.W.K. Chan et al. / Journal of Constructional Steel Research 65 (2009) 260268

264
Table 4
Selected results of cyclic tests
Specimen

Elastic stiffness (kN/mm)

Yield (kN)

Pmax + (kN)

Pmax (kN)

10

Energy dissipated (kJ)

Buckled

100-2C
100-2CS
100-3CS
120-2C
120-3C

19.8
16.7
24.0
24.2
16.8

22.42
10.44
20.77
12.38
23.87

42.0
41.0
65.5
46.2
57.3

36.6
40.6
59.9
41.8
54.3

1.73
3.62
2.78
3.50
2.20

6.99
5.73
5.94
6.23
6.51

Y
Y
N
Y
Y

Fig. 8c. Specimens with stiffened flange.

Fig. 7. Monotonic test results.

Fig. 8a. Monotonic tests specimens.

tests, significant localized deformation was observed in these

specimens. It is believed that the local deformations of the SHS
have caused the poor performance of these specimens. The tests
of these two specimens were terminated after the 10 mm cycles.
Specimens 120-2C and 120-3C performed satisfactorily with fairly
stable and large forcedisplacement hysteresis. Both of these
specimens buckled due to their large plate slenderness. 120-4C did
not buckle but demonstrated poor behavior due to similar reasons
as discussed above. Specimens with stiffened flanges (Fig. 9(gi))
offered certain improvement when compared to their unstiffened
counterparts. The Bauschinger effect is noticeable, and on average,
the positive peak strength are 8.4% higher than negative peaks.
Table 4 lists additional observations obtained from testing the
specimens shown in Fig. 9. An over strength factor 10 is also listed
in Table 4 and is much higher than 1 for all the specimens.
5. Discussion

Fig. 8b. Cyclic tests specimens.

A total of 19 specimens with 6 different combinations

of diaphragm plate-to-SHS were experimentally tested under
monotonic and quasi-static loading regimes that simulate strong
ground motion. All tests achieved high over strength factor with a
shear strain in the range of 15%20%. All the specimens have passed
the tests without failure. Since our objective was to examine
different combinations of plate-SHS under specified loading, no
attempt was made to test any of these specimens to failure. We
are about to embark on another testing program to evaluate the
fatigue characteristics of this proposed device.

attributed to the relative flexibility of the connected flanges of the

SHS and possible bolt slippage.

5.1. Elastic stiffness and strength

4.3.2. Cyclic tests

Results of cyclic tests are shown in Fig. 9 (ai). Positive
force and displacement represent downward stroke. Similar to
the monotonic tests, yield points are not distinctly noticeable.
Specimen 100-2C exhibited reasonably stable hysteresis with
yield strength approximately predicted by the simple analytical
expression (Eq. (3)). Shear buckling was noticeable in the last cycle
(20 mm amplitude) and its effect can be observed from the slightly
pinched hysteresis near zero displacement. It corresponded to the
buckle forming in the reverse direction. Specimens with thicker
diaphragm plates (100-3C and 100-4C) did not show buckling, but
exhibited seriously pinched hysteresis loops. As in the monotonic

Based on the obtained experimental results (Tables 3 and 4),

it appears that experimental elastic stiffness of the specimens
is much lower than the simple analytical value of Eq. (2). A
comparison between analytical and experimental response for
monotonic test 100-2M is shown in Fig. 10 as an example. The
lower elastic stiffness obtained from the test can be attributed
to bolt slippage, geometric imperfection and deformation of the
SHS section. The bolt slippage in particular is expected to have
large influence since the theoretical yield displacement (Eq. (4))
is quite small ranging from 0.15 to 0.3 mm for specimen 100-2M
and 120-4M respectively. Such a small displacement could easily
be exceeded by bolt slippage alone.

R.W.K. Chan et al. / Journal of Constructional Steel Research 65 (2009) 260268

(a) 100-2C.

(b) 100-3C.

(c) 100-4C.

(d) 120-2C.

(e) 120-3C.

(f) 120-4C.

(g) 100-2CS.

(h) 100-3CS.

(i) 100-4CS.
Fig. 9. Forcedisplacement hysteresis of cyclic tests.

265

R.W.K. Chan et al. / Journal of Constructional Steel Research 65 (2009) 260268

266

Fig. 10. Comparison between experimental and predicted results for 100-2M.

Fig. 13. Cumulative energy dissipation of specimens.

Fig. 11. Maximum principal strain measured on diaphragm plates.

Fig. 14. Effective stiffness and energy dissipated in a cycle.

5.3. Energy dissipation

Fig. 12. Principal tensile strain angle.

