The 3rd International Conference on Earthquake Engineering and Disaster Mitigation 2016 (ICEEDM-III 2016)

Recent Developments on Seismic Performance of Steel Plate Shear Walls

Ronny Purbaa,*, Michel Bruneaub
a
Department of Civil Engineering, Universitas Bandar Lampung, Bandar Lampung, 35142, INDONESIA
b
Department of Civil, Structural and Environmental Engineering, University at Buffalo, Amherst, NY 14260, U.S.A

Abstract
Steel Plate Shear Walls (SPSWs) are rapidly becoming an appealing alternative lateral force resisting system for building
structures in high seismic areas. This paper presents results of recent research to expand the understanding of seismic
performance of SPSWs. Emphasis is on assessing the collapse potential of SPSWs having infill plates designed to resist
different percentages of the applied lateral loads. In this case, comparing the cases when neglecting and considering the
contribution of their boundary moment resisting frames to resist story shear forces. Seismic performance factors for both
types of SPSWs were identified and compared. Findings from these analyses demonstrate that the infill plates of SPSWs
should be designed to resist the total specified story shears, and that SPSWs designed by sharing those story shears
between the boundary frame and infill plates will undergo significantly larger and possibly unacceptable drifts. Results
of analytical and experimental research to investigate impact of in -span plastic hinge formation along the span of
horizontal boundary elements (HBEs) on the seismic behavior of SPSWs are also presented. This formation has
significant consequences, namely lower lateral strength due to partial yielding of the infill plates and significant plastic
incremental deformations on the HBEs.
Keywords: Steel plate shear walls; horizontal boundary elements; seismic performance; collapse potential; collapse fragility; in-span
plastic hinges; plastic incremental deformations.

1. INTRODUCTION

In seismic design applications, the primary energy dissipating elements of steel plate shear walls (SPSWs) resisting
lateral loads are their unstiffened infill plates (webs), which buckle in shear and form a series of diagonal tension field
actions (TFAs). From a capacity design perspective, the tension force from the infill plates must be resisted by the
surrounding horizontal and vertical boundary elements (HBEs and VBEs). When rigid connections are specified between
HBEs and VBEs, and between VBEs and the ground (as specified in many applications of SPSWs), SPSWs also benefit
from the moment resisting action of the boundary frame to resist the applied lateral loads. Nonetheless, it is specified in
the current Canadian Standard [1] for the design of steel structures that infill plates of SPSWs must be designed to resist
the entire lateral loads, without considering the possible contribution from the surrounding boundary moment resisting
frame. Such a statement is not explicitly included in the American Seismic Provisions [2], but one possible interpretation
of the design specifications could lead to the same design approach.

*
Corresponding author. Tel.: +62-721-789-825; Fax.: +62-721-770-261.
E-mail address: [email protected]
 As reported in past experiments, this overstrength in conventional SPSWs can be quite significant. For example,
Driver et al. [3] reported that boundary frame moment resisting action contributed about 25% of the global plastic
strength of their four-story SPSW specimen. Qu and Bruneau [4] demonstrated that boundary frame moment resisting
action can contribute up to 50% of the total strength of a SPSW with aspect ratio of 2.0 when its boundary elements are
designed per capacity design principles. This provides a significant incentive to reduce overstrength by explicitly
considering boundary frame moment resisting action as contributing to the SPSW overall lateral strength. However, the
consequences of reducing this overstrength are unknown, and opinions vary as to whether this should be permitted.
Moreover, the American Seismic Provisions [2] requires that HBEs and VBEs shall be designed to remain essentially
elastic under the maximum tension forces from the yielded infill plates, with the exception of plastic hinging at the ends
of HBEs. Implicitly, this indicates that in-span plastic hinges should be avoided. However, the provision does not specify
an analysis procedure to guarantee that this intent is met (although the commentary provides some guidance that could be
used for this purpose).
As a result, structural engineers might not anticipate that their designs may lead to in-span HBE plastic hinges (unless
these analyses are complemented by the use of nonlinear analysis programs to predict the plastic mechanism of
structures). In parallel, some structural engineers fully recognize the potential for in-span hinging to develop, but
question the merit of limiting the location of plastic hinges to only occur at the ends of HBEs because, in general, this
design requirement results in a relatively substantial size of boundary elements. Thus, to achieve more economical
designs, structural engineers may try to minimize overstrength by allowing plastic hinges to occur along HBE span as
this leads to relatively smaller boundary elements. Whether or not in-span hinging is acceptable has been a contentious
issue, particularly in the absence of factual data to support either position.
This paper presents recent investigations on the seismic performance of SPSWs to address the above two concerns,
namely: sharing of lateral loads between the boundary frame and infill plates, and impact of in-span HBE plastic hinging.
To investigate the first concern, this paper investigates collapse potential of SPSWs having infill plates designed to
sustain different percentages of the applied lateral loads. In this case, comparing the cases when neglecting and
considering the contribution of their boundary moment resisting frames to resist story shear forces. To address the second
concern, one must first determine whether in-span HBE hinging, when it happens, can impact in any way to the seismic
performance of SPSWs – irrespectively of whether it develops in a SPSW intentionally or as a result of unintended
design consequences. Here, analytical and experimental research were conducted to investigate the second concern.

