Article

Advances in Structural Engineering

2016, Vol. 19(5) 795–805
Experimental study on application of Ó The Author(s) 2016
Reprints and permissions:
reduced section in long link-to-column sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1369433216630374

connections ase.sagepub.com

Ali Naserifar and Fakhreddin Danesh

Abstract
Link-to-column connections in steel eccentrically braced frames must sustain large moments and shear forces to support link rotation.
Recent studies showed that link-to-column connections are prone to failure at low drift levels due to their susceptibility to fracture at
the link flange-to-column connections. The most promising method to overcome this problem is to utilize reduced beam section in
link, in order to keep away stress concentration from the link ends. In this investigation, the application of reduced beam section in
long link beams is studied experimentally. Two full-scale specimens with different reduced beam section locations are constructed and
tested under cyclic loading. In one specimen, reduced beam section is placed at both ends of the link beam; and in the other specimen,
reduced beam section is solely placed at the link end which is near the column. The results show that both specimens have satisfied
the seismic code requirements for the link inelastic rotation, which in turn leads to improvement in seismic performance.

Keywords
cyclic behavior, eccentrically braced frames, long link-to-column connection, reduced link section

Introduction The yielding behavior of EBF links is governed by

the normalized link length, r defined as
Moment resisting steel frames have suitable ductility.
However, dimensions of frame members in such a sys- e
r= ð1Þ
tem must be taken substantially large in order to con- Mp =Vp
trol story drifts which in turn will increase the cost of
construction. On the other side, the inverse concentri- where Mp is the plastic moment capacity of the link
cally braced frames exhibit large enough lateral stiff- section, Vp is the plastic shear capacity, and e is the link
ness and easily satisfy code requirements for story length. According to the link normalized length, they
drifts, but they do not have enough ductility to absorb are categorized into three types: short shear yielding
the input energy during a severe earthquake. The link (r ł 1:6), long flexural yielding link (r ø 2:6), and
eccentrically braced frames (EBFs) were postulated by intermediate link (1:6 ł r ł 2:6) that yields in combi-
Roeder and Popov (1978) to relief defects of two nation of shear and flexure. Between above-mentioned
above-mentioned earthquake resistance systems. EBF link beams, shear yielding one is more recommended
offers the enough ductility to absorb the input seismic due to its better ductility, stiffness, and strength. Most
energy and at the same time maintains large enough of the previous researches focused on the behavior of
stiffness to satisfy code requirements for story drifts. this type of links (Chao et al., 2006; Dusicka et al.,
So, there is more than three decades that EBFs have 2004, 2010; Kasai and Popov, 1986). However, there is
been used as earthquake resistance structural systems a tendency to utilize long links instead of short ones in
in many countries. Initial research on this system has
been done by Popov et al. from 1978 to 1984 Faculty of Civil Engineering, K. N. Toosi University of Technology, Tehran,
[Hjelmstad and Popov (1983), Hjelmstad and Popov Iran
(1984), Malley and Popov (1984), Popov (1983),
Popov and Malley (1983)]. The key elements in such Corresponding author:
Ali Naserifar, Faculty of Civil Engineering, K. N. Toosi University of
system are link beams (as shown in Figure 1) that must Technology, No. 1346, Vali Asr Street, Mirdamad Intersection, Tehran,
have stable and ductile hysteretic behavior during Iran.
severe earthquake loading. Email: [email protected]
796 Advances in Structural Engineering 19(5)

