Engineering Structures 298 (2024) 117072

Contents lists available at ScienceDirect

Engineering Structures
journal homepage: www.elsevier.com/locate/engstruct

A replaceable sandwiched metallic fuse damper for seismic protection of

braced frames
Rafiq Ullah a, Mohammadreza Vafaei a, *, Sophia C Alih b, Abdul Waheed c
a
Faculty of Civil Engineering, Universiti Teknologi Malaysia, Skudai, Malaysia
b
Institute of Noise and Vibration, Faculty of Civil Engineering, Universiti Teknologi Malaysia, Skudai, Malaysia
c
Department of Civil Engineering, National University of Science and Technology, Balochistan Campus, QTA87300, Pakistan

A R T I C L E I N F O A B S T R A C T

Keywords: In this study, a replaceable Sandwiched Fuse Damper referred to as SFD is introduced to enhance the seismic
Replaceable fuse damper performance of braced frames. The SFD consists of one or several fuses made by Hollow Structural Section (HSS),
Flexural-yielding behavior sandwiched between two steel plates. The fuses are welded to the steel plates in a parallel pattern. An axial force
Seismic-resistant brace
in a brace equipped with the SFD imposes a relative movement on the steel plates welded to the fuses. The fuses
Component cyclic test
Hysteretic performance
yield as a result of such relative movements and dissipate the input energy. In contrast to other metallic dampers,
Analytical modeling the fabrication and installation of the SFD are much simpler, resulting in lower construction and maintenance
costs. The seismic performance of the SFD was investigated through a series of displacement-controlled cyclic
tests, in which the SFD was installed within a brace and its cyclic response was compared with that of the bare
brace. The experimental results showed that the SFD-brace system possessed stable hysteresis loops, a large
deformation capacity, and an excellent energy dissipation capacity. An analytical model was developed to es­
timate the SFD’s yield strength and its corresponding displacement. Experimental results were employed to
examine the efficiency of the proposed model. It was shown that the yield strength of tested specimens was
estimated with good accuracy.

1. Introduction Metallic yielding dampers as one of the most cost-effective devices

absorb the input energy through axial, flexural, or shear deformation
Seismic design codes often require collapse prevention under rare based on their yielding modes [8,9]. A widespread and practically
earthquakes for structures located in seismic-prone regions. Therefore, implemented buckling restrained brace (BRB) is a type of axial (tensile-
primary structural elements are usually designed to dissipate seismic compressive) yielding metallic damper [10]. BRBs usually consist of an
energy by plastic deformations at specific locations such as both ends of inner steel core restrained by an external restraining mechanism to
the beams and columns in moment-resisting frames and along braces in overcome the global buckling of the inner core and provide symmetrical
concentrically braced frames. It was realized after the Northridge hysteresis under cyclic load [11,12]. Furthermore, all-steel assembled
earthquake that the energy dissipation components are generally a part BRB was introduced to avoid the fabrication complexity and heavy
of the gravity-resisting system consequently making the post-earthquake weight of conventional BRB [13,14]. However, on the subassembly or
repair of structure usually difficult and costly [1,2]. Thus, an effective system level, excessive residual drift and drift concentration in a single
way to overcome seismic-induced damage to structures is to utilize story were observed through experimental tests and numerical analysis
energy-dissipation devices that are decoupled from the gravity-resisting of the BRB system [15,16]. As another type of metallic damper, the shear
system [3,4]. Under seismic excitations, plastic deformation is concen­ panel dampers (SPD) dissipate the seismic input energy through shear
trated on specially designed energy-dissipation devices which can be deformation of the diaphragm steel plate welded inside a square or
easily replaced without disruption to the gravity-resisting system [5]. rectangular rigid boundary [17,18]. Chan [19] studied the performance
Passive energy dissipation system has been recognized as effective and of the SPD through a series of sub-assemblage tests and reported
inexpensive damping devices inclusive of metallic yielding, friction, excellent seismic performance of the damper. The typical failure modes
viscous, and viscoelastic damper [6,7]. for the test specimens include panel-flange weld fracture, panel crack,

* Corresponding author.
E-mail address: [email protected] (M. Vafaei).

https://doi.org/10.1016/j.engstruct.2023.117072
Received 20 October 2022; Received in revised form 27 September 2023; Accepted 17 October 2023
Available online 23 October 2023
0141-0296/© 2023 Elsevier Ltd. All rights reserved.
R. Ullah et al. Engineering Structures 298 (2024) 117072

and panel buckling. In some other metallic dampers such as honeycomb addition, Shao [46] proposed brace-type shear fuses (BSF), in which core
[20], slit dampers [21,22], and comb-teeth [23] seismic input energy is perforated plates were connected in series with brace through bridging
dissipated through in-plane flexural or shear yielding of the steel strips. plates at the location of the brace-gusset connection. BSF was designed
Yang [24] presented a kind of honeycomb structural fuse that absorbs to dissipate seismic input energy via shear and bending deformation of
energy via shear deformation. They also proposed equations for the its steel strips in perforated plates. Similarly, a metallic damper with a
design of the damper’s key parameters. Sahoo [25] proposed a com­ shear-yielding mechanism is introduced by Bakhshayesh [47] to
bined shear-flexure yielding damper (SAFYD) consisting of two X-plates enhance the hysteretic performance of the concentric brace. Despite the
yielding under flexural action, and a rectangular panel yielding under invention of many different types of metallic dampers, only a few have
shear action. The main purpose of the X-plate was to constrain the out- been used in real structures. Complex fabrication and installation
of-plan buckling of the shear panel. Although the X-plates increased methods, difficult post-earthquake replacement, complicated design
energy dissipation capacity and improved the deformation mode of the methods, and expensive fabrication costs are among the main reasons
damper, its failure was due to the out-of-plane local buckling. for the limited application of invented dampers. Therefore, more cost-
A well-known assembly of hourglass shape steel plates, known as effective dampers that have easy fabrication, installation, and design
ADAS damper [26], or triangular shape plates referred to as TADAS methods need to be developed.
damper [27], dissipate energy through flexural yielding of plates under This study proposes a new replaceable metallic damper named
seismic excitations. Moreover, U-shaped dampers [28,29], make use of sandwiched fuse damper (SFD), for improving the seismic performance
rolling-bending deformation to dissipate the input energy and enhance of concentrically braced frames (CBFs). The proposed damper is con­
the deformation capacity. Deng [30] employed U-shaped steel plates to nected to a brace from one end and to a gusset plate from the other end.
develop a crawler steel damper for bridges. Although the results The connection between the damper and the brace or gusset plate is
demonstrated good performance of the U-shaped steel plate damper, made by bolts, making its installation, inspection, and post-earthquake
plasticity concentration of the energy dissipated elements and through- replacement easy and simple. This damper dissipates seismic energy
depth rupture of the end plates were still observed at large displace­ through the flexural yielding of its fuses resulting in stable hysteresis
ments. The majority of these devices are installed on the soffit of a beam loops and excellent energy dissipation. A simple design procedure and
supported by chevron braces [31]. Therefore, under seismic excitation, ease of fabrication and installation are the main advantages of this
the beam of the frame structure not only is displaced horizontally but damper compared with existing metallic dampers. Moreover, the
also vertically, which causes the beams to be subjected to high addi­ employed fuses which are made of HSS profiles can be conveniently
tional forces [32]. replaced after earthquakes.
Some researchers invented metallic yielding dampers applied in In the subsequent sections, at first, the details of the proposed
diagonally braced frames [4,33]. Qu [34] introduced a diagonal brace damper are presented. Then, the analytically driven equations for the
with a replaceable U-shaped steel plate to dissipate the seismic-induced damper’s initial stiffness, yield strength, and corresponding displace­
energy. Gray [35] introduced the yielding brace system (YBS), in which ment are described. Finally, the conducted experiments are illustrated,
specially designed cast steel triangular fingers were installed at the end and the obtained results are discussed.
of concentric braces. YBS dissipates seismic input energy via the flexural
yielding of triangular fingers. Palermo et al. [36] proposed the Cresent 2. Sandwiched fuse damper (SFD)
Shaped Brace (CSB), a bent steel brace at mid-length to produce a
yielding lever arm. The authors explored the cyclic behavior of a single 2.1. Details and working principles of SFD
diagonal CSB extensively and discovered its asymmetric hysteresis
response due to non-linear geometrical factors such as hardness in ten­ The geometry of the replaceable sandwiched fuse damper (SFD) is
sion and softening in compression [37]. In another study, installing CSBs shown in Fig. 1. An SFD unit consists of fuses (cut-lengths of HSS profile)
in chevron configuration over two adjacent frames, produced an overall that are sandwiched between two capping plates. The connection be­
symmetric hysteretic response under cyclic load, since one CSB worked tween the HSS profile and the capping plate is through welding. The
in compression and the other in tension [38]. Hsu and Halim [39] capping plates are slotted to provide access for a fillet weld that connects
proposed a Steel Curve Damper (SCD), a laser-curt curved steel plate of them to HSS profiles. The SFD unit is connected to a gusset plate and a
variant geometries placed at the beam-to-column connection to improve brace through bolts so that after a strong seismic event one can easily
frames’ strength and stiffness. The authors also employed the SCD in the open the SFD and replace it with a new one. One of the capping plates, as
middle of a brace member through a hinged connection, to amplify the shown in Fig. 1, directly joins the SFD to the gusset plate via bolts.
brace’s deformation capacity [40]. The geometrical configuration of the Through a mounting plate and an I-shaped part, the second capping
brace in the frame was considered such that it provided a lever arm plate links the SFD to the brace. An I-shaped section’s web is welded to
mechanism in the damper under the axial deformation of the brace. the end of the brace via longitudinal slots, and its flange is bolted to the
Dehghani [41] proposed a curved damper truss moment frame (CDTMF) mounting plate. It should be noted that all of the bolts used in the
for enhancing the seismic performance of a long-span moment frame by connections (gusset plate-capping plate, capping plate-mounting plate,
introducing the curved damper with the help of steel trusses into a semi- and mounting plate-flange of I-shaped section) are post-tensioned to
rigid beam-column connection. prevent sliding between steel plates.
Similarly, Demir [32] designed a saw-type brace with a series of As soon as the axial force of a brace is transferred to the mounting
triangular plates, capable of dissipating energy by flexural yielding. plates, a relative movement is imposed on the webs of HSS profiles
Jarrah [42] introduced the piston metallic damper (PMD), which is resulting in their double curvature flexural yielding and energy dissi­
comprised of a set of parallel hollow circular plates that interconnect the pation in the damper as illustrated in Fig. 2. The yielding of webs leads
inner shaft of the PMD to its outer pipe. In these devices, flexural to large plastic deformation and may deviate the capping plates from
yielding is transformed into tensile yielding at large displacements, their parallel configuration. Therefore, the pure flexural yielding of
resulting in a significant increase in the stiffness and strength of the webs is altered to a flexural-tensile mechanism. In such conditions, the
system. In another study, Maleki [43] introduced the dual-pipe damper SFD exhibits an increase in its strength due to the strain-hardening
(DPD), which formed a tension diagonal in the middle of the damper at behavior. To obtain the desired strength and stiffness of the SFD sys­
large displacements. Guo [44] incorporated a varied yielding section tem, several numbers and variant shapes of the HSS profile can be
referred to as an S-shaped steel damper (VYSSD) and Tabatabaei [45] utilized.
used a bar-fuse damper at the location of brace-gusset plate connection, The damper can be built in a workshop and then fitted to the
to provide a damper with a more stable flexural-tensile behavior. In structure’s diagonal bracing on-site during or after construction. As a