2. ASSESSMENT OF COLLAPSE POTENTIAL

Using the FEMA P695 methodology [5], which defines the performance in terms of collapse potential under maximum
considered earthquake (MCE), the assessment was first conducted on SPSWs designed neglecting the contribution of
their boundary moment resisting frames to resist story shear forces (a.k.a. conventional design). Then, this assessment of
collapse potential was repeated for SPSWs designed considering the sharing of story shear forces between the boundary
frames and infill plates such that the sum of the strength of the two SPSW components was exactly equal to the required
strength to resist the designed lateral loads (a.k.a. balanced design).
Twelve SPSW archetypes with various structural configurations (i.e., panel aspect ratio, seismic weight intensity, and
number of story) were prepared. Their loading information, floor plans, and elevations were taken as similar to the SAC
model building [6]. Each SPSW archetype was designed to have one bay width, 13 ft story height, and low to moderate
aspect ratio (i.e., aspect ratio of either 1.0 or 2.0). All SPSWs had moment resisting HBE-to-VBE connections. Detail of
those archetypes can be found in [7]. Fig. 1 shows an example two-dimensional nonlinear model for collapse simulation
of 3-story SPSW archetypes developed in OpenSees [8] with the deterioration material models for SPSW components
(i.e., strips and boundary elements) and the gravity leaning column elements to capture the P- effects [9].

2.1. Collapse Performance Evaluation of SPSW Archetypes

The assessment started by conducting nonlinear pushover analysis and incremental dynamic analysis (IDA) on each
SPSW archetype. The former was performed to estimate system overstrength ( o) and period-based ductility ( T) factors.
The latter was performed to obtain collapse margin ratio (CMR) of which each archetype was subjected to 44 “Far-Field”
ground motions. Fig. 2 presents an example of IDA results for conventional and balanced three story SPSW archetypes. The
CMR values obtained from the IDA were then adjusted to consider frequency content of the selected ground motion
records (i.e., the effect of spectral shape). Spectral shape factor (SSF) values used to modify the CMR to the adjusted
collapse margin ratio (ACMR) are a function of the archetype fundamental period (T) and T factor obtained from the
pushover analysis. The resulting ACMR was compared to the acceptable ACMR for 10% collapse probability under
MCE ground motions (i.e., ACMR10%) of 2.16 for a total system collapse uncertainty ( TOT) of 0.6.