(2014a) derived an analytical formulation to evaluate

the internal actions and the plastic deformations of
intermediate links and established appropriate hierar-
chy criteria to avoid undesired collapse mechanisms in
such types of EBFs.
Engelhardt and Popov (1992) investigated the per-
formance of link-to-column connections experimen-
tally. The results showed that long link-to-column
connection failed due to fracture at the link flange-to-
column groove weld or at the link flange in vicinity of
groove weld. This failure caused sudden drop in the
load-bearing capacity of connection and prevent the
connection to support the expected levels of inelastic
rotation. Consequently, they recommended that long
Figure 1. Typical eccentrically braced frames with (a) mid- links (r ø 1:6) attached to columns should not be used
span links and (b) links with column connection. (Prinz and in EBFs. Okazaki et al. (2006) conducted 12 full-scale
Richards, 2009). tests to achieve the behavior of link-to-column connec-
tion with various lengths. They observed that the most
order to provide larger openings which in turn lead to of the specimens failed by fracture of the link flange
satisfaction of functional and architectural require- near the groove weld. Only one of the 12 specimens
ments. Daneshmand and Hosseini Hashemi (2012) met the plastic link rotation requirement and the oth-
investigated the performance of intermediate and long ers did not provide the levels of inelastic rotation
links through finite element analysis. Consequently, required by the 2005 American Institute of Steel
they showed that some intermediate links that have Construction (AISC, 2005b) Seismic Provisions,
1:8 ł r ł 2:2 and called controversial length region referred to as The Provisions herein. They concluded
failed to achieve the required rotation capacities. They that further research is needed to identify satisfactory
advised that the designers shall not use the links in the link-to-column connection details. They also recom-
controversial length region; however, some considera- mended that EBFs with links attached to the columns
tions have been suggested. should be avoided. Consequently, on the basis of these
According to their location, link elements are studies, The Provisions advises the designers to avoid
divided into two forms as shown in Figure 1. In the EBF configurations with links attached to columns
first form, link element is placed between the diagonal until further research has been conducted to identify
brace and column (Figure 1(a)), while in the second suitable details.
form link is placed between the two braces (Figure To overcome these problems, a number of improved
1(b)). The application of EBFs with link beams beam-to-column connection design strategies can be
attached to the column is more interested than middle used in link-to-column connection, many of which
link beams. This is due to the fact that this form has have shown to exhibit satisfactory levels of ductility in
more flexibility from architectural point of view. But numerous tests. Two key concepts have been devel-
due to the fracture of link beam-to-column connection, oped in order to provide highly ductile response and
EBFs with link beam attached to the column have less reliable performance: strengthening the connection
plastic rotation capacity than other one (Engelhardt and/or weakening the beam framing into the connec-
and Popov, 1992; Prinz and Richards, 2009). tion (FEMA 350, 2000).
Therefore, an important issue in design of this type of In the tests by Engelhardt and Popov (1992), only
EBF is the development of suitable link-to-column one of the test specimens, where triangular cover plate
connection details. has been used in the connection, reached required plas-
In addition to the above-mentioned configuration tic rotation. Although one test specimen is not enough
of EBFs that the link belongs to the beam member, to conclude that these types of connections satisfy
there is Y-scheme of EBFs that link does not belong to requirements, but this can be considered as an appro-
the beam member. A modern application of the capac- priate approach and further research should be con-
ity design principle was applied to these types of EBFs ducted. Okazaki et al. (2014) and Hong et al. (2014)
by Montuori et al. (2014b). They presented a design reinforced the connection region using a pair of dou-
methodology aiming to assure the development of a bler plates in the first link web panel next to the col-
collapse mechanism of global type for seismic-resistant umn, parallel to, but offset from the link web, called
EBFs with inverted Y-scheme. Also Montuori et al. the supplemental web doubler connection. These
Naserifar and Danesh 797