2
R. Ullah et al. Engineering Structures 298 (2024) 117072

Components of SFD A replaceable SFD

Gusset Plate
Bolt Hole Slot for a fillet weld
F/2=P

F F

F/2=P HSS Tube

HSS Brace Capping Plate
Mounting Plate Bolt I-Section

Fig. 1. Components of SFD unit and installation.

replaced by right-angle corners, and a simplified mechanical model in

the form of a two-legged frame was used for calculations as shown in
Fig. 4(b). The corresponding three-fold statistically indeterminate sys­
tem is presented in Fig. 4(c).
The axial force, shear force, and bending moment at the supports
were shown by X1 , X2 and X3 , respectively. The force method was used
to analyze the statistically indeterminate frame. In this method,
compatibility and equilibrium equations are used to determine the un­
known support reactions [48]. For the given frame in Fig. 4(c), the
following compatibility equations can be written:
δ11 X1 + δ12 X2 + δ13 X3 + Δ1C = 0
Fig. 2. Working mechanism of SFD.
δ21 X1 + δ22 X2 + δ23 X3 + Δ2C = 0 (1)
result, overhead welds are not required on-site, ensuring high con­
struction quality. The primary goal of this research is to employ the SFDs δ31 X1 + δ32 X2 + δ33 X3 + Δ3C = 0
in chevron or diagonal-type braces. As illustrated in Fig. 3, the SFD can
where, δij are the resulting displacements along the direction of Xi under
be attached at one or both ends of a diagonal brace. It can also be used to
link coupling walls or as a connection between precast walls and the action of unit load in Xj direction; ΔiC are the displacements of the
columns. determinate frame along the direction of Xi due to the support motion.
For a channel-shaped steel plate shown in Fig. 4(a), only displacement
along the longitudinal direction is permitted; therefore, Δ2C = Δ3C = 0,
2.2. Derivation of analytical equations of SFD and Δ1C = Δ. Besides, since the flexural yielding dominated the cyclic
response of the SFD, the axial and shear deformation will have an
In this section, some analytical equations are derived to estimate the insignificant effect on the results and can be ignored. Therefore, the
elastic stiffness and the yield point (yield strength and its corresponding required displacements can be calculated using the virtual work method
displacement) of the SFD. As mentioned earlier, a single SFD unit con­ as given below:
sists of one or more HSS profiles sandwiched between capping plates.
∑ ∫ M2 ∫L 2 ∫R 2
These HSS profiles work together in series, consequently, they linearly δ11 = 1
dx =
x2
dx2 +
L
dx3 =
L3 L2 R
+ (2)
contribute to the elastic stiffness and the yield point of an SFD unit. L EI 0 EI 0 EI 3EI EI
Therefore, for the derivation of the equations, only an HSS profile is
∑ ∫ M1 M2 ∫L ∫R
needed to be considered. Moreover, a single HSS profile when welded to δ12 = δ21 = dx =
− Rx2
dx2 +
− L(R − x3 )
dx3
capping plates provides two symmetrical channel-shaped elements that L EI 0 EI 0 EI
control the stiffness and yield point of the SFD, as shown in Fig. 4(a). L2 R LR2
= − − (3)
When the axial force of the brace is transferred to the SFD, it results in a 2EI 2EI
relative movement between the upper and lower flanges of the channel-
shaped element (see Δ in Fig. 4(a)). It should be noted under the relative
movement the HSS webs experience a double curvature bending
resulting in two plastic hinges at the corners. For the sake of simplifi­
cation, the small round corners of the channel-shaped element were

Mounting Plate

HSS Brace
Gusset plate

Gusset plate

I-Section Brace

SFD Unit Mounting Plate

Fig. 3. Illustration of proposed damper connected to a brace.

3
R. Ullah et al. Engineering Structures 298 (2024) 117072

(a) Yielding mechanism (b) Simplified model description (c) Statistically indeterminate frame
Fig. 4. Mechanical model employed for the derivation of analytical equations.

∑ ∫ M2 ∫R 2
x1
∫L 2
R
∫R
(R − x3 )2
2
δ22 = dx = dx1 + dx2 + dx3
L EI 0 EI 0 EI 0 EI
2R3 LR2
= + (4)
3EI EI

∑ ∫ M2 ∫R 2
1
∫L 2
1
∫R 2
1 2R L
δ33 = 3
dx = dx1 + dx2 + dx3 = + (5)
L EI 0 EI 0 EI 0 EI EI EI

∑ ∫ M1 M3 ∫L
x2
∫R
L L2 LR
δ13 = δ31 = dx = dx2 + dx3 = + (6)
L EI 0 EI 0 EI 2EI EI

∑ ∫ M2 M3 ∫R
− x1
∫L
− R
∫R
− (R − x3 )
δ23 = δ32 = dx = dx1 + dx2 + dx3
L EI 0 EI 0 EI 0 EI
R2 RL
= − −
EI EI
(7)

By substituting Eqs. (2)–(7) into Eq. (1), we can get: Fig. 5. Internal force diagrams of the frame model subjected to a sup­
( 3 ) ( ) ( ) port motion.
2L + 6L2 R X1 − 3L2 R + 3LR2 X2 + 3L3 + 6LR X3 = − 6EIΔ (8)
( ) ( ) ( )
− 3L2 R + 3LR2 X1 + 4R3 + 6LR2 X2 − 6R2 + 6LR X3 = 0 (9) calculated as:

( ) ( ) 6EIΔy (L + 2R) κbt2 σ y

L2 + 2LR X1 − 2R2 + 2LR X2 + (4R + 2L)X3 = 0 (10) Mmax = X3 = = (13)
L3 2
+ 8L R + 12LR 2 6

Solving Eqs. (8), (9), and (10) simultaneously provide the support where Δy is the yield displacement of the plate; κ is the shape factor of
reactions: the section, which is adopted as 1.5 for the rectangular cross-section of
− 12EIΔ(L + 2R) 6EIΔ(L + 2R) HSS profiles [4950], σy is the yield stress of the HSS profile. By solving
X1 = , X2 = 0, X3 = 3 (11) Eq. (13), the yielding displacement of SFD is obtained:
L4 + 8L3 R + 12L2 R2 L + 8L2 R + 12LR2
σ y (L3 + 8L2 R + 12LR2 )
Subsequently, the shear force and bending moment diagram of the frame Δy = (14)
Et(L + 2R)
segment can be drawn as shown in Fig. 5.
Therefore, the initial stiffness of an SFD can be estimated from the
From Eq. (12) and Eq. (14), the yield force of SFD can be estimated as:
following equation:
bt2 σy
X1 12EI(L + 2R) Ebt3 (L + 2R) Fy = Δy ke = (15)
ke = = = (12) L
Δ L4 + 8L3 R + 12L2 R2 L4 + 8L3 R + 12L2 R2
It should be noted that the above equations have been obtained for one
where E is the elastic modulus of steel material; I is the moment of web of an HSS profile. Depending on the number of HSS profiles in an
inertia; R is the radius of the HSS profile’s round corner; L(= H − t) is the SFD unit, these equations should be modified by multiplying them by the
height; b is the width, and t is the thickness of the HSS profile. For number of HSS profiles’ web.
estimating the yield force of the SFD, it was assumed that the plate The brace of the SFD-brace system should be designed and con­
deems to yield if the bending moment exceeds the elastic limit of the structed such that it remains in the elastic range while the SFD fuse
cross-section. Based on Eq.(11), the plastic moment (Mmax ) can be experiences large plastic deformations. Therefore, the axial buckling