Strips

Boundary
Elements

Fig. 1. Nonlinear Model for Collapse Simulation

Fig. 2. Incremental Dynamic Analysis (IDA) Results: (a) SW320; (b) SW320K

Table 1 presents performance evaluation of each SPSW archetypes per the FEMA P695 methodology. Detail
information can be found in [7]. All conventional archetypes passed the performance criterion. The computed ACMR for
each archetype was larger than the acceptable ACMR10% of 2.16. These results indicate that each archetype has a
reasonable safety margin against collapse (i.e., a lower probability of collapse) as a result of the overstrength reserve
provided by the boundary frame. For this type of SPSW, results indicate that the R factor of 7 used in design is adequate
(i.e., satisfied the ACMR requirement). The o factor for the archetypes considered (based on the pushover analysis
results) varied from 2.3 to 3.1. Considering the limited numbers of SPSW archetypes designed in this research, the o
factor of 2.0 can be considered adequate for conventional SPSW. Assuming the inherent damping available in SPSW to
be 5% of critical damping, a Cd factor of 7 can be considered for conventional SPSWs. Note that the resulting seismic
performance factors for conventional SPSW obtained in this case are somewhat similar to those specified in the ASCE 7-
10 (i.e., R, o, and Cd factors are 7, 2, and 6, respectively).
For the balanced archetypes, except for the 10-story archetype and 5-story archetype design with high seismic weight
(i.e., SW1020K and SW520GK), all other archetypes did not meet the performance criterion because their computed
ACMR was smaller than ACMR10%. These results indicate that the R factor of 7 used in the initial step to design the
balanced SPSW would not lead to an adequate design (i.e., the resulting did not satisfy the ACMR requirement). Hence,
another 3-story balanced archetype with R factor of 6 was designed (i.e., denoted as SW320KR6). As shown in Table 1,
compared with the results for SW320K, SW320KR6 show a slight increase in the calculated ACMR. The calculated
ACMR of 2.06 is approximately 5% below the acceptable ACMR10% of 2.16. Although some could consider that
difference acceptable, to be rigorous, another design iteration was performed using an R factor of 5 (i.e., denoted as
SW320KR5). As hoped, SW320KR5 satisfied the performance criteria. Here, the calculated ACMR of 2.39 is 11%
higher than the threshold ACMR10% (Table 1).

Table 1. Performance Evaluation for SPSW Archetypes with Various Structural

Configurations

Pushover Results IDA Results Performance Evaluation

Archetype
ID Vd Vmax y,eff u 0= T= CMR = Pass/
ŜCT (g) SSF1 ACMR2
(kips) (kips) (in) (in) Vd/Vmax u/ y,eff ŜCT/SMT Fail3
SW310 155 401 2.1 11.7 2.6 5.5 3.14 2.10 1.26 2.64 Pass
SW320 176 495 1.8 8.9 2.8 4.9 3.60 2.40 1.25 3.00 Pass
SW320G 465 1440 1.8 9.9 3.1 5.5 4.08 2.72 1.26 3.43 Pass
SW520 255 578 3.9 16.3 2.3 4.2 3.40 2.42 1.25 3.03 Pass
SW520G 766 1924 4.1 19.5 2.5 4.8 4.26 3.03 1.27 3.85 Pass
SW1020 681 1975 7.8 40.6 2.9 5.2 3.40 4.08 1.36 5.58 Pass
SW310K 155 236 2.1 10.5 1.5 5.0 2.28 1.52 1.25 1.90 Fail
SW320K 176 226 1.8 8.6 1.3 4.8 2.29 1.53 1.24 1.90 Fail
SW320GK 465 618 1.7 8.9 1.3 5.1 2.32 1.55 1.25 1.93 Fail
SW520K 255 254 3.8 16.1 1.0 4.3 2.10 1.50 1.25 1.80 Fail
SW520GK 766 837 3.8 17.9 1.1 4.7 2.64 1.88 1.27 2.39 Pass
SW1020K 681 953 7.9 41.1 1.4 5.2 1.92 2.30 1.36 3.16 Pass
SW320KR6 205 270 1.7 8.6 1.3 5.0 2.47 1.65 1.25 2.06 Fail
SW320KR5 246 334 1.8 9.1 1.4 5.1 2.87 1.91 1.25 2.39 Pass
Note:
3)
1) SSF obtained from FEMA P695 table for a given T and T Acceptance criteria: ACMR10% for TOT of 0.6 = 2.16
2) ACMR = SSF (T, T) CMR Pass if ACMR ACMR10%, otherwise Fail
SMT = 1.5g, 1.4g, and 0.83g for 3-, 5-, and 10-story archetypes, respectively.

Based on the above results, seismic performance factors for SPSW designed with balanced are recommended to be
smaller compared to that for conventional SPSW (i.e., the 100% design case, ). Results above indicate that an R
factor of 5 should be used for the design of balanced SPSWs. No system overstrength factor is available in balanced
SPSWs (i.e., o = 1). Like for conventional SPSWs, the Cd factor for balanced SPSWs should be taken as similar to the
assigned R factor (i.e., Cd = 5.0).

2.2. Interstory Drift as Damage Measure (DM)

Considering the above results, it is also meaningful to interpret the IDA results in terms of drift demands.
Specifically, fragility curves can be constructed for the probability of exceeding certain drift values in terms of spectral
acceleration of the ground motions, for selected fixed values of interstory drifts up to the drift at the collapse. The
resulting “drift-exceedance” fragility curves for SW320 and SW320K, using interstory drifts as DMs, are plotted in Figs
3a and 3b, respectively. As a reference, the results using the collapse point as the DM are superimposed in these curves.
At the
MCE level (i.e., SMT = 1.5g), there is approximately a 50% probability that drifts will exceed 2% and 3.5% interstory
drifts for SW320 and SW320K, respectively. More significantly at a 20% probability of exceedance, the respective
archetypes will exceed 3% and 7% interstory drifts. The results indicate that SW320K has higher probability to suffer
significant larger interstory drift, which can be associated with larger structural and non-structural damages. The same
results were also obtained when comparing SW1020 and SW1020K in Figs. 3c and 3d.