specimens have good cyclic performance and most of

them achieved target inelastic rotation; however, they
are in the range of shear yielding links.
As second strategy, improved performance of link-
to-column connections may be achieved by reducing
the severe stress concentration in the region near the
critical link flange welds. For this purpose, reduced
beam sections (RBSs) such as the ones used for moment
resisting frames may be utilized in the link. The typol-
ogy and the technology of the RBS beam-to-column
connection, as well as its behavior under cyclic load Figure 2. Schematic representation of test setup.
conditions, were comprehensively investigated, in the
United States, after the Northridge earthquake (Moore
configuration is selected such that the link is supposed
et al., 1999; Popov et al., 1996, 1998). However, the
to minimize axial loads, a constant shear force and
large moment gradient along the relatively short length
almost reverse equal end moments. The test setup and
of the link makes the RBS difficult if not impossible to
its schematic representation are shown in Figures 2
implement for link-to-column connections. But apply-
and 3, respectively. This test setup is similar to the one
ing the RBS concept to long links can move the loca-
used by Okazaki (2004) experimental work. However,
tion of severe plastic strain and associated localized
due to the limitations of reaction frame, some changes
flange deformation away from the column face and sen-
are applied. A cyclic displacement is imposed to the tip
sitive link flange welds. Berman et al. (2010) investi-
of the column by a hydraulic actuator. The hydraulic
gated the behavior of reduced link section (RLS)
loading system, mounted between the reaction frame
numerically and observed that the behavior of the link
and the column tip, consisted of a 1000 KN hydraulic
beams is improved when RBSs are used in all types of
actuator with 150 mm equal stroke for both forward
link beams. This is due to the fact that the stress and
and backward directions. The test setup is equipped
plastic strain concentration were at the RLS region
with a load cell which is placed on the tip of a hydraulic
instead of the link ends. Although the analysis results
actuator to record the applied loads. Also, the actuator
by Berman et al. (2010) indicate the promise of the RLS
displacement is measured via both an internal displace-
concept, an experimental verification study was recom-
ment transducer of the actuator and an external linear
mended by the authors before the proposed method be
variable displacement transducer (LVDT).
implemented in link-to-column connections. Also
In order to restrain the out of plane movement of
numerical investigation by Danesh and Faridalam
the specimen, lateral bracings are utilized in the vici-
(2012) showed that the RLS approach is more efficient
nity of the link to beam connection and also near the
in long links when compared to the other types.
beam end. Furthermore, the column is laterally braced
In this article, it is aimed to investigate the applica-
near the point of loading and its far end as shown in
tion of reduced section in long link-to-column connec-
Figure 4. Also, the link is welded to heavy end plate at
tions through experimental study. For this purpose,
the link end opposite to the column where it is bolted
two specimens with long link-to-column connection
to the outside beam.
are tested. At the first specimen, RBS is used at the
both end of the link while at the second RBS is solely
used at the link end which is near the column. These Design and construction of specimens
specimens are called 2SIDE-RLS and 1SIDE-RLS,
respectively. As already mentioned, two specimens are constructed
To the knowledge of the authors, the tests con- which are consisted of an IPB240 as column, an
ducted in this investigation are the first series of experi- IPE300 as link, and beam segment outside of the link
ments specifically aimed to study the seismic using ST37 steel. Full-depth 10 mm web stiffeners are
performance of large-scale EBF with RLS to column provided on both sides of the link web at the start and
connections. end of RBSs, as shown in Figure 6(d). The column is
provided with 12 mm continuity plates and a 10 mm
doubler plate in the panel zone to restrict inelastic
Experimental program behavior in this region. Table 2 summarizes the char-
acteristics of constructed specimens.
Test setup According to the capacity design principles, the spe-
The test setup is prepared to simulate the boundary cimens are designed such that majority of inelastic
conditions of the link-to-column connection. The behavior is concentrated at the link, minor inelastic
798 Advances in Structural Engineering 19(5)

Figure 3. Test setup.

Figure 5. Geometry of RBS.

Intermediate Steel Moment Resisting Frames for

Seismic Applications (AISC, 2005a), referred to as
Prequalified Connections herein, to be within the limits

0:5bbf ł a ł 0:75bbf
0:65d ł b ł 0:85d
0:1bbf ł c ł 0:25bbf

where bf is the width of link beam flange and d is the

depth of link beam.
In this study, values of above-mentioned parameters
are considered as follows
Figure 4. Overall view of the test setup.
a = 0:6bbf
rotations occurred at the panel zone, and the beam b = 0:75d
segment outside of the links. To achieve this goal, the
c = 0:2bbf
beam outside of the link is strengthened with added
plates on both flanges. And the values of them are listed in Table 1.
Montuori (2014) derived a closed form solution for
the location of RBS in order to avoid the plasticization
RLS design of beam-to-column connections. From this study, it
In this study, link beam flanges are cut such as RBS can be seen that when the value of gravity loads is
approach in the moment frames, as seen in Figure 5. equal to zero and the ratio of the plastic moment of
Dimensions a, b, and c in Figure 5 are constrained by the weakened beam section to the plastic moment of
the AISC Prequalified Connections for Special and the complete beam section is equal to 0.6 (in the case
Naserifar and Danesh 799