4
R. Ullah et al. Engineering Structures 298 (2024) 117072

strength of the brace should be larger than the ultimate strength of the Cauchy similitude law [53]. The dimensions of the scaled frame are
damper to ensure the occurrence of such behavior in the frame. For this shown in Fig. 7.
purpose, a safety factor equal to 20% is suggested to consider the effects The experimental program consisted of four specimens of brace with
of strain hardening and residual stresses during the welding process. SFD (SFD-brace) and two specimens of bare brace as a reference struc­
Furthermore, the employed fuses in the SFD-brace system should yield ture for comparison. The test specimens were designed and fabricated
earlier than the surrounding frame. This consideration is to avoid according to AISC 341–16 seismic provisions [54] and the bare brace
damage to the frame before the initiation of plastic deformation in fuses. specimens were designed for the highly ductile brace member. The
Besides, since the flexural yielding of the SFDs controls the deformation slenderness ratio of bare brace specimens (for B1 = 55 and B2 = 40) and
of the SFD-brace system, the brace should possess a relatively large axial their sectional width-to-thickness ratio (for B1 = B2 = 10.9) satisfied the
stiffness to concentrate deformations in fuses. In short, the SFD-brace requirements of seismic provisions for a highly ductile brace (i.e.,
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
system should be designed based on the following criteria: KL/r ≤ 200, b/t ≤ 0.65 E/Ry Fy ). According to the capacity design
principle stipulated in the seismic provisions, the gusset plate, brace-
(1) The design axial deformation of the SFD-brace (Δd,d ) should be gusset plate weld connections, and reinforcement of the slotted ends
less than the frame’s yield displacement (Δf,y ), as shown in Fig. 6 of the brace were designed for the ultimate tensile strength of the brace.
and Eq. (16). To design the SFDs, at first, the compressive buckling capacity of the
scaled-down braces was estimated. Then, the ultimate flexural capacity
Δd,d < Δf ,y (16)
of the SFD fuses was equalled 80% of the compressive buckling capacity
of the braces. An over-strength factor of 1.2 was assumed to estimate the
(2) The ultimate flexural capacity of SFD (Fu ) should be at least 20% ultimate flexural strength of SFDs from their yield strengths. Following
less than the compressive buckling capacity of the brace (Pcre ). the AISC standard, the compressive buckling strength of the scaled-
down brace was obtained 42 kN, consequently, the ultimate flexural
Fu < 0.8Pcre (17)
capacity of the SFD was considered 33.6 kN, 20% smaller than the
compressive buckling capacity. Besides, the ratio of brace to fuse stiff­
(3) The axial stiffness of the brace (Kb ) should be significantly (i.e., 2 ness was considered to be larger than 2 so that the majority of defor­
to 3 times) larger than the flexural stiffness of the SFD fuse (Kd ). mation concentrated on the SFD fuses. Moreover, a story drift of 0.5%
was set for the initiation of yielding in the SFD fuses.
Kb ≥ Kd (18)
All other parts of the SFD-brace (capping plates, mounting plates, I-
section, weld, and bolt connections) were designed based on the brace’s
3. Experimental program
ultimate tensile strength to remain in the elastic range during the tests.
In the design and fabrication of test specimens of SFD-brace speci­
3.1. Test specimens
mens, two parameters were considered as variables. The first parameter
was related to the number of fuses (i.e., HSS profiles) in the SFDs. Two
The test specimens were assumed to be extracted from a prototype
types of SFDs were fabricated; the first type had only one fuse while the
three-story and three-bay office building located in high a seismicity
second type had two fuses as shown in Fig. 8. The second parameter was
region. As illustrated in Fig. 7, special concentrically braced frames
the place where SFDs were installed. As can be seen from Fig. 9, the
(SCBFs) in chevron configuration were assumed to be used in the
employed SFDs were installed once only at one end of the braces and
building as a lateral force-resisting system. The story height and bay
then at both ends. It should be mentioned that the total length of
width of the prototype building were considered 3 m and 6 m in both
employed fuses was similar in both types of SFDs. These two parameters
principal directions. The dead and live loads of the building were
were considered as variables to investigate their effect on the energy
assumed to be 6.25 kN/m2, and 3.75 kN/m2, respectively. The building
dissipation capacity and ductility ratio of SFD-brace systems.
was designed by the Equivalent Lateral Force procedure (ELF) according
Table 1 summarizes the dimensions of employed steel plates, braces,
to ASCE 7–16 [51]. In the design of the building, it was assumed that the
and SFDs, which have been described in Figs. 8 and 9. The test speci­
peak ground acceleration is 0.3 g, and the building is constructed on stiff
mens were given a name based on the number of employed SFDs and
soil, i.e., site class D. The structural elements’ sizes were determined
their installation location. For example, SFB1 refers to an SFD-brace
based on the calculated internal forces using AISC 360–16 [52]. The
system that comprises a single fuse damper (i.e., S) installed at one
bottom story of the building was assumed as the reference frame to be
side of the brace (i.e., 1), and DFB2 refers to an SFD-brace system that
equipped with the SFD-brace system. Based on the capacity of the
comprises a double fuse damper (i.e., D) at both ends of the brace (i.e.,
employed cyclic testing machine, a geometric scale of 1:4 was used to
2). Fillet-type welding with a throat size of 2 mm connected the HSS
determine the scaled-down sizes of the test specimens following the

Fig. 6. Lateral deformation of braced frame equipped with SFD fuse.

5
R. Ullah et al. Engineering Structures 298 (2024) 117072

Fig. 7. (a) Plan view of the prototype building (b) selected reference frame (c) scaled-down frame.

Fig. 8. Detail of single and double fuse SFD unit.

sections to capping plates through the provided slots. Besides, high- The lateral movement of braces was monitored by a Linear Variable
strength post-tensioned bolts with 8 mm diameters provided the Differential Transformer (LVDT). Besides, the probable relative move­
connection between the capping plates and the mounting plates. ment between the capping plate and the mounting plate was controlled
by another LVDT.
3.2. Test setup and loading history
4. Experimental results
The test setup of the conducted experiments is shown in Fig. 10. As
4.1. The failure mechanism of test specimens
can be seen from the figure, all test specimens were fixed within an MTS
testing machine with a 100 kN loading capacity. A similar quasi-static
All the specimens were subjected to the cyclic load depicted in
cyclic load was applied to all the test specimens through a
Fig. 11. The deformed shapes under various displacements of applied
displacement-controlled loading protocol. As shown in Fig. 11, the ATC
cyclic load and failure mode of the test specimens are shown in Figs. 12,
24 [55] loading protocol was adopted in this study. In this loading
13, and 14. As can be seen from these figures, the bare-brace specimens
protocol the yield deformation (i.e., Δy ) controls the amplitude of ex­
demonstrated visible buckling under the compression load and SFD-
citations in each cycle of loading. The yield deformation was calculated
brace specimens yielded symmetrically under tension and compression
as; Δy = σy Lb /E, where σ y was the yield strength of employed steel, Lb
excursions.
was the brace length, and E was the steel’s modulus of elasticity. The
Fig. 12(a) and (b) present the deformed shapes and failure mode of
employed steel’s yield strength and modulus of elasticity were deter­
bare-brace specimens, B1 and B2, respectively. As shown in Fig. 12(a),
mined through tensile strength test according to ASTM E8M-00 [56].
buckling of the B1 specimen was initiated when the axial displacement
Table 2 summarizes the obtained results from three tensile strength tests
approached 2 mm and it occurred exactly at the mid-length of the brace.
conducted on HSS fuses (denoted by T) and braces (denoted by B). In this
The buckling resulted in a local denting at the mid-length of the brace
table εu and σ u are respectively, the ultimate strain and strength while εy
and a crack initiated at the dented location when the axial deformation
and σ y are the yield strain and strength, respectively.
reached 6 mm. Subsequently, the B1 specimen developed a visible
As shown in Fig. 10 all test specimens were equipped with strain
fracture at the region where the crake occurred at the first tensile cycle
gauges. A strain gauge was installed on braces and another one on fuses.

6
R. Ullah et al. Engineering Structures 298 (2024) 117072

Fig. 9. Detail drawing of SFD-brace and bare brace test specimens.

of 8 mm displacement. In the case of B2, the buckling was initiated in the yielding webs of the left fuse. At the end of the test, the left fuse of the
middle of the brace when the axial displacement passed 2 mm. A crack SFB1 was completely detached from its capping plates.
was observed at the dented zone when the imposed displacement In the case of the SFB2, cracks were initially observed under the
reached 4 mm. Under the second tensile cycle of 5 mm displacement, the second cycle of 8 mm displacement at the bottom and top corners of the
B2 specimen developed a visible fracture at the mid-length of the brace, lower left fuse. These cracks were propagated deeper and longer during
as shown in Fig. 12(b). the subsequent cycles of 10 mm displacement. Finally, the lower left fuse
The deformed shapes and failure modes of SFB1 and SFB2 under was fractured along the observed cracks under the second cycle of 10
different displacements are shown in Fig. 13(a) and (b). The cracks were mm displacement. As shown in Fig. 13(b), the fractures have occurred at
initiated at the top and bottom corners of the left fuse at the second cycle the top and bottom of the left fuse. Similar to the SBF1, at the end of the
of 6 mm displacement. The cracks were propagated and got deeper and test, the lower left fuse of SFB2 was completely detached from its
longer under the subsequent cycle of 8 mm displacement and finally, the capping plates.
left fuse was fractured at the second cycle of 8 mm displacement. As Fig. 14(a) and (b), depict the deformed shapes and failure mode of
shown in Fig. 13(a), the fractures were observed at the top and bottom the DFB1 and DFB2 specimens, respectively. As can be seen from Fig. 14

7
R. Ullah et al. Engineering Structures 298 (2024) 117072

Table 1
Details of test specimens (unit: mm).
No Specimen Type SFD Brace

Fuse (HSS) Capping plate

H b t Lp Bp Tp HSS profile Lb

1 SFB1 Single fuse at one end of brace 32 100 2.3 110 130 5 25 × 25 × 1.6 530
2 SFB2 Single fuse at both ends of brace 32 100 2.3 110 130 5 25 × 25x1.6 380
3 DFB1 Double fuse at one end of brace 32 50 2.3 110 130 5 25 × 25 × 1.6 530
4 DFB2 Double fuse at both ends of brace 32 50 2.3 110 130 5 25 × 25 × 1.6 380
5 B1 Bare brace – – – – – – 25 × 25 × 1.6 530
6 B2 Bare brace – – – – – – 25 × 25 × 1.6 380

Notes: H is the height, b is the length, t is the thickness of the HSS profile, Lp is the length, Bp is the width, Tp is the thickness of the capping plate and Lb , represents the
length of steel braces. It is worth noting that the dimensions of gusset plates, cover plates, capping plates, and mounting plates are the same in all test specimens.