Fig. 3. Exceedance Fragility Curves using Various Level of Inter-story Drift as Damage Measure:
(a) SW320; (b) SW320K; (c) SW1020; (d) SW1020K

It should be emphasized that even though the 10-story balanced archetype (i.e., SW1020K) had a calculated
ACMR that met the acceptable ACMR limit, its probability to undergo significantly large interstory drift (i.e., 3%) can
be as high as 50% under MCE ground motions (Fig. 3). While this SPSW designed with balanced case and R factor of 7
have sufficient margin to collapse, its ability to prevent damage to the structure and to drift-sensitive non-structural
components is significantly less than for its counterpart archetype (i.e., SW1020). Hence, the need to design balanced
archetypes with smaller R factor is deemed necessary.
In terms of the probability of exceeding the damage measures of 2, 3, and 4% interstory drift, results indicate that
reducing the R factor from 7 to 6 resulted in an improvement of exceedance probability of no more than 10% for
SW320KR6 compared to SW320K. More specifically, whereas half of the considered ground motions at the MCE lev
resulted in approximately 3.5% interstory drifts for SW320K, this slightly improved to 3.0% interstory drifts for
SW320KR6. Moreover, half of the considered ground motions at the MCE level caused approximately 2.5% maximum
interstory drifts for SW320KR5, which is tolerable and closer to what is expected for conventional SPSWs.
In terms of the total steel weight for archetypes designed with different R factors, the “reference” conventional SPSW
(i.e., SW320, designed per [2] with an R factor of 7) requires a total of 10,459 pounds of steel. The case designed with
balanced with R factor of 7, SW320K, requires a total of 5737 pounds of steel, which is approximately 55% less than what
is required for the conventional design, but, as indicated above, SW320K did not meet the collapse performance criterion
according to the FEMA P695 methodology and a lower R factor must be used. Designed with R factors of 6 and 5,
SW320KR6 and SW320KR5 require 17 and 31% more steel than SW320K, but SW320KR5 still provides a 28%
reduction in the total weight of steel from that is required for the conventional SPSW. However, that savings in steel
comes at the cost of the SPSW designed for balanced developing larger interstory drifts compared to the conventional
SPSWs (i.e., 2.5% versus 2.0% interstory drift) under MCE ground motions.

3. IMPACT OF IN-SPAN PLASTIC HINGE

As a case study to investigate the possible significance of in-span HBE plastic hinges, a three-story single-bay SPSW
was selected. Bay width and typical story height were arbitrarily chosen equal to 20 and 10 ft, respectively. It was also
assumed that the structure is located on Class D soil in downtown San Francisco, California and designed for an office
building. Total weight of the structure Wt was 1085 kips and the total base shear V resisted by the structure was 176 kips.
Two design procedures were applied to design the boundary elements: the Indirect Capacity Design approach [2] and the
capacity design approach which combines the procedure proposed by [10] for HBEs and that proposed by [11] for VBEs.
The resulting SPSWs obtained by the two different design procedures are denoted as SPSW-ID and SPSW-CD,
respectively. Detail of member sizes and strip models in SAP2000 used for this study can be found in [12].

