This article appeared in a journal published by Elsevier.

The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
 Author's personal copy

Soil Dynamics and Earthquake Engineering 31 (2011) 478–490

Contents lists available at ScienceDirect

Soil Dynamics and Earthquake Engineering

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

Comparative seismic performance of steel frames retrofitted with

buckling-restrained braces through the application of Force-Based
and Displacement-Based approaches
Amador Terán-Gilmore a,n, Jorge Ruiz-Garcı́a b
a
Universidad Autónoma Metropolitana, México
b
Universidad Michoacana de San Nicolás de Hidalgo, México

a r t i c l e i n f o a b s t r a c t

Article history: This paper presents an analytical study aimed at evaluating the feasibility of using buckling-restrained
Received 23 June 2010 braces as a retrofit scheme for existing multi-bay multi-story steel buildings. For that purpose, the seismic
Received in revised form response of four two-dimensional frame models representative of typical steel buildings designed in a
1 November 2010
region of high seismicity was analyzed prior to and after including buckling-restrained braces as a retrofit
Accepted 3 November 2010
strategy. The braces were designed following Force-Based and Displacement-Based approaches. The
structural performance of the different versions of the frames was evaluated by subjecting each one to a
set of twenty ground motions representative of the design earthquake with 10% exceedance probability in
fifty years. It was observed that buckling-restrained braces allow for an efficient reduction in the peak
drift demands in the retrofitted frames. However, since the beneficial effect of the braces cannot be fully
controlled under a Force-Based design approach, it was concluded that a Displacement-Based design
approach is the best option to achieve optimum structural performance.
& 2010 Elsevier Ltd. All rights reserved.

1. Motivation shows schematically the concept of a buckling-restrained brace

Several techniques, such as reinforced concrete and steel jacketing,
as well as the addition of walls and/or braces, have been used to retrofit
buildings that have experienced structural damage as a consequence
of moderate or severe earthquake ground shaking, or for the seismic
upgrading of outdated buildings. Among these techniques, diagonal
steel bracing has been considered an attractive option to enhance the
lateral strength and stiffness of existing multi-story steel buildings.
Nevertheless, it should be kept in mind that traditional braces (ductile
or not) tend to exhibit global buckling when subjected to compressive
strains, which in turn results in local buckling, fracture of the base
material, and a highly unstable behavior under cyclic loading [1,2].
Under these circumstances, adequate earthquake-resistance can only
be achieved through conservative design, or more rationally, through
combining the well-known efficiency of triangular sub-structures
with innovative buckling-restraining mechanisms.
The idea behind a buckling-restrained brace is to fabricate a
structural element that is able to work in a stable manner when
subjected to compressive deformations. Because braces are nor-
mally able to behave in a stable manner when subjected to tensile
forces, a buckling-restrained brace is capable of dissipating large
amounts of energy in the presence of multiple yield reversals. Fig. 1
n
Corresponding author.
E-mail address: tga@correo.azc.uam.mx (A. Terán-Gilmore).
and shows its different components: (A) a ductile steel core that
dissipates energy through axial deformation; (B) mortar, concrete
or grout fill that restricts buckling of the core; and (C) a steel jacket
that confines the mortar, concrete or grout fill and provides further
restriction from buckling. Under severe ground motion, only the
core of the brace should yield.
Usually, the steel core is isolated from the mortar, concrete or
grout fill in an attempt to minimize or eliminate the transfer of axial
stresses between both materials. This is done so that the compres-
sion strength of the brace is similar to its tension strength. Further
discussion regarding the concept and use of buckling-restrained
braces can be found in Black et al. [3] and Uang and Nakashima [4].
Experimental testing on buckling-restrained braces indicates a
highly stable behavior under severe cyclic loading.
Several studies have focused on the use of buckling-restrained
braces to provide efficient seismic resistance to moment-resisting
steel frames, and as a result several Displacement-Based design
methodologies have been formulated [5–8]. Within this context, the
effects of several parameters have been studied to better explain the
dynamic nonlinear response of buildings with dual structural systems
composed of steel frames and buckling-restrained braces, among
them: (A) the lateral stiffness of the frames relative to that of the
braces; (B) the yield strength of the steel used to fabricate the braces;
(C) the number of stories; and (D) the distribution through height of
braces. Although the studies have been useful in clarifying many issues
in terms of the earthquake-resisting potential of buckling-restrained

0267-7261/$ - see front matter & 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.soildyn.2010.11.003
 Author's personal copy
A. Terán-Gilmore, J. Ruiz-Garcı́a / Soil Dynamics and Earthquake Engineering 31 (2011) 478–490
Plan View
Unbonding
agent Steel Core
Steel core
Elevation View
Steel tube
Steel Mortar, concrete
tube or grout fill
A SectionA-A
A Lc
Fig. 1. Schematic configuration of buckling-restrained brace (after Tremblay et al. [14]).
braces, some issues still need clarification. Particularly, all methodol-
ogies referenced above have conformed to a direct Displacement-
Based design format that is based on an equivalent linearization
approach [9]. Inherent to this format is the use of equivalent damping
to represent the integrated plastic hysteretic energy dissipation
capacity of frames and braces. In spite of the relative simplicity of
this approach, the fact is that there is no physical principle that justifies
the existence of a stable relationship between hysteretic energy
dissipation and an equivalent viscous damping [5,10] and, because
of this, empirical calibrations need to be carried out with different
hysteretic behaviors and structural systems [8,11]. Worldwide, the
format of several seismic codes (such as the International Building
Code and the Mexican Building Code) does not allow for the use of
direct Displacement-Based approaches, in such a manner that alter-
native approaches need to be considered and developed. An issue that
still needs to be well understood is the relative advantages or
disadvantages that may result from using steel frames with different
structural properties in terms of the structural efficiency and perfor-
mance of dual systems.
Although the practical implementation of buckling-restrained
braces was originally focused on providing primary lateral resistance
in new buildings, these structural elements have been employed
recently as a retrofit option for existing buildings [12,13]. Within this
context, the primary objective of this paper is to study the effec-
tiveness of a system of buckling-restrained braces in adequately
retrofitting a series of steel moment-resisting frames through the
use of a simple Displacement-Based format that characterizes the
nonlinear behavior of the braces through their maximum ductility