14
12
10
8
6

Displacement (mm)
4
2
0
-2
-4
-6
-8
Lateral Brace -10
LVDT Strain -12
Gauge -14
0 2 4 6 8 10 12 14 16 18 20 22
Cycles

Fig. 11. Employed cyclic load protocol [55].

Table 2
Mechanical properties of steel materials.
Sample E(GPa) σy (MPa) σu (MPa) εy (%) εu (%) Elongation (%)

T1 209 430.6 453.1 0.253 12.9 25.3

Axial
T2 197 425.7 463.7 0.304 12.1 23.6
LVDT
T3 215 422.2 496.5 0.285 16.1 31.6
Mean 207 426.1 471.1 0.28 13.7 26.8
B1 203 283.1 430.1 0.19 17.3 34.2
SFD B2 206 280.4 428.7 0.18 17.4 34.4
Strain B3 198 381.7 425.6 0.171 17.1 33.8
Gauge Mean 202 281.8 428.2 0.18 17.2 34.1

seen from Fig. 14(b), the fractures in both fuses are below the corners of
Fig. 10. Test setup. the HSS profile adjacent to the weld lines. At the end of the test, none of
the fractured fuses were completely detached from their capping plates,
(a), the DFB1 exhibited the initial cracks under the second compressive and at least one of the two webs of the fuses was still connected to the
cycle of 6 mm displacement at the top corner of the lower left fuse and capping plates.
the bottom of the lower right fuse. The cracks were propagated during It should be mentioned that none of the SFD-brace specimens
the subsequent loading cycles, resulting in the fracture of the lower fuses exhibited a buckling or yielding in their brace and plates (i.e., capping,
during the second cycle of 8 mm displacement. As shown, the fractures mounting, and gusset plates) before reaching the failure of their fuses.
in lower fuses are adjacent to the weld lines of the capping plates. At the Compared with bare brace specimens that experienced a sudden buck­
end of the test, none of the fractured fuses were completely detached ling, the SFD-brace specimens exhibited a gradual yielding with a larger
from their capping plates. deformation capacity. In addition, specimens with SFD at one end of the
In the case of the DFB2, the cracks were initially developed under the brace (i.e., SFB1 and DFB1) had a smaller deformation capacity than
first compressive cycle of 10 mm displacement at the top corner of the specimens with SFD at both ends of the brace (i.e., SFB2 and DBF2).
upper left fuse and the top corner of the upper right fuse. Under the By comparing the single-fuse (SFB1 and SFB2) and double-fuse
second tensile cycle of 10 mm displacement, the upper left fuse got (DFB1 and DFB2) SFD-brace specimens, it can be found that in the
fractured along the crack, and soon after, the fracture occurred at the single-fuse SFD-brace system, both webs of HSS fuses were fractured at
upper right fuse during the first cycle of 12 mm displacement. As can be the end of the tests such that one of the fuses was completely detached

8
R. Ullah et al. Engineering Structures 298 (2024) 117072

(a)

-1 mm -2 mm -4 mm -6 mm -8 mm +8 mm +8 mm 1st cycle

(b)

-1 mm -2 mm -4 mm +5 mm +5 mm 2nd cycle

Fig. 12. Deformation and failure mode of bare brace specimens (a) B1 and (b) B2. Note: The positive and negative signs indicate the tension and compression
excursions, respectively.

from the capping plate. However, in the double-fuse SFD-brace speci­ exhibited an in-cycle strength degradation under compression. The SFD-
mens, at least one web of a fuse was still connected to the capping plate, brace specimens, unlike bare braces, have shown a symmetrical
allowing for load transfer even after passing the ultimate load. This behavior under compression and tensile forces that indicate almost a
difference in the failure mechanism which was mainly due to the higher similar energy dissipation under compassion and tension. The SFD-brace
redundancy of double-fuse specimens increased the deformation ca­ specimens were yielded in flexure under relatively small displacement
pacity of double-fuse SFD-brace specimens compared with single-fuse and exhibited stable and plump hysteretic loops with stable post-yield
SFD-brace specimens. stiffness at large displacement. An increase in the displacement up to
the failure load has increased the energy dissipation of SFD-brace
specimens without any sign of stiffness degradation.
4.2. Hysteretic behavior
Fig. 15, further illustrates that the bare brace specimens reached
their ultimate tensile and compression strength at 3 mm and 2 mm
Fig. 15 depicts the force–displacement response of the tested speci­
displacement, respectively. The ultimate tensile and buckling load of
mens. As shown, the bare brace specimens B1 and B2 were buckled in
bare brace specimens were almost 54 kN and 44 kN, respectively. On the
compression under 2 mm displacement cycles. Thus, their strength and
other hand, the SFD-brace specimens reached their ultimate strength at
stiffness were progressively degraded under the subsequent loading
6 mm (i.e., SFB1 and DFB1) and 10 mm (i.e., SFB2 and DFB2) dis­
cycles. After buckling, the drop in the strength of the B1 specimen is
placements. Beyond their ultimate load, a gradual reduction in their
significantly larger than the B2 specimen. However, under tensile forces,
load-resisting capacity was noticed due to the initiation of cracks in the
the strength degradation of the B1 specimen starts at a larger displace­
fuses. The strength degradation in specimens SFB1 and DFB1 was
ment when compared with the B2 specimen. Besides, both bare braces

9
R. Ullah et al. Engineering Structures 298 (2024) 117072

(a)

-1 mm -2 mm -4 mm -6 mm -8 mm +8 mm -8 mm -8 mm 2nd cycle

(b)

-1mm -2 mm -4 mm -6 mm -8 mm -10 mm +10 mm +10 mm +10 mm 2nd cycle

Fig. 13. Deformation and failure mode of single-fuse SFD-brace (a) SFB1 and (b) SFB2.

initiated at the second cycle of 6 mm displacement. In the case of SFB2 obtained from the force–displacement hysteresis loops. It is seen that all
and DFB2 specimens, their strength degraded at the second cycle of 10 the SFD-brace specimens exhibit smaller initial stiffness compared with
mm displacement. No in-cycle strength degradation was observed for the bare brace specimens because of the lower stiffness of their fuses.
SFD-brace specimens. Specimens SFB1 and DFB1 failed at the 16th Besides, specimens with the SFD at one end of the brace (SFB1 and
loading cycle, and failure of specimens SFB2 and DFB2 occurred at the DFB1) exhibited relatively larger initial stiffness than the specimens
18th and 20th loading cycles, respectively. Therefore, since the total with the SFD at both ends of the brace (SFB2 and DFB2). However,
length of fuses was similar in all SFD-brace specimens, an increase in the similar initial stiffness was exhibited by single and double fuses SFD-
number of fuses enhanced the ductility of specimens through delay in brace specimens, which indicates that the number of fuses in an SFD
the initiation of cracks and fracture of fuses. The maximum strength of has no significant effect on their initial stiffness. It is seen that under
all the SFD-brace specimens was close to each other and in the range of compressive forces, the post-yield stiffness and strength of bare brace
31–34 kN, indicating that the number of fuses had an insignificant effect specimens degrade rapidly soon after the occurrence of the buckling.
on the ultimate loads. It should be noted that the ultimate loads of bare However, the post-yield stiffness and strength of the SFD-brace system
braces are larger than the SFD-brace system because the SFD units had a degrade gradually as the fuses yield.
smaller yield load than the buckling load of the bare braces. However, To determine the yield point of tested specimens, their backbone
the observed differences in the ultimate loads can be decreased by curves were tri-linearized using the equivalent energy method [57,47].
designing fuses for a yield load slightly less than the buckling load of the In the proposed method, the area under the trilinear model equaled to
braces. the area under the backbone curve. In addition, the trilinear model
Fig. 16(a), depicts the backbone curves of the test specimens intersected the backbone curve at an offset of 0.6 Fy , as shown in Fig. 16

10
R. Ullah et al. Engineering Structures 298 (2024) 117072

(a)

-1 mm -2 mm -4 mm -6 mm -8 mm +8 mm -8 mm 2nd cycle

(b)

-1 mm -2 mm -4 mm -6 mm -8 mm -10 mm +10 mm -12 mm -12 mm 1st cycle

Fig. 14. Deformation and failure mode of double-fuse SFD-brace (a) SFB1 and (b) SFB2.