3.1. Nonlinear Static Analysis (Pushover Analysis)

Obtained from the monotonic pushover results at 4% drift, the base shears were 311 and 477 kips for SPSW-ID and
SPSW-CD, respectively. For comparison, their respective theoretical values were 351 and 488 kips, obtained using the
plastic analysis equations for uniform plastic sway mechanism [13]. While the theoretical and calculated base shears for
SPSW-CD were less than 3% differences, those for SPSW-ID were about 13% differences. This significant discrepancy
on SPSW-ID was attributed to the fact that in-span plastic hinges developed on its HBEs thus SPSW-ID did not follow
the assumed uniform plastic sway mechanism (also known as ‘panel mechanism’) but rather consists of a ‘sway’ and
‘beam’ combined mechanism. An equation to calculate the theoretical base shear strength of SPSW having in-span
plastic hinges considering their actual plastic mechanism was derived in [12]. Using that equation, a theoretical base
shear for SPSW-ID was 304 kips, which agreed within 2.2% with the aforementioned 311 kips result from the SAP2000
analysis.
To investigate whether plastic hinging along an HBE span could lead to progressively increasing deformations in the
HBEs of both SPSW-ID and SPSW-CD, and whether it may affect structural performance, cyclic pushover analysis was
conducted with a progressively increasing cyclic displacement history of up to 3% drift (in increment of 0.5%). Fig. 4a
shows the plastic hinge and strip yielding distributions on SPSW-ID. When the structure experienced +1% and 1%
lateral drift, respectively, a total of four and five plastic hinges occurred at the HBE ends. In addition, three strips (the
right- or left-leaning strips for the positive or negative direction, respectively) on the second and the third floor remained
elastic and only two strips on the first floor had yielded. Though more strips yielded as the pushover displacement
increased, some strips remained elastic. Beyond the plastic hinges that occurred at the HBE ends, three locations of in-
span plastic hinges were also observed on HBE2 and HBE3 (2nd and 3 rd floor) at the end of 2% drift cycle; and the
yielding condition occurred along the span of HBE0 and HBE1 (base and 1st floor). At the end of the 3% drift cyclic, in-
span plastic hinges on the HBEs occurred at 4 locations for both positive and negative drift excursions. In contrast with
SPSW-CD (presented in Fig. 4b), most of the strips had yielded at the end of the 1% drift cycle and only four right-
leaning strips and five left-leaning strips in total had remained elastic. All strips have completely yielded at the end of the
3% drift cycle. In addition, all plastic hinges have developed at its HBE ends and no in-span plastic hinge developed.
 (a)

yielding condition

yielding condition

(b)

1% Drift 2% Drift 3% Drift

Legend:
| = partial plastification | fully plastic in positive drift | = partial plastification | fully plastic in both directions
| = partial plastification | fully plastic in negative drift = yielding condition (yielding remains within HBE flanges)
= strip yielding (P = Py)
Fig. 4. Plastic Hinge and Strip Yielding Distributions on (a) SPSW-ID; (b) SPSW-CD

A most significant phenomenon observed is the HBE vertical downward deformation of SPSW-ID, progressively
increasing and of significant magnitude as the lateral drift increased, as one example is shown in Fig. 5a. This figure
compares vertical displacement history at the mid-span of the top HBE for both SPSWs. The HBE vertical downward
displacement for SPSW-ID increases faster than that for SPSW-CD. In other words, this accumulative plastic
incremental deformation due to cyclic pushover displacement detrimentally affects the structural performance of SPSW-
ID. For example at +3% drift, the HBE3 vertical displacement of SPSW-ID was 2.3 in.; about 2.6 times larger than that
of SPSW-CD, which was 0.9 in.
n
(

)
i

Lateral Drift Lateral Drift

m
D

n
s

a
c
e

-4% -3% -2% -1% 0% 1% 2% 3% 4%

t
i

-4% -3% -2% -1% 0% 1% 2% 3% 4%

0.0 0.0
(a) (b)
-0.5 -0.5
Vertical Displacement (in) .

V
e

c
a
r
t
i

-1.0 -1.0

-1.5
-1.5
-2.0
-2.0
-2.5
-3.0
SPSW-CD -2.5
SPSW-CD
SPSW-ID SPSW-ID
-3.0 -3.5
-14.4 -10.8 -7.2 -3.6 0 3.6 7.2 10.8 14.4 -14.4 -10.8 -7.2 -3.6 0 3.6 7.2 10.8 14.4
Lateral Displacement (in) Lateral Displacement (in)