All the specimens offered quite good over strength factors

which were generally much higher than 1. Although the simple
analytical model (Eqs. (2) and (3)) over-estimates the elastic
stiffness and under-estimate the capacity, it remains a useful tool
for preliminary selection of the device.
5.2. Principal strains
Principal strains in the plate diaphragms under monotonic
loading were obtained using rosette gages. Fig. 11 shows the
principal strain plotted against the normalized applied load for
various monotonic tests. From this figure it appears that nearly
all specimens start yielding at approximately 0.7F /Fy , with the
buckled specimens 100-2M and 120-2M offering the highest strain
hardening. Fig. 12 shows the measured direction of principal strain
at the center of the diaphragm plates with respect to the horizontal
direction. Specimen 100-2M clearly maintains the principal strain
close to 45 degrees in both pre- and post-buckling stages. This
observation indicates an approximate pure-shear condition was
set in the pre-buckling stage and a tension field was formed
diagonally within the square plate confined by the SHS in the postbuckling stage.

The cumulative energy dissipation of the cyclic test specimens

is plotted as a function of cumulative device displacement in
Fig. 13. Specimen 100-2C dissipated the largest amount of energy
(close to 7 kJ) among the tested combinations. This specimen also
offers the highest rate of dissipation. Specimen 120-3C and 120-2C
followed. Again, specimens with higher plate slenderness offer superior performance, and inelastic shear buckling did not cause instability of the device. Stiffened specimen 100-2CS dissipated the
least energy among the tests. The decrease in energy dissipation
may be attributed to an increase in eccentricity due to the stiffening plates. This increase in eccentricity between the centerline
of applied load and the centerline of the specimen will make the
stress field in the diaphragm plate deviate from a pure shear state.
This will require further investigation on the effect of eccentricity.
5.4. Equivalent device stiffness and damping
It is generally accepted that energy dissipated in cyclic straining
of metals is rate-independent. For practical use it is sometimes
preferable to express the device properties in an equivalent viscous
system. This is basically a single degree of freedom oscillator with
an equivalent stiffness keff defined as (see Fig. 14),
keff =

|Pmax | |Pmin |
.
|max | |min |

(7)

R.W.K. Chan et al. / Journal of Constructional Steel Research 65 (2009) 260268

267

Fig. 15. Effective damping ratios of selected specimens.

When using the experimental results (Fig. 9), the maximum

and minimum force, P, and displacement, , were obtained by
averaging the experimental results from three repeated cycles at
particular amplitude. Cycles at the largest four amplitudes (3, 5, 10
and 20 mm (Fig. 6)) were used since not much dissipation takes
place at the lower cycles.
The damping ratio for the equivalent system, eq can be
obtained by equating the measured energy dissipated per cycle (ED )
in the experiment to that of a viscously damped oscillator [17],

eq =

1 ED

(8)

4 ES0

where ES0 is the energy stored in an elastic spring with a stiffness

keff and displacement max .
Plots of equivalent device damping ratio versus device effective
stiffness are shown in Fig. 15 (for different loading cycles). Each
point represents a feasible stiffness and equivalent damping ratio
of the proposed device. Effective stiffness decreases as the device
undergoes larger displacement. It can be observed that equivalent
damping ratio is inversely related to the effective stiffness. At large
displacement range, the device provides a damping ratio in excess
of 30%. It must be noted that this damping ratio is related to the
device itself and does not represent the damping ratio that would
be achieved if such a device were fitted in a real structure. The
main purpose of evaluating this ratio is to have an indication of
the energy dissipation capability of the proposed device.
5.5. Hysteretic model for the YSPD
A single equation continuous Bouc-Wen model [18] is commonly used to model the hysteretic behavior of passive devices.
This model has been incorporated in commercial software for modeling supplementary energy dissipation.
The restoring force developed in the YSPD is expressed by,
P (t) =

Ke u(t) + (1 )Ke uy z(t)

(9)

where is the ratio of post-yield stiffness to the elastic stiffness; Ke

is the elastic stiffness of the device; u(t) is the displacement of the
device at time t; uy is the yield displacement of device; and z(t) is an
evolutionary variable defined by a first-order nonlinear differential
equation which possesses the hysteretic properties,
uy z(t) + |u(t)| z(t) |z(t)|

n1

+ u (t) |z(t)|n u (t) = 0

(10)

in which , and n are model parameters whose values control the

shape of the hysteresis loops. These parameters can be calibrated
using the obtained experimental results (Fig. 9).

Fig. 16. Bouc-Wen model of 100-2C.

Using the experimental results for specimen 100-2C, the

following parameters for the Bouc-Wen model can be obtained
Ke = 19 777 N/mm

= 0.035
uy = 1.61 mm

= 0.9
= 0.12
n = 1.

(11)

This model is compared with the experimental results in Fig. 16.