(b). In this figure, Fy and Δy represent, respectively, the strength and addition, the maximum displacement of the one-end SFD-brace system
displacement of the effective yield point, and the slope of the curve in tensile and compressive directions are, respectively, 66.7% and 150%
before the yield point is the effective stiffness (Ke ). Fu shows the ultimate larger than that of the bare brace. This implies that SFDs have signifi­
load and KP is the post-yield stiffness equivalent to the ascending slope cantly increased the deformation capacity of the bare braces.
of the trilinear curve from the yield point to the ultimate strength. Δu is As shown in Table 3, the average ductility coefficient of the bare
the maximum displacement and equals to a displacement corresponding brace in tension is 3.85, and in compression equals 2.45. On the other
to 80% of the ultimate force on the descending slope of the tri-linearized hand, the SFD-brace specimens have an average ductility coefficient of
curve [57,47]. The obtained results from the linearization have been 4.07 in tension and 4.25 in compression. Therefore, the ductility of SFD-
presented in Table 3. In this table, the ductility coefficient (μ) is the ratio brace specimens under tensile force is somewhat greater than that of
of displacement corresponding to the 80% of the ultimate strength to the bare braces. The ductility of the SFD-brace under compression force, on
displacement at the effective yield point. the other hand, is much greater than that of the bare brace.
Table 3 shows that the bare braces had a larger ultimate load, yield The effective yield and ultimate displacements of both-end SFD-
strength, and effective stiffness than the SFD-brace systems. As designed, brace specimens (i.e., SFB2 and DFB2) are approximately 60% larger
the ultimate flexural strength of the fuse was to be smaller than the than one-end SFD-brace specimens for both directions. However, the
buckling load of braces to ensure the full yielding of the fuse before effective stiffness of both-ends SFD-brace specimens is 31.5% and 38%
brace buckling. On the other hand, the effective yield displacement of smaller than one-end SFD-brace specimens for tensile and compressive
the SFD-brace is slightly larger than the bare brace. However, the ulti­ direction, respectively. This observation indicated that when SFD units
mate displacement of SFD-brace specimens is significantly larger than were added to both ends of braces the yield and ultimate displacements
bare brace specimens. The yield displacement of one-end SFD-brace were increased while the effective stiffness was decreased. Results also
specimens (i.e., SFB1 and DFB1) in tensile and compressive directions show that the post-yield stiffness of one-end SFD-brace specimens is
are, respectively, 26.4% and 74% larger than that of bare brace (B1). In significantly larger than that of both-ends SDF-brace specimens. Table 3

11
R. Ullah et al. Engineering Structures 298 (2024) 117072

60 60
50 50
40 40
30 30
20 20
Force (kN)

Force (kN)
10 10
0 0
-10 -10
-20 -20
-30 -30
-40 -40
-50 -50
-60 -60
-10 -8 -6 -4 -2 0 2 4 6 8 10 -8 -6 -4 -2 0 2 4 6 8
Displacement (mm) Displacement (mm)
(a) B1 (b) B2
40 40
30 30
20 20
10 10

Force (kN)
Force (kN)

0 0
-10 -10
-20 -20
-30 -30

-40 -40
-10 -8 -6 -4 -2 0 2 4 6 8 10 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12
Displacement (mm) Displacement (mm)
(c) SFB1 (d) SFB2
40 40
30 30
20 20
10 10
Force (kN)

Force (kN)

0 0
-10 -10
-20 -20
-30 -30
-40 -40
-10 -8 -6 -4 -2 0 2 4 6 8 10 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14
Displacement (mm) Displacement (mm)
(e) DFB1 (f) DFB2
Fig. 15. Force-displacement hysteresis curve of the test specimens.

further illustrates that adding the SFD units to both ends of braces had no design stage of SFDs.
significant effect on the yield strength Fy , ultimate strength Fu , and Table 4 compares the obtained results from analytically driven
displacement ductility ratio of the SFD-brace specimens. It is also equations with the average values of those obtained from the conducted
noteworthy that single-fuse specimens (i.e., SFB1 and SFB2) have a close experiments. As can be seen, the yield displacements of SFB1 and DFB1
effective stiffness, yield and ultimate displacements, and effective yield have been estimated, respectively, 12% and 16% smaller than the
strength to that of double-fuse specimens (i.e., DFB1 and DFB2). experimentally obtained values. On the other hand, the yield displace­
Table 3 also shows that the ratio of ultimate strength to yield strength ments of SFB2 and DFB2 have been estimated, respectively, 13% and
of the SFD-brace specimens is nearly identical and is in the range of 11% larger than the results of conducted experiments. In addition, the
1.13–1.33. This range is close to the assumption made at the design stage estimated elastic stiffness of SFB1 and DBF1 are respectively, 19% and
(i.e., 1.2). Moreover, comparing the compressive yield capacity of bare 18% larger than the experimental results. For the SFB2 and DFB2, the
braces (i.e., 38.8–39.8 kN) with the ultimate strength of SFD-brace estimated elastic stiffness is smaller than the test values by 6% and 9%,
specimens (i.e., 28.9–34 kN) shows an average safety margin of 24%, respectively. It is noteworthy that the theoretical yield forces of SFB1
which reasonably agrees with the 20% safety margin employed in the and DFB1 are, respectively, 9% and 4% larger than their test results.

12
R. Ullah et al. Engineering Structures 298 (2024) 117072

60
B1
50
B2
Force (kN)
40 SFB1 Backbone curve
30 SFB2
20 DFB1 Trilinear curve
Force (kN)

DFB2
10
0
-10
-20
Displacement (mm)
-30
-40
-50
-60
-4 -3.5 -3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4
Story drift ratio (%)
(a) Backbone curves of the test specimens (b) Trilinear model of backbone curves
Fig. 16. Force-displacement backbone curves of the test specimens.

Table 3
The main mechanical parameters of the tested specimens.
Excursion Specimen Δy (mm) Fy (kN) Ke (kN/mm) Δu (mm) Fu (kN) KP (kN/mm) Fu /Fy KP /Ke μ
Tension SFB1 1.78 25.6 14.38 7.58 34 2.61 1.33 0.18 4.3
DFB1 1.91 27.3 14.29 8 33.3 1.94 1.22 0.14 4.2
SFB2 2.63 25.9 9.85 10 31.5 1.04 1.22 0.11 3.8
DFB2 2.81 27.7 9.86 11.2 33.6 1.14 1.21 0.12 4
B1 1.49 42.1 28.26 5.95 53.4 7.48 1.27 0.26 4
B2 1.43 42.1 29.44 5.28 54.9 8.15 1.3 0.28 3.7
Compression SFB1 1.61 25.4 15.78 8 33.8 2.48 1.33 0.16 5
DFB1 1.63 26.6 16.32 6.7 31.6 1.48 1.19 0.09 4.1
SFB2 2.62 25.6 9.77 10 28.9 0.61 1.13 0.06 3.8
DFB2 2.61 26.8 10.27 10.75 31.1 0.8 1.16 0.08 4.1
B1 1.08 38.8 35 2.46 43.6 6.3 1.15 0.18 2.3
B2 0.98 39.8 36.5 2.51 45.4 9.51 1.27 0.26 2.6

Table 4
Comparison between estimated and experimentally obtained results.
Specimen Δy,cal (mm) Fy,cal (kN) Ke,cal (kN/mm) Δy,exp. (mm) Fy,exp. (kN) Ke,exp. (kN/mm) Δy,exp. /Δy,cal Fy,exp. /Fy,cal Ke,exp. /Ke,cal

SFB1 1.52 28.04 18.52 1.7 25.5 15.05 1.12 0.91 0.81
DFB1 1.52 28 18.48 1.77 26.95 15.23 1.16 0.96 0.82
SFB2 3.04 28.04 9.26 2.63 25.75 9.81 0.87 0.92 1.06
DFB2 3.04 28 9.24 2.71 27.25 10.06 0.89 0.97 1.09

1
Force
0.9
Dissipated energy ratio (Eloop/EKH)

0.8

0.7

0.6

0.5

0.4
Displacement B1
0.3 B2
SFB1
0.2 SFB2
DFB1
0.1 DFB2

0
0 1 2 3 4 5 6 7 8 9 10 11 12
Displacement (mm)
(a) Energy ratio concept (b) Energy ratio
Fig. 17. Hysteresis energy dissipation capacity of the test specimens.