Fig. 5. History of HBE3 Vertical Displacement (a) Cyclic Pushover; (b) Time History
 Another approach that can be used to examine the behavior of the two SPSWs is by comparing the moment-rotation
hysteresis of their HBEs, as one example shown in Fig. 6. Unlike the general hysteresis curve for special moment resisting
frames, which is typically symmetric with respect to positive and negative rotations developed under a symmetric cyclic
pushover displacement history, the hysteresis curves of both SPSWs considered here are not symmetric but looping with a bias
toward one direction. The tension forces from the infill plates contribute to this behavior by always pulling the HBE in the
direction of the tension forces. Interestingly, except for the bottom HBE, all the moment-resisting ends of the HBEs of SPSW
ID developed a cross-section rotation (i.e., cross-section curvature multiplied by plastic hinge length) greater than 0.03 radians
after the structure was pushed cyclically up to a maximum lateral drift of 3%. In one case (i.e., HBE2), the total rotations even
reached 0.062 radians. Such a significantly high cyclic rotation demand would be difficult to achieve using the type of moment
resisting connections used in SPSW (the AISC Seismic Specifications [2] only require that Ordinary-type connections be used
in SPSW). In fact, it might also be difficult to achieve with special moment resisting frame (SMRF) beam-to-column
connections approved by [2], which are experimentally verified to perform well up to +/ 0.04 radians total rotations, or +/ 0.03
radians plastic rotations. By comparison for1.5 SPSW CD, all HBE total rotations obtained were1.5 less than or equal to 0.03
radians under the same cyclic
pushover displacements up to 3% drift.

3.2. Nonlinear Time History Analysis

Nonlinear time history analysis was conducted to investigate whether those previous results would be replicated
during earthquake excitations and whether additional seismic behaviors for the aforementioned SPSW systems would
emerge as a consequence of the random nature of earthquake records. Three synthetic time histories of ground
acceleration were generated for this purpose, which their spectra matched the design basis earthquake (DBE) spectra.
The accumulative plastic incremental deformation is still observed (Fig. 5b), with maximum and residual vertical
deformations more apparent on SPSW-ID than on SPSW-CD. For example, when SPSW-CD reached a lateral drift of
1% for the first time, the largest HBE3 vertical displacement at the same drift for SPSW-ID was 2.25 larger. This implies
that the HBE3 vertical downward displacement for SPSW-ID increased faster than that for SPSW-CD as the lateral drift
increased.
The nonlinear time history analyses were then extended to investigate the performance of both SPSWs under the more
severe (MCE). It was observed that as the severity of the synthetic ground motions increased for the MCE case
(consequently generating higher lateral drifts on both SPSWs), HBE vertical deformations of SPSW-ID especially at the
top two floors significantly increased compared to the corresponding magnitudes in the DBE case. For example, HBE3
maximum vertical deformation increased from 3.2 inches in the DBE case to 5.1 inches in the MCE case. By comparison
for SPSW-CD, only minor changes of HBE vertical deformations occurred. Hence, when formation of in-span plastic
hinges on HBEs is possible, such as in the case of SPSW-ID, the more severe the ground excitations, the more
accumulation of plastic incremental deformation observed.
3.3. Cyclic Pushover Test of a Three-Story SPSW Specimen
An experimental program was deemed desirable to investigate whether the above undesirable behaviors could also be
observed in an actual SPSW. Here, as shown in Fig. 7a, a 1/3-scaled three-story single-bay SPSW specimen was
prepared for this purpose. Special moment resisting connections were designed for all HBE-to-VBE connections. A
gravity column system was used to apply loads and provide lateral supports to the specimen (Figs. 7b-7e). The specimen
was instrumented to collect experimental data, including displacement transducers, uniaxial and triaxial Rosette strain
gauges, load cells, and video cameras. The loading protocol for this experimental program was developed as a
combination of the ATC 24 protocol [14] and the AISC Seismic Provisions [2] requirements. Details of the specimen can
be found in [15]. In addition, a finite element investigation of the tested specimen was performed.

(a) 1'-0" 1'-0"

10" W6 12

4'-2¼" 19 Ga. Plate

t = 0.0418"

S5 10

W6 25
Height

¾ 14 Ga. Plate
:

13'-4" 4'-2 " t = 0.0747"

Full
W

S5 10
6
2
5
H

h
g
e

t
:
i

13 Ga. Plate
t = 0.0897"
F
u
l
l

3'-6"

W8 13

7"

7'-0"

8'- 0¾"
8'-85/8"

Fig. 7. Three-Story SPSW Specimen: (a) Elevation View; (b) Experimental Setup with Gravity Column System; (c) Angles for
Load Transfer Mechanism; (d) and (e) Angles for Lateral Supports at HBE Bottom Flange and Both Sides of VBEs