of the specimens presented in this study), the distance applied quasi-statically in displacement-controlled
of RBS center from the beam end to the beam length mode of the actuator by low rate to identify the beha-
((a + b=2)=L) should be limited to 0.25. This ratio is vior of the connection during the load cycles and
approximately equal to 0.127 in the specimens adopted record all the deformations and probable events.
in the experimental tests. Therefore, they satisfy the
abacus obtained in the above-mentioned publication.
As said, the yielding behavior of EBF links is gov- Test results
erned by r (the normalized link length). In the both As noted earlier, incremental cyclic static loading is
specimens, link beam section is designed with IPE300 applied on the specimens and results are presented
with r = 3:02 and therefore the behavior of link is herein. The hysteresis during the displacement-
dominated by flexural yielding. controlled testing for both specimens is shown in
Figure 6 shows the details of two specimens of RLS Figures 7 and 8, which plots link shear versus elastic
to column connection. In these specimens, link beam is and inelastic rotations. It is apparent from this figure
directly connected to the column through complete that the 2SIDE-RLS reached the 5% link total rota-
joint penetration welds of link flanges to column tion and 3% link inelastic rotation without any failure
flange. The link web is also connected to the column
in the vicinity of welds or at the RLS region. The
flange using fillet welds. Weld access hole has been
inelastic rotation, g p , is evaluated by removing the
considered according to The Provisions requirements
contributions of elastic response from the link rota-
as shown in Figure 6(g). In the 1SIDE-RLS specimen,
tion, g. Due to multiple degrees of freedom in the test
stiffeners are used on both sides of link web within the
setup, some pinching phenomena can be seen in the
distance of 1.5bf from the end far from the column
cyclic response of specimens (Figures 7 and 8). The
where RBS is not placed. Note that these stiffeners are
specimen 1SIDE-RLS reached the 7% link total rota-
in addition to the stiffeners used for RBS end. Table 2
tion and 3% link inelastic rotation without any failure
lists the specifications of both specimens.
in the vicinity of welds or at the RLS region. It is
worthwhile to note that due to limitations of the actua-
Material properties tor, the maximum drift of 7% is applied to the speci-
men. According to The Provision, specimen 1SIDE-
Table 3 lists the mechanical properties of the link sec-
RLS achieved 5% link total rotation and 3% link plas-
tions, determined based on tension coupons sampled
tic rotation.
from the edges of the flange, and mid-depth of the
The acceptance criteria for links are defined in The
web. The average of the modulus of elasticity was cal-
Provisions based on inelastic rotation. The Provisions
culated to be about 206 GPa.
specify that the flexural yielding links should be capa-
ble of an inelastic rotation of 0.02 rad. The inelastic
Loading protocol rotation capacity of the link specimens is defined per
The loading protocol which is used in this study fol- The Provisions, as the maximum level of inelastic rota-
lows the protocol outlined in The Provisions. This tion sustained for at least one full cycle of loading prior
loading protocol presented in Table 4. However, due to the link shear strength drops below the nominal link
to limitations of actuator, the maximum rotation of shear strength. The target inelastic rotation of 0.03 rad
7% is imposed to the specimens. The loading is was exceeded by 50%. Therefore, the link-to-column
connection of both specimens satisfied the performance
requirements of The Provisions.
Table 1. Reduced link section dimension.

Link beam (IPE300) RBS dimensions Observation

Depth (mm) Width (mm) a (mm) b (mm) c (mm) In the 2SIDE-RLS specimen, no considerable white-
wash flaking is observed until the end cycle of 1% link
300 150 90 225 30 total rotation angle. During the first cycle of 1.5%

Table 2. Specifications of specimens.

Specimens Column Beam Doubler plate thickness (mm) Continuity plate thickness (mm) Stiffener thickness (mm)

2SIDE-RLS IPB240 IPE300 10 12 10

1SIDE-RLS IPB240 IPE300 10 12 10
800 Advances in Structural Engineering 19(5)

Figure 6. Details of specimens: (a) 2SIDE-RLS, (b) 1SIDE-RLS, (c) link beam properties of specimen 2SIDE-RLS, (d) link beam
properties of specimen 1SIDE-RLS, (e) details of RLS from link top view-Specimen 1SIDE-RLS (in cm), (f) details of RLS from link top
view-Specimen 2SIDE-RLS (in cm), and (g) dimension of weld access hole (in mm).
Naserifar and Danesh 801