13
R. Ullah et al. Engineering Structures 298 (2024) 117072

Similarly, the theoretical yield force of SFB2 and DFB2 specimens are observation is because the yield deformation of one-end SFD-brace
8% and 3% larger than their experimental values, respectively. In short, specimens is smaller than the both-ends SFD-brace specimens. It is also
this can be concluded that the obtained equations can estimate the yield noteworthy that the energy dissipation per unit volume of SFD is
point and elastic stiffness of the SFD-brace systems with good accuracy. increased exponentially with an increase in the displacement which is
The observed variations can be attributable to the simplified mechanical similar to other dampers such as Bar-fuse [60], S-shaped [49], Piston
model used, the assumptions made, and the changes in actual yield [42], Dual-pipe [61], and Honeycomb metallic damper [24]. Moreover,
stress of the HSS profile owing to residual stresses imparted during the to compare the energy dissipation and deformation capacity of the
manufacturing process. The highest value of the transverse residual proposed SFD with other dampers, their cumulative energy per unit
stress distributions in the hot-formed square hollow section, according volume and cumulative drift ratios were calculated and shown in Fig. 18
to Abathun [58] was 14.1% of the material yield stress at the corner and (b). The cumulative drift ratio was calculated using the imposed
16.6% of the material yield stress in the flat sections that need to be displacement in the direction of loading divided by the clear height of
considered when designing the SFDs. the yielding web of each damper. It is seen that the SFD specimens have
an excellent cumulative energy dissipation per unit volume with a
relatively large cumulative drift ratio, putting them among metallic
4.3. Energy dissipation
dampers with great deformation capacity.
The equivalent damping ratios (ξeq ) of test specimens were also
The energy dissipation capacity of the test specimens was computed
calculated and compared. The calculations were made based on the
as the area enclosed within the hysteresis loops at a particular
following equations and the principle shown in Fig. 19(a).
displacement cycle. Since the ultimate loads of the bare braces and SFD-
brace specimens were not similar their dissipated energy was normal­ 1 Eloop 2 Eloop
ξeq = = (25)
ized using an energy ratio, as shown in Fig. 17(b). The energy ratio is the 4π Ees π Keq (|Δ+ | + |Δ− | )2
ratio of the hysteretic energy absorbed by the test specimen (Eloop ) to the
hysteretic energy absorbed by an equivalent Kinematic-Harding hys­ (|P+ | + |P− | )
Keq = (26)
teretic model (EKH ), subject to the same cyclic loading, as illustrated (|Δ+ | + |Δ− | )
graphically in Fig. 17(a). The energy ratio varies from one for the
kinematic-hardening model to zero for the nonlinear-elastic model [59]. where, Eloop is the energy dissipated in on hysteresis loop of the test
It can be seen that the energy ratio of the bare braces B1 and B2 reduced specimen, and Ees is the corresponding elastic strain energy; P+ and P−
abruptly when they reached their ultimate load at 3 mm displacement. are the maximum and minimum force respectively; Δ+ and Δ− are the
Consequently, the energy dissipation ratio of B1 and B2 dropped to 0.4 maximum and minimum displacement, respectively [49].
and 0.31 at the failure of specimens, respectively. In contrast to the bare Fig. 19(b) compares the equivalent damping ratio of test specimens.
braces, the energy dissipation of the SFD-brace specimens showed an As shown, the equivalent damping ratio of the bare braces increased
increase after 3 mm displacement and remained almost constant after rapidly during small displacement cycles compared to SFD-brace spec­
passing their ultimate load. At the end of the test, the energy ratios of imens. The bare brace specimens B1 and B2 reached their maximum
SFD-brace specimens SFB1, DFB1, SFB2, and DFB2, are 0.61, 0.56, 0.66, equivalent damping ratio of 26% and 29.5% at a displacement of 4 mm.
and 0.60, respectively. However, beyond this displacement, their equivalent damping ratio was
Fig. 18(a) presents the relationship between cumulative dissipated significantly reduced. On the other hand, the equivalent damping ratio
energy per unit volume and cumulative displacement of the test speci­ of SFD-brace specimens progressively increased with increasing loading
mens. As shown, the bare brace specimens dissipated more energy cycles, until the failure of the specimens. Moreover, the one-end SFD-
during small displacement cycles compared to SFD-brace specimens. brace specimens (SFB1 and DFB1) had a larger equivalent damping ratio
However, due to their buckling, the energy dissipations of bare braces compared to both-ends SFD-brace specimens (SFB2 and DFB2) up to 8
were reduced when the displacement of cyclic loading increased. mm displacement. The specimens SFB1 and DFB1 reached their
Moreover, the energy dissipation per unit volume of one-end SFD-brace maximum equivalent damping of 29% and 30% at a displacement of 8
specimens (i.e., SFB1 and DFB1) is larger than the both-ends SFD-brace mm. However, the specimens SFB2 and DFB2 obtained their maximum
specimens (i.e., SFB2 and DFB2) under small displacement cycles. This

0.09 0.2

0.18
Cumulative energy per unit volume (J/mm3)

0.08
Cumulative energy per unit volum (J/mm3)

0.07 0.16

0.14
0.06
0.12 SFB1
0.05 SFB2
0.1
DFB1
0.04
B1 0.08 DFB2
0.03 B2 Bar-fuse damper [50]
SFB1 0.06
Dual-pipe damper [46]
0.02 SFB2
0.04 Piston-metallic damper [36]
DFB1
S-shaped damper [51]
0.01 DFB2 0.02
Honeycomb damper [24]
0 0
0 50 100 150 200 250 300 350 400 450 0 2 4 6 8 10 12 14 16 18
Cumulative displacement (mm) Cummulative drift ratio

(a) Cumulative displacement (b) Cumulative drift ratio

Fig. 18. Cumulative dissipated energy per unit volume of the test specimens.

14
R. Ullah et al. Engineering Structures 298 (2024) 117072

Force 0.35

0.3

Equivalent damping ratio (ξeff)

1
0.25

0.2

0.15 B1
B2
0.1 SFB1
Displacement SFB2

0.05 DFB1
DFB2

0
0 1 2 3 4 5 6 7 8 9 10 11 12
Displacement (mm)
(a) Equivalent damping ratio concept (b) Equivalent damping ratio
Fig. 19. Equivalent damping ratio of the test specimens.

equivalent damping of 32% and 34% at displacements of 10 mm and 12 4.5. Residual displacement
mm, respectively. The maximum equivalent damping ratio of the SFD
specimens is 0.34. The SFD specimens possess an equivalent damping The residual displacement of the test specimens is shown in Fig. 21.
ratio of 28% at the ultimate load of the damper, which is almost similar Because the specimens are in the elastic range, the residual
to other metallic dampers such as PMD (27%) [42], S-shaped (25%)
[49], and DPD (23%) [61]. 10
B1
8 B2
Residual displacement (mm)

4.4. Stiffness degradation 6 SFB1

SFB2
4 DFB1
The stiffness degradation curve is an important index of a metallic
damper from the perspective of structural design. To compare the 2 DFB2

stiffness degradation curves, the equivalent stiffness of the test speci­ 0

mens was normalized by their initial stiffness and plotted in Fig. 20(a). It -2
is seen that the SFD-brace specimens exhibit a lower stiffness degrada­ -4
tion rate compared with the bare braces. Moreover, the stiffness
-6
degradation of one-end SFD-brace specimens (i.e., SFB1 and DFB1) is
-8
slightly faster than that of both-ends SFD-brace specimens (i.e., SFB2
and DFB2). As shown in Fig. 20(b), the bare-brace specimens B1 and B2 -10
lost 50% of their initial stiffness, respectively, at displacements of 2.62 -4 -3.5 -3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4
mm and 2.42 mm. In the case of one-end and both-end SFD-brace Story drift ratio (%)
specimens, the 50% reduction of their initial stiffness occurred at a
Fig. 21. Residual displacement of the tested specimens.
displacement between 3.16 mm and 3.77 mm and 5.44 mm to 5.57 mm.

8
1
25% Degradation of intial stiffness
B1 7 50% Degradation of initial stiffness
Normalised stiffness (K/Kintial)

B2
0.8 SFB1 6 5.57
5.44
Displacement (mm)

SFB2
DFB1 5
0.6 DFB2
3.9 3.77
4
3.16 3.23
0.4 3 2.62 2.42
2.17 2.11
2 1.67
1.42
0.2
1

0 0
0 1 2 3 4 5 6 7 8 9 10 11 12 B1 B2 SFB1 SFB2 DFB1 DFB2
Displacement (mm) Test specimen

(a) Normalized stiffness degradation (b) Displacement at percent degraded stiffness

Fig. 20. Equivalent stiffness degradation of the test specimens.

15
R. Ullah et al. Engineering Structures 298 (2024) 117072

displacements of the test specimens are modest up to a drift ratio of recorded during tests using uniaxial strain gauges. As depicted in Fig. 10,
0.5%. From this point onward, the residual displacements of the bare three strain gauges were installed on each SFD-brace specimen. For one-
brace specimens are marginally larger than those of the SFD-brace end SFD-brace specimens, two strain gauges were installed on the bot­
specimens. Also, the residual displacements of SFD-brace specimens tom of the right and left fuses, and one on the center of the brace. In the
are somewhat larger than those of bare brace specimens at the ultimate case of both-end SFD-brace specimens, two strain gauges were installed
loads. Notably, when the drift ratio exceeds 1%, the residual displace­ on the bottom left and top left fuses and another one on the center of the
ments of SFD-brace specimens are around 1 mm. In other words, more brace.
than 70% of the enforced drift has been temporary, with less than 30% Fig. 22 displays the maximum measured strain in the brace (i.e., SFD-
being permanent. Brace-SG) and fuse (i.e., SFD-Fuse-SG) of SFD-brace specimens in com­
Under the same amplitude of imposed displacement, the residual parison to that of the bare brace (i.e., Bare Brace-SG). It can be seen from
displacements of both-end SFD-brace specimens (i.e., SFB2 and DFB2) the displacement–strain relationships that the bare braces reached their
are relatively smaller than one-end SFD-brace specimens for both di­ yield strain in tension and compression directions at 1.69 mm and 1 mm
rections. However, the residual displacements of both-ends SFD-brace displacements, respectively. On the other hand, in SFD-brace specimens,
specimens are approximately 17% to 45% larger than one-end SFD- the fuses of one-end and both-end SFD-brace specimens reached their
brace specimens at the maximum displacement of loading cycles. Since yield strain at 1.55 mm and 2.75 mm displacements, respectively. After
both-end SFD-brace specimens underwent larger displacement the yielding of fuses, their strain increased gradually till the initiation of
compared to one-end SFD-brace specimens, they sustained compara­ crakes, beyond that it decreased with the propagation of cracks until the
tively larger residual displacement. On the other hand, the single-fuse rupture occurred.
specimens (i.e., SFB1 and SFB2) have nearly identical residual dis­ Fig. 22 further illustrates that none of the braces of SFD-brace
placements to that of double-fuse specimens (i.e., DFB1 and DFB2). This specimens reached their yield strain even after the failure of their
implies that the number of fuses has no significant effect on the residual fuses. The ratio of strain in the fuse to strain in the brace of SFB1, SFB2,
displacement of the SFD-brace system. DFB1, and DFB2, at the ultimate load, are approximately 4, 2, 5.5, and 2,
respectively. This implies that the fuses of the SFD-brace systems have
provided a safety margin for their brace to remain in the elastic zone
4.6. Strain analysis when reaching the ultimate states. Moreover, the fuse-to-brace strain
ratio of one-end SFD-brace specimens is 2 to 2.5 times larger than that of
The strain variation in fuses and braces of SFD-brace specimens was

4000 4000
SFD-Brace-SG SFD-Brace-SG
3000 3000
SFD-Fuse-SG SFD-Fuse-SG
2000 2000
Bare Brace-SG Bare Brace-SG
1000 1000
Strain (με)

Strain (με)

0 0

-1000 -1000
Brace Yield Strain Brace Yield Strain
-2000 Fuse -2000 Fuse

-3000 -3000

-4000 -4000
-8 -6 -4 -2 0 2 4 6 8 -10 -8 -6 -4 -2 0 2 4 6 8 10
Displacement (mm) Displacement (mm)
(a) SFB1 (b) SFB2
4000 4000
SFD-Brace-SG SFD-Brace-SG
3000 3000
SFD-Fuse-SG SFD-Fuse-SG
2000 2000
Bare Brace-SG Bare Brace-SG
1000 1000
Strain (με)

Strain (με)

0 0

-1000 -1000
Brace Yield Strain Brace Yield Strain
-2000 Fuse Yield Strain
-2000 Fuse Yield Strain

-3000 -3000

-4000 -4000
-8 -6 -4 -2 0 2 4 6 8 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12
Displacement (mm) Displacement (mm)
(c) DFB1 (d) DFB2
Fig. 22. Relationships between measured strains and imposed displacements.