In general, the hysteretic loop resembles that expected for unstiffened thin SPSWs reported in past experimental
research, namely: exhibiting pinching in the hysteretic curve, stable and ductile behavior when undergoing large lateral
drifts, and relatively small strength degradation between cycles at the same displacement step. Inspection of the
specimen after the test showed that plastic hinges developed at both ends of each HBE and at the base of each VBE.
While it was not as pronounced as that on HBE0 (Fig. 8), evidence of in-span plastic hinge was also observed on HBE3
indicated by yielding initiated at its top flange and spread to its web around the ¼- and ¾ -points. Moreover, yield lines
distributed along HBE flange indicated that plastification of the upper two HBEs was not localized within a finite length
as in the case of HBE0 but rather distributed within a longer span. Qualitative indication suggested that complete infill
plate yielding occurred in each panel as shown by flaking of whitewash roughly over the entire surface of each panel.
However, quantitative measurements indicated that plate yielding did not develop simultaneously, but rather
progressively. In span plastification of HBE contributed to this delay of plate yielding. At the conclusion of the test, it
appeared that the entire infill plates had yielded. Hence, the theoretically predicted incomplete yielding of the infill
seems to not have happened. However, it is important to emphasize that the load transfer set-up used in the experimental
program is partly responsible for this difference in behavior.
 Fig. 8. Sign of In-Span Plastification on HBE0 at 3= 2.5% Drift: (a) West Side; (b) East Side

Vertical deformations recorded during this cyclic pushover testing and that obtained from a FEM analysis are
presented in Fig. 9 for HBE3, HBE2 and HBE0. Although evidence of in-span plastic hinge (inferred from yield line in
the whitewash) was only clearly visible on HBE0, the results in those figures indicate that accumulation of plastic
incremental deformations were apparent not only on HBE0 but also on HBE2 and HBE3. The “backbone” displacements
(defined as the displacement when the structure reached the maximum target drift of every cycle) progressively increased
from practically zero in the first displacement step to 1.64, 0.76, and 1.11 in. respectively for HBE0, HBE2, and HBE3
when the specimen was cycled up to 3.33% top story drift amplitude. The same trend was also observed on the
progression of the “residual” displacement (defined as HBE residual vertical displacement when the structure returned to
its original position at 0% drift). Incidentally, the finite element results show the accumulation of plastic incremental
deformation similar to that was observed during the experiment. However, their magnitudes were significantly smaller
compared to those recorded during the experiment. It is suspected that the decision to exclude several displacement steps
from the analysis and to conduct only one cycle per displacement step; as well as the fact that the FEM model was not
developed to consider material fatigue life, contributed to this discrepancy.
The experimental results shown in Fig. 10 for HBE3 and HBE2 confirm the development of un-symmetric hysteretic
curves, making it possible to experimentally verify this analytically predicted behavior. Note that unidentified factors
contributed to some irregularities in the hysteresis curves of HBE3 in its earlier cycles, introducing an early “drift” in the
results, but this did not affect the subsequent one-sidedness behavior of the moment-rotation hysteretic loops.

CONCLUSIONS

Seismic performance of SPSWs having infill plates designed to sustain different percentages of the applied lateral
loads (i.e., conventional and balanced designs) was investigated using the FEMA P695 methodology. All conventional
archetypes met the FEMA P695 performance criteria for the R factor of 7 used in their design. By contrast, the balanced
archetypes designed with an R factor of 7 did not meet the FEMA P695 performance criteria. To rigorously meet the
performance criteria, an R factor of 5 was required for the balanced SPSWs. Most importantly, the balanced archetypes
were found to have a higher probability to suffer significantly larger (and unacceptable) interstory drift than the
conventional archetypes. Savings in steel when designing balanced SPSWs with a lower R factor came at the cost of the
SPSWs developing such larger interstory drifts compared to the conventional SPSWs under MCE ground motions. These
findings suggest that the infill plates of SPSWs should be designed to resist the total specified story shears, rather than be
designed by sharing those forces between the boundary frame and infill.
Using the nonlinear static analysis and nonlinear time history analyses, significant consequences to having in-span
plastic hinges were identified. It was demonstrated that plastification along HBE spans can induce significant
accumulation of plastic incremental deformations on the HBEs, themselves leading to partial yielding of the infill plates
and correspondingly lower global plastic strength compared to the values predicted by code equations. The experiment
demonstrated the development of in-span plastification and accumulation of plastic incremental deformations. These
findings suggest that development of in-span hinges should be explicitly avoid in the design of HBEs. This research also
allowed to experimentally observe that the moment-rotation hysteresis curves of HBEs connections in SPSWs are not
symmetric, but rather lopsided toward one direction, unlike what is typically observed in special moment-resisting
frames.
ACKNOWLEDGEMENTS