Table 3. Link section properties. rotation, the yielding is initiated in the beam flanges
after a minor flaking within the reduced section region,
Fy (Mpa) Fu (Mpa) Elongation (%) as seen in Figure 9, and extended to the entire reduced
region at the end of 2% story drift and 3% drift.
Web 282.0 416.7 27.7
Flange 271.8 411.4 29.3 Figure 10 shows the RLS region of the 2SIDE-RLS
at the end of the test. This RLS is the one near the col-
umn. Yielding of RLS is evidenced by flaking of the
whitewash. Also, local buckling can be seen in the link
Table 4. Cyclic loading protocol. flange at RLS region. Initiation of web local buckling
in the RLS induces flange local buckling followed by
Amplitude: g (rad) Number of cycles
strength degradation in the specimen and reduced
0.00375 6 demand on the panel zone. Finally, 2SIDE-RLS speci-
0.005 6 men failed due to fractures at the fillet welds connect-
0.0075 6 ing the link flanges to the end plate in the beginning of
0.01 6
0.015 4
cycle 7% of loading as is seen in Figure 11. This failure
0.02 2 is considered an artifact of the test setup, as the link
0.03 2 end connection used for the specimen was not repre-
0.04 1 sentative of conventional link end connection used in
0.05 1 actual EBFs. Significant yielding is visible inside the
0.07 1
0.09 1 link reduced section panels. Yielding in RLS region at
0.11 1 the link end opposite to the column is shown in
Figure 12. As be seen from this figure and obvious

Figure 7. Cyclic response of specimen 2SIDE-RLS.

400 400
300 300
200 200
LINK SHEAR (KN)

LINK SHEAR (KN)

100 100
0 0

-100 -100

-200 -200

-300 -300

-400 -400
-8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6

LINK ROTATION γ (%) LINK INELASTIC ROTATION γp (%)

Figure 8. Cyclic response of specimen 1SIDE-RLS.

802 Advances in Structural Engineering 19(5)

Figure 9. Yielding at the RLS region, specimen 2SIDE-RLS at (a) 1.5%, (b) 2%, and (c) 3% drift.

Figure 10. RLS region near the link-to-column connection.

Figure 12. RLS region near the link end opposite to the
column.

Figure 13 shows the RLS region in 1SIDE-RLS at

the end of the test. Once again, yielding in the RLS is
evidenced by flaking of the whitewash and local buck-
ling can be seen in the link flange and web of RLS
region. Similar to the results of 2SIDE-RLS, initiation
of web local buckling in the RLS induces flange local
buckling in the specimen. Finally, 1SIDE-RLS speci-
men failed, because of rupture occurrence at the link to
end plate weld in the cycle 7% of loading (Figure 14).

Figure 11. Fracture at the link to end plate weld.

Strain distribution
from whitewash flaking, yielding in this region is a few According to Figure 15, the normalized longitudinal
less than the column side of the link. strains along link flange center versus drift ratio in the
Naserifar and Danesh 803

Figure 15. Distribution of strains along the link length.

Figure 13. RLS panel near the link-to-column connection.

Figure 16. Distribution of the flexural strains across the link

flange.

Distribution of the flexural strains across the link

flange at the back of groove welds shows that after
1% drift ratio its value at the center of the link flange
is higher than flange edge (Figure 16). Therefore, it is
more susceptible that the fracture to be initiated from
the center of the link flange than from its edge.
The comparison of flexural strains on the link
flange near the column reveals that the flexural strains
at the end of the link are much more than in the RLS
Figure 14. Fracture at the link to end plate weld near the link region at the start of the loading until 2% of drift
outside beam. (Figure 17). But the growth rate of the longitudinal
strains on the reduced region of the link become so
extensive than the link end by increasing the loading
2SIDE-RLS specimen are compared. The results show after 2% drift. This appears that reducing the link
that up to drift 4%, the maximum flexural strain is flange is effective in decreasing strains and stresses
observed in the center of the link flange near link-to- demands in high rotation level at the vicinity of the
column connection groove welds. At this point, flex- groove welds and therefore increases ductility of the
ural strain exceeds the elastic limit in the drift ratio link-to-column connection. These results clearly indi-
within 0.5%–0.75% while strains at the other points cate that the RLS did not eliminate all inelastic strain
are in the range of elastic limit and the yield strain is demands at the link flange groove welds but reduction
achieved after 1% drift ratio. Also, this figure shows in this demand is significant. These results are com-
that the values of the flexural strain on the link flange mon in the RBS to column connection in the moment
near the column are more than column opposite side frames and are previously remarked by the other
up to 4% drift ratio. researchers (e.g. Deierlein and Chi, 1999; Engelhardt
804 Advances in Structural Engineering 19(5)