It is clear that a good agreement between the model and the
experimental results is achieved.
6. Conclusion
An experimental study on a metallic passive energy dissipative
device which utilizes plastic shear deformation of diaphragm
steel plate is presented. Nineteen tests comprised of monotonic
and cyclic load regimes were conducted. Devices with six
different diaphragm plate-to-SHS combinations and with various
diaphragm plate slenderness in the range of 2459 were tested.
Major findings are summarized as follows.
1. Devices with high diaphragm plate slenderness offered good
energy dissipation, strength and ductility. In particular, device
100-2C (slenderness ratio 49.5) offered the best performance in
terms of strength and energy dissipation.
2. Inelastic shear buckling was observed in devices with plate
slenderness greater than 38.9. Plate buckling is accompanied
by a slight pinch in the observed hysteresis but did not cause
overall instability of the device.
3. Devices with lower slenderness did not buckle but offered
unsatisfactory strength and energy dissipation due to localized
deformation of the SHS and possible bolt slippage.
4. The experimental initial stiffness of the specimens were
significantly lower than values predicted by the simple
analytical equation (Eq. (2)). This is thought to be mainly due
to bolt slippage in addition to geometric imperfection and
deformation of the SHS.
5. Stiffening of the SHS connecting flanges provided some beneficial performance in terms of the stability of the hysteresis. However, the resulting increase in eccentricity reduces the amount
of energy dissipation.

268

R.W.K. Chan et al. / Journal of Constructional Steel Research 65 (2009) 260268

6. The energy dissipation capability of the YSPD can be represented by an equivalent damping ratio. This damping ratio can
be in excess of 30% at large displacement range.
7. The YSPD response can be simulated using a Bouc-Wen
hysteretic model calibrated against the obtained experimental
results.
References
[1] Soong TT, Dargush GF. Passive energy dissipation systems in structural
engineering. John Wiley & Sons; 1997.
[2] Soong TT, Spencer Jr BF. Supplemental energy dissipation: State-of-the-art and
state-of-the-practice. Eng Struct 2002;24:24359.
[3] Whittaker AS, Bertero VV, Thompson CL, Alonso LJ. Seismic testing of steel
plate energy dissipation devices. Earthq Spectra 1991;7(4):563604.
[4] Bergman DM, Goel SC. Evaluation of cyclic testing of steel plate devices for
added damping and stiffness. Report no. UMCE87-10. Ann Arbor (MI, USA):
The University of Michigan; 1987.
[5] Tsai K, Chen H, Hong C, Su Y. Design of steel triangular plate energy absorbers
for seimic-resistant construction. Earthquake Spectra 1993;9(3):50528.
[6] De la Llera J, Esguerra C, Almazan JL. Earthquake behavior of structures with
copper energy dissipators. Earthq Eng Struct Dyn 2004;33:32958.
[7] Chan RWK, Albermani F. Experimental study of steel slit damper for passive
energy dissipation. Eng Struct 2007. doi:10.1016/j.engstruct.2007.07.005.

[8] Clark PW, Aiken ID, Tajirian F, Kasai K, Ko E, Kimura I. Design procedures
for buildings incorporating hysteretic damping devices. In: Proc. int. postSmiRT conf. seminar on seismic isolation, passive energy dissipation and active
control of vibrations of structures. 1999.
[9] Basler K. Strength of plate girders in shear. J Struct Div, ASCE 1961;87(7).
[10] Elgaaly M, Caccese V, Du C. Postbuckling behavior of steel-plate shear walls
under cyclic loads. J Struct Engrg 1993;119(2).
[11] Driver RG, Kulak GL, Kennedy DJK, Elwi AE. Cyclic test of four-story steel plate
shear wall. J Struct Engrg, ASCE 1998;124(2).
[12] Lubell AS, Prion HGL, Ventura CE, Rezai M. Unstiffened steel plate shear wall
performance under cyclic loading. J Struct Engrg, ASCE 2000;126(4).
[13] Vian D, Bruneau M. Testing of special LYS steel plate shear walls. Paper no. 978.
In: 13th WCEE. 2004.
[14] Schmidt K, Dorka UE, Taucer F, Magnonette G. Seismic retrofit of a steel
frame and an RC frame with HYDE systems. Institute for the Protection and
the Security of the Citizen European Laboratory for Structural Assessment.
European Commission Joint Research Centre; 2004.
[15] Williams MS, Albermani F. Monotonic and cyclic tests on shear diaphragm
dissipaters for steel frames. Adv Steel Constr 2006;2(1):121.
[16] European Convention for Construction of Steelworks. Recommended testing
procedure for assessing the behaviour of steel elements under cyclic loads.
Technical Committee 1. TWG 1.3. 1986.
[17] Chopra AK. Dynamics of structures: Theory and applications to earthquake
engineering. Englewood Cliffs (NJ): Prentice Hall; 1995.
[18] Wen YK. Method for random vibration of hysteretic systems. J Engr Mech
1976;102:24963.