Incremental dynamic analyses were executed using the NEES-HUB supercomputer. This work was supported in part
by the George E. Brown, Jr. Network for Earthquake Engineering Simulation (NEES) Program of the National Science
Foundation under NSF NEESR Award Number CMMI-0830294 and by the Multidisciplinary Center for Earthquake
Engineering Research (MCEER), University at Buffalo. W-steels used for this study were donated by the American
Institute of Steel Construction (AISC). Niagara Testing performed nondestructive inspection on several special moment
connections. The financial support of the Fulbright Indonesia Presidential Scholarship and the Universitas Bandar
Lampung, Indonesia to the first author is gratefully appreciated. However, any opinions, findings, conclusions, and
recommendations presented in this paper are those of the writers and do not necessarily reflect the views of the sponsors.

REFERENCES

[1] AISC (2010): Seismic Provisions for Structural Steel Buildings. ANSI/AISC 341-10, American Institute of Steel
Construction, Inc., Chicago, Illinois.
[2] Canadian Standards Association (CSA) (2009): Design of Steel Structures. CAN/CSA S16-09, Willowdale, Ontario,
Canada.
[3] Driver, R. G., Kulak, G. L., Kennedy, D. J. L., and Elwi, A. E. (1997): Seismic Behavior of Steel Plate Shear Walls.
Structural Engineering Report 215, Department of Civil Engineering, University of Alberta, Edmonton, Alberta,
Canada.
[4] Qu, B., and Bruneau, M. (2009): Design of Steel Plate Shear Walls Considering Boundary Frame Moment Resisting
Action. Journal of Structural Engineering, ASCE, Vol. 135, No. 12, pp. 1511-1521.
[5] FEMA. (2009): Quantification of Building Seismic Performance Factors. FEMA Report No. P695, Prepared by the
Applied Technology Council for the Federal Emergency Management Agency, Washington, D.C.
[6] FEMA. (2000): State of the Art Report on Systems Performance of Steel Moment Frames Subject to Earthquake
Ground Shaking. FEMA Report No. 355C, Prepared by the SAC Joint Venture Partnership for the Federal
Emergency Management Agency, Washington, D.C.
[7] Purba, R., and Bruneau, M. (2014): Seismic performance of steel plate shear walls considering two different design
philosophies of infill plates. II: Assessment of collapse potential. . Journal of Structural Engineering, ASCE,
10.1061/(ASCE)ST .1943-541X.0001097, 04014161.
[8] Mazzoni, S., McKenna, F., Scott, M. H., and Fenves, G. L. (2009): Open System for Earthquake Engineering
Simulation (OpenSees) User Command-Language Manual Version 2.0. Pacific Earthquake Engineering Research
Center, University of California, Berkeley, Berkeley, CA.
[9] Purba, R., and Bruneau, M. (2014): Seismic performance of steel plate shear walls considering two different design
philosophies of infill plates. I: Deterioration model development. Journal of Structural Engineering, ASCE,
10.1061/(ASCE) ST.1943-541X.0001098, 04014160.
[10] Vian, D., and Bruneau, M. (2005): Steel plate shear walls for seismic design and retrofit of building structures.
Tech. Rep. MCEER-05-0010, Multidisciplinary Center for Earthquake Engineering Research, State University
of New York at Buffalo, Buffalo, New York.
[11] Berman, J. W., and Bruneau, M. (2008): Capacity Design of Vertical Boundary Elements in Steel Plate Shear Walls.
Engineering Journal, AISC, First Quarter, pp. 57-71.
[12] Purba, R., and Bruneau, M. (2012): Case Study on the Impact of Horizontal Boundary Elements Design on Seismic
Behavior of Steel Plate Shear Walls. Journal of Structural Engineering, ASCE, Vol. 138, No. 5, pp. 645-657.
[13] Berman, J. W., and Bruneau, M. (2003): Plastic Analysis and Design of Steel Plate Shear Walls. Journal of
Structural Engineering, ASCE, Vol. 129, No. 11, pp. 1448 1456.
[14] ATC-24 (1992): Guidelines for Cyclic Seismic Testing of Components of Steel Structures. Applied Technology
Council, Redwood City, CA.
[15] Purba, R., and Bruneau, M. (2015): Experimental Investigation of Steel Plate Shear Walls with In-Span
Plastification along Horizontal Boundary Elements. Engineering Structures, Vol. 97, No. 8, pp. 68-79.