16
R. Ullah et al. Engineering Structures 298 (2024) 117072

both-end SFD-brace specimens, which indicates that the one-end SFD- installed within a frame.
brace specimens better concentrate the plastic deformations to their
fuses. It is also noteworthy that the fuse-to-brace strain ratios of SFD- CRediT authorship contribution statement
brace specimens are larger than one even after passing the ultimate
load of the systems. This observation implies their contribution to en­ Rafiq Ullah: Conceptualization, Methodology, Investigation,
ergy dissipation even at large displacements. Writing – original draft. Mohammadreza Vafaei: Supervision, Formal
analysis, Visualization, Writing – review & editing. Sophia C. Alih:
5. Conclusions Supervision, Funding acquisition, Project administration. Abdul
Waheed: Investigation, Data curation.
This study presents a replaceable sandwiched fuse metallic damper
to enhance the seismic performance of the concentrically braced frames Declaration of Competing Interest
(CBF). The seismic performance of the SFD was investigated through a
series of displacement-controlled cyclic tests, in which the SFD was The authors declare that they have no known competing financial
installed within a brace and its cyclic response was compared with that interests or personal relationships that could have appeared to influence
of the bare brace. An analytical model was developed to estimate the the work reported in this paper.
SFD’s yield strength and its corresponding displacement. Experimental
results were employed to examine the efficiency of the proposed model. Data availability
Based on the obtained results the following conclusions can be derived:
Data will be made available on request.
(1) The SFD-brace system possessed stable hysteresis loops, large
deformation capacity, and excellent energy dissipation per unit Acknowledgments
volume compared to the bare brace.
(2) The failure mode of SFD-brace specimens was dominated by the The authors would like to acknowledge the the Ministry of Higher
flexural yielding of HSS profiles’ web. Because of their higher Education of Malaysia under Fundamental Research Grant Scheme with
redundancy, the double-fuse SFDs exhibited a more ductile fail­ registration number FRGS/1/2020/TK0/UTM/02/70 (Vot 5F365) and
ure mode and larger deformation capacity than the single-fuse FRGS/1/2022/TK06/UTM/02/33 (Vot 5F545).
SFDs.
(3) The SFD-brace system enhanced the ductility ratio of the bare
References
braces particularly under compression around 180% to 200%
larger than that of the bare brace. The single-fuse SFDs exhibited [1] Mirtaheri M, Gheidi A, Zandi AP, Alanjari P, Samani HR. Experimental
a similar displacement ductility ratio to that of the double-fuse optimization studies on steel core lengths in buckling restrained braces. J Constr
SFDs. Similarly, the installation of SFD units at both ends or Steel Res Aug. 2011;67(8):1244–53. https://doi.org/10.1016/j.jcsr.2011.03.004.
[2] Guo W, Ma C, Yu Y, Bu D, Zeng C. Performance and optimum design of replaceable
one end did not significantly change the displacement ductility steel strips in an innovative metallic damper. Eng Struct 2020; 205(December
ratio of SFD-brace systems. 2019): p. 110118, doi: 10.1016/j.engstruct.2019.110118.
(4) SFD-brace systems had a gradual cyclic strength deterioration [3] Soong TT, Jr BFS. Supplemental energy dissipation : state-of-the-art and state-of-
the- practice. Eng Struct 2002; 24; 243–259, 2002, doi: 24 (2002) 243–259.
rate with no in-cycle strength degradation. The double-fuse and [4] Symans MD, et al. Energy Dissipation Systems for Seismic Applications: Current
single-fuse SFDs exhibited a similar strength degradation rate. Practice and Recent Developments. J Struct Eng Jan. 2008;134(1):3–21. https://
(5) The cumulative energy dissipation per unit volume of the SFD- doi.org/10.1061/(ASCE)0733-9445(2008)134:1(3).
[5] Wada A, Iwata M, Huang H. Damage-controlled structures.I: Preliminary design
brace specimen was around 70% to 130% larger than that of methodology for seismically active regions. Eng Struct 1997; APRIL: 423–431, doi:
bare brace specimens. When the SFD units were installed only on 123:423-431.
one end of a brace, their cumulative energy dissipation per unit [6] Zheng J, Li A, Guo T. Analytical and experimental study on mild steel dampers with
non-uniform vertical slits. Earthq Eng Eng Vib Mar. 2015;14(1):111–23. https://
volume was 36% larger than SDF units that were installed on both doi.org/10.1007/s11803-015-0010-9.
ends of the brace. However, the single-fuse SFD units had a [7] Chen Y, Chen C, Jiang H, Liu T, Wan Z. Study of an innovative graded yield metal
similar cumulative energy dissipation per unit volume to that of damper. J Constr Steel Res Sep. 2019;160:240–54. https://doi.org/10.1016/j.
jcsr.2019.05.028.
the double-fuse SFDs. The cumulative drift ratios of the SFD-
[8] Aiken ID, Nims DK, Whittaker AS, Kelly JM. Testing of Passive Energy Dissipation
brace systems were comparable to other types of metallic Systems. Earthq Spectra Aug. 1993;9(3):335–70. https://doi.org/10.1193/
dampers with a high deformation capacity. 1.1585720.
(6) The SFD-brace systems exhibited a much slower stiffness degra­ [9] Vafaei M, Omar Sheikh AM, Alih SC. Experimental study on the efficiency of
tapered strip dampers for the seismic retrofitting of damaged non-ductile RC
dation rate than the bare braces. When the SFD units were frames. Eng Struct 2019;199(March):109601. https://doi.org/10.1016/j.
installed on both ends of a brace, their stiffness degradation rate engstruct.2019.109601.
was slower than the cases in which SFD units were installed on [10] Guo Y, Zhou P, Wang M, Pi Y, Bradford MA, Tong J. Experimental and numerical
studies of hysteretic response of triple-truss-confined buckling-restrained braces.
one end of the brace. Eng Struct Oct. 2017;148:157–74. https://doi.org/10.1016/j.
(7) The recorded strains indicated that the SFD units significantly engstruct.2017.06.058.
decreased the strain in the braces with a safety margin of 1.5 to [11] A. W. and M. F. Atsushi Watanabe, Yasuyoshi Hitomi, Eiichiro Saeki, “properties of
brace encased in buckling-restraining concrete and steel tube.” 1988.
2.5 and contributed to the energy dissipation even after passing [12] Wu B, Mei Y. Buckling mechanism of steel core of buckling-restrained braces.
their ultimate loads. J Constr Steel Res Apr. 2015;107:61–9. https://doi.org/10.1016/j.
(8) The analytically derived equations predicted the yield strength of jcsr.2015.01.012.
[13] Ghowsi AF, Sahoo DR. “Experimental Study of All-Steel Buckling-Restrained Braces
the SFD-brace systems with less than 10% difference from the Under Cyclic Loading. Geotech, Geol Earthq Eng 2019;47:67–80. https://doi.org/
actual values. The yield displacement and elastic stiffness, on the 10.1007/978-3-319-78187-7_6.
other hand, were estimated with maximum deviations of 16% [14] Mirtaheri SM, Nazeryan M, Bahrani MK, Nooralizadeh A, Montazerian L,
Naserifard M. Local and global buckling condition of all-steel buckling restrained
and 19%, respectively.
braces. Steel Compos Struct Feb. 2017;23(2):217–28. https://doi.org/10.12989/
scs.2017.23.2.217.
It should be mentioned that the comparisons made in this study were [15] Della Corte G, D’Aniello M, Landolfo R, Mazzolani FM. Review of steel buckling-
based on the tests conducted on the bare-brace specimens. Therefore, to restrained braces. Steel Constr Jun. 2011;4(2):85–93. https://doi.org/10.1002/
stco.201110012.
better understand the seismic response of the SFDs, further in­ [16] M. N. and K.-C. T. Chia-Ming Uang, “Research and application of buckling
vestigations should be carried out on the SFD-brace system when restrained braced frames. Int J Steel Struct 2004; 4(2004): 301-313.