The results show that both specimens have satisfied

the seismic code requirements for inelastic rotation of
the long link, which in turn leads to improvement in
the seismic performance. According to studied speci-
mens, it is also shown that this method prevents the
fracture to be occurred at the link-to-column connec-
tion. However, the strain gage values showed that the
stress concentration was being mitigated but not thor-
oughly removed from the link ends.
Since typically in practice, a continuous member is
used for the link and the beam segment outside of the
Figure 17. Distribution of strains on the link flange near the link, the RLS can be solely utilized at the link end near
column connection. the column. The results of 1SIDE-RLS specimen show
good performance of this form of RLS.
The experimental results confirm that the cyclic per-
formance of the RLS to column connection is excellent
and the plastic hinge is formed at the RLS region.
Also, no weld fracture was observed while both of spe-
cimens reached over the acceptable 0.02 rad plastic
rotation without fracture, supporting the overall valid-
ity of the RLS application. It is important to note that,
due to the limitations of this study to investigate the
influence of other parameters such as RLS dimension,
link length, and column size and section on the beha-
vior of RLS, further experimental researches are
required to be conducted prior to its application.

Figure 18. Comparison of strains in test specimens.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with
et al., 2000). Some of strain gages are separated from respect to the research, authorship, and/or publication of this
the flange due to high strain values in this region after article.
3% of the link rotation. Thus, there are no records to
be compared. Funding
The comparison of the longitudinal strains across The author(s) received no financial support for the research,
the center of the RLS region at both end of the link authorship, and/or publication of this article.
beam indicates that strain values in the edge of the
flange are more than the flange center. According to
test images (Figures 10 and 13), the local buckling in References
the RLS flange edge near the column approves this AISC (2005a) Prequalified connections for special and inter-
statement. In Figure 18, the normalized values of flex- mediate steel moment resisting frames for seismic applica-
ural strain of 2SIDE-RLS and 1SIDE-RLS specimens tions. Report no. ANSI/AISC 358-05. Chicago, IL: AISC.
near the groove weld of the link flange up to the drift AISC (2005b) Seismic provisions for structural steel buildings.
level 4% are compared. Much higher strain in the Standard ANSI/AISC 341-05, March. Chicago, IL: AISC.
Berman JW, Okazaki T and Hauksdottir HO (2010) Reduced
2SIDE-RLS specimen is evident.
link sections for improving the ductility of eccentrically
braced frame link-to-column connections. Journal of
Summary and conclusion Structural Engineering: ASCE 136(5): 543–553.
Chao SH, Khandelwal K and El-Tawil S (2006) Ductile web
In this investigation, RBS concept is utilized in a long fracture initiation in steel shear links. Journal of Structural
link in order to keep away stress concentration from Engineering: ASCE 132(8): 1192–1200.
the link ends. Two full-scale specimens with different Danesh F and Faridalam M (2012) Behavior of link-to-
RBS locations are constructed and tested under cyclic column connections in eccentrically braced frames. In:
loading. In one specimen, RBS is placed at both ends Proceedings of the fifteenth world conference on earthquake
of the link beam; and in the other specimen, RBS is engineering (15WCEE), Lisbon. Available at: http://
solely placed at the link end which is near the column. www.iitk.ac.in/nicee/wcee/article/WCEE2012_5284.pdf
Naserifar and Danesh 805

Daneshmand A and Hosseini Hashemi B (2012) Perfor- Montuori R, Nastri E and Piluso V (2014a) Rigid-plastic
mance of intermediate and long links in eccentrically analysis and moment–shear interaction for hierarchy cri-
braced frames. Journal of Constructional Steel Research teria of inverted Y EB-Frames. Journal of Constructional
70: 167–176. Steel Research 95: 71–80.
Deierlein GG and Chi WM (1999) Integrative analytical Montuori R, Nastri E and Piluso V (2014b) Theory of plastic
investigations on the fracture behavior of welded moment mechanism control for eccentrically braced frames with
resisting connections. Final report, SAC TASK 5.3.3. Palo inverted y-scheme. Journal of Constructional Steel
Alto, CA: The John A. Blume Earthquake Engineering Research 92: 122–135.
Center, Department of Civil and Environmental Engi- Moore KO, Malley J and Engelhardt M (1999) Design of
neering, Stanford University. Reduced Beam Section Moment Frame Connection. Struc-
Dusicka P, Itani AM and Buckle IG (2004) Finite element tural Steel Educational Council: Technical Information
investigation of steel built-up shear links subjected to and Product Service, Berkeley, CA, July.
inelastic deformations. Earthquake Engineering and Engi- Okazaki T (2004) Seismic performance of link-to-column con-
neering Vibration 3(2): 195–203. nections in steel eccentrically braced frames. PhD Disserta-
Dusicka P, Itani AM and Buckle IG (2010) Cyclic behavior tion, Department of Civil Engineering, The University of
of shear links of various grades of plate steel. Journal of Texas at Austin.
Structural Engineering: ASCE 136(4): 370–378. Okazaki T, Engelhardt MD, Hong JK, et al. (2014)
Engelhardt MD and Popov EP (1992) Experimental perfor- Improved link-to-column connections for steel eccentri-
mance of long links in eccentrically braced frames. Journal cally braced frames. Journal of Structural Engineering:
of Structural Engineering: ASCE 118(11): 3067–3088. ASCE 141(8): 04014201.
Engelhardt MD, Fry G, Johns S, et al. (2000) Behavior and Okazaki T, Engelhardt MD, Nakashima M, et al. (2006)
design of radius-cut, reduced beam section connections. Experimental study on link-to column connections in steel
SAC report 00/17. Sacramento, CA: SAC Joint Venture. eccentrically braced frames. Journal of Structural Engi-
FEMA 350 (2000) Recommended seismic design criteria for neering: ASCE 132(8): 1201–1211.
new steel moment-frame buildings. Popov EP (1983) Recent research on eccentrically braced
Hjelmstad KD and Popov EP (1983) Cyclic behavior and frames. Engineering Structures 5(1): 3–9.
design of link beams. Journal of Structural Engineering: Popov EP and Malley JO (1983) Design of links and beam-to-
ASCE 109(10): 2387–2403. column connections for eccentrically braced steel frames.
Hjelmstad KD and Popov EP (1984) Characteristics of eccen- EERC report no. 83/03. Berkeley, CA: University of Cali-
trically braced frames. Journal of Structural Engineering: fornia, Berkeley.
ASCE 110(2): 340–354. Popov EP, Blondet M and Stepanov L (1996) Application of
Hong J-K, Uang C-M, Okazaki T, et al. (2014) Link-to-col- dog bones for improvement of seismic behavior of steel con-
umn connection with supplemental web doublers in eccen- nections. Report no. UCB/EERC 96/05. Berkeley, CA:
trically braced frames. Journal of Structural Engineering: University of California, Berkeley.
ASCE 141(8): 04014200. Popov EP, Yang TS and Chang SP (1998) Design of steel
Kasai K and Popov EP (1986) Cyclic web buckling control MRF connections before and after the 1994 Northridge
for shear link beams. Journal of Structural Engineering: earthquake. Engineering Structures 20(12): 1030–1038.
ASCE 112(3): 362–504. Prinz GS and Richards PW (2009) Eccentrically braced frame
Malley JO and Popov EP (1984) Shear links in eccentrically links with reduced web sections. Journal of Constructional
braced frames. Journal of Structural Engineering: ASCE Steel Research 65: 1971–1978.
110(9): 2275–2295. Roeder CW and Popov EP (1978) Eccentrically braced steel
Montuori R (2014) The influence of gravity loads on the seis- frames for earthquakes. Journal of the Structural Division:
mic design of RBS connections (special issues). Open Con-
ASCE 104(3): 391–412.
struction and Building Technology Journal 8: 248–261.
Copyright of Advances in Structural Engineering is the property of Multi-Science Publishing
Co Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv
without the copyright holder's express written permission. However, users may print,
download, or email articles for individual use.