17
R. Ullah et al. Engineering Structures 298 (2024) 117072

[17] Abebe DY, Jeong SJ, Getahune BM, Segu DZ, Choi JH. Hysteretic characteristics of [39] Hsu H, Halim H. Improving seismic performance of framed structures with steel
shear panel damper made of low yield point steel. S5–902-S5-910 Mater Res Innov curved dampers. Eng Struct 2017;130:99–111. https://doi.org/10.1016/j.
2015;19(sup5). https://doi.org/10.1179/1432891714Z.0000000001219. engstruct.2016.09.063.
[18] Chen Z, Ge H, Usami T. Hysteretic perfomance of shear panel dampers. Adv Steel [40] Hsu HL, Halim H. Brace performance with steel curved dampers and amplified
Struct 2005;II:1223–8. deformation mechanisms. Eng Struct 2018;175(August):628–44. https://doi.org/
[19] Chan RWK, Albermani F, Williams MS. Evaluation of yielding shear panel device 10.1016/j.engstruct.2018.08.052.
for passive energy dissipation. J Constr Steel Res Feb. 2009;65(2):260–8. https:// [41] Dehghani S, Fathizadeh SF, Yang TY, Farsangi EN, Vosoughi AR. Performance
doi.org/10.1016/j.jcsr.2008.03.017. evaluation of curved damper truss moment frames designed using equivalent
[20] Yang TY, Li T, Tobber L, Pan X. Experimental and numerical study of honeycomb energy design procedure. 2021; 226(October 2020), doi: 10.1016/j.
structural fuses. Eng Struct 2020;204(September):109814. https://doi.org/ engstruct.2020.111363.
10.1016/j.engstruct.2019.109814. [42] Jarrah M, Khezrzadeh H, Mofid M, Jafari K. Experimental and numerical
[21] Chan RWK, Albermani F. Experimental study of steel slit damper for passive energy evaluation of piston metallic damper (PMD). J Constr Steel Res 2019;154:99–109.
dissipation. Eng Struct Apr. 2008;30(4):1058–66. https://doi.org/10.1016/j. https://doi.org/10.1016/j.jcsr.2018.11.024.
engstruct.2007.07.005. [43] Cheraghi A, Zahrai SM. Cyclic Testing of Multilevel Pipe in Pipe Damper. J Earthq
[22] Lee J, Kim J. Development of box-shaped steel slit dampers for seismic retrofit of Eng Nov. 2019;23(10):1695–718. https://doi.org/10.1080/
building structures. Eng Struct 2017;150:934–46. https://doi.org/10.1016/j. 13632469.2017.1387191.
engstruct.2017.07.082. [44] Guo W, Li S, Zhai Z, Li Z, Tan S, Ding F. Seismic performance of a new S-shaped
[23] Garivani S, Aghakouchak AA, Shahbeyk S. Numerical and experimental study of mild steel damper with varied yielding cross-sections. J Build Eng October 2021;
comb-teeth metallic yielding dampers. Int J Steel Struct Mar. 2016;16(1):177–96. 2022(45):103508. https://doi.org/10.1016/j.jobe.2021.103508.
https://doi.org/10.1007/s13296-016-3014-z. [45] Tabatabaei Mirhosseini R, Hamzeh S. Modified bar-fuse damper in gusset plate
[24] Yang TY, Li T, Tobber L, Pan X. Experimental and numerical study of honeycomb detail to improve seismic behavior of bracing system. Structures 2021; 29
structural fuses. Eng Struct 2021;204(November):2020. https://doi.org/10.1016/j. (December 2020): 954–965, doi: 10.1016/j.istruc.2020.12.011.
engstruct.2019.109814. [46] Shao F, Gu T, Jia L-J, Ge H, Taguchi M. Experimental study on damage detectable
[25] Sahoo DR, Singhal T, Taraithia SS, Saini A. Cyclic behavior of shear-and-flexural brace-type shear fuses. Eng Struct 2020;225(March):111260. https://doi.org/
yielding metallic dampers. J Constr Steel Res Nov. 2015;114:247–57. https://doi. 10.1016/j.engstruct.2020.111260.
org/10.1016/j.jcsr.2015.08.006. [47] Bakhshayesh Y, Shayanfar M, Ghamari A. Improving the performance of
[26] Hanson CXD. Influence of ADAS element parameters on building seismic response. concentrically braced frame utilizing an innovative shear damper. J Constr Steel
J Struct Eng 1992;118(7):1903–18. Res 2021;182:106672. https://doi.org/10.1016/j.jcsr.2021.106672.
[27] Tsai K. Design of steel trinagular plate energy absorbers for seismic-resistant [48] Hibbeler K-H-T. Russell Charles, Structural analysis. Upper Saddle River: Pearson
construction. Earthq Spectra 1993; 9(3): 1990. Prentice Hall; 2006.
[28] Ebadi Jamkhaneh M, Ebrahimi AH, Shokri Amiri M. Experimental and Numerical [49] Zhai Z, Guo W, Yu Z, He C, Zeng Z. Experimental and numerical study of S-shaped
Investigation of Steel Moment Resisting Frame with U-Shaped Metallic Yielding steel plate damper for seismic resilient application. Eng Struct 2020;221(May):
Damper. Int J Steel Struct Jun. 2019;19(3):806–18. https://doi.org/10.1007/ 111006. https://doi.org/10.1016/j.engstruct.2020.111006.
s13296-018-0166-z. [50] Dal Lago B, Biondini F, Toniolo G. Experimental tests on multiple-slit devices for
[29] Yüksel E, et al. Behaviour of steel cushions subjected to combined actions. Bull precast concrete panels. Eng Struct Jul. 2018;167(April):420–30. https://doi.org/
Earthq Eng Feb. 2018;16(2):707–29. https://doi.org/10.1007/s10518-017-0217- 10.1016/j.engstruct.2018.04.035.
4. [51] ASCE/SEI-7-10. Minimum Design Loads for Buildings and Other Structures, Boca
[30] Deng K, Pan P, Wang C. Development of crawler steel damper for bridges. JCSR Raton, FL: American Society of Civil Engineers (ASCE); 2016.
2013;85:140–50. https://doi.org/10.1016/j.jcsr.2013.03.009. [52] ANSI/AISC 360-16. Specification for Structural Steel Buildings, Chicago, Illinois
[31] Amiri HA, Najafabadi EP, Estekanchi HE. Experimental and analytical study of 60601: American Institute of Steel Construction (AISC); 2016.
Block Slit Damper. J Constr Steel Res 2018;141:167–78. https://doi.org/10.1016/ [53] Lu X, Fu G, Shi W, Lu W. Shake table model testing and its application. Struct
j.jcsr.2017.11.006. Design Tall Spec Build Mar. 2008;17(1):181–201. https://doi.org/10.1002/
[32] Demir S, Husem M. Saw type seismic energy dissipaters: development and cyclic tal.338.
loading test. J Constr Steel Res 2018;150:264–76. https://doi.org/10.1016/j. [54] ANSI/AISC 341-16. Seismic Provisions for Structural Steel Buildings, Chicago,
jcsr.2018.08.015. Illinois 60601: American Institute of Steel Construction (AISC),” 2016.
[33] Aghlara R, Tahir MM. A passive metallic damper with replaceable steel bar [55] Applied Technology Council, Guidelines for Cyclic Seismic Testing of Components
components for earthquake protection of structures. Eng Struct 2018;159 of Steel Structures for Buildings, Report No. ATC-24; 1992.
(January):185–97. https://doi.org/10.1016/j.engstruct.2017.12.049. [56] ASTM. Standard test methods for tension testing of metallic materials (Metric) 1.
[34] Qu B, Dai C, Qiu J, Hou H, Qiu C. Testing of seismic dampers with replaceable U- ASTM International 2013; E8/E8M-13a, doi: 10.1520/E0008.
shaped steel plates. Eng Struct August 2018;2019(179):625–39. https://doi.org/ [57] Kachooee A, Kafi MA. A Suggested Method for Improving Post Buckling Behavior
10.1016/j.engstruct.2018.11.016. of Concentric Braces Based on Experimental and Numerical Studies. Structures
[35] Gray MG, Christopoulos C, Packer JA. Cast steel yielding brace system for 2018;14(March):333–47. https://doi.org/10.1016/j.istruc.2018.04.003.
concentrically braced frames: Concept development and experimental validations. [58] Abathun MZ, Han J, Yu W. “Effects of manufacturing methods and production
J Struct Eng (United States) 2014;140(4):pp. https://doi.org/10.1061/(ASCE) routes on residual stresses of rectangular and square hollow steel sections: a
ST.1943-541X.0000910. review”, Archives of Civil and Mechanical. Engineering 2021;21(3):pp. https://doi.
[36] Palermo M, Silvestri S, Gasparini G, Trombetti T. Crescent shaped braces for the org/10.1007/s43452-021-00193-8.
seismic design of building structures. Mater Struct/Materiaux et Constructions [59] Kazantzi AK, Vamvatsikos D. The Hysteretic Energy as a Performance Measure in
2015;48(5):1485–502. https://doi.org/10.1617/s11527-014-0249-z. Analytical Studies. Earthq Spectra May 2018;34(2):719–39. https://doi.org/
[37] Palermo M, Pieraccini L, Dib A, Silvestri S, Trombetti T. Experimental tests on 10.1193/112816EQS207M.
Crescent Shaped Braces hysteretic devices. Eng Struct 2017;144:185–200. https:// [60] Aghlara R, Tahir MMd. A passive metallic damper with replaceable steel bar
doi.org/10.1016/j.engstruct.2017.04.034. components for earthquake protection of structures. Eng Struct 2018;159
[38] Mokhtari E, Laghi V, Palermo M, Silvestri S. Quasi-static cyclic tests on a half- (December 2017):185–97. https://doi.org/10.1016/j.engstruct.2017.12.049.
scaled two-storey steel frame equipped with Crescent Shaped Braces. Eng Struct [61] Maleki S, Mahjoubi S. Dual-pipe damper. J Constr Steel Res Jun. 2013;85:81–91.
2021;232(January):111836. https://doi.org/10.1016/j.engstruct.2020.111836. https://doi.org/10.1016/j.jcsr.2013.03.004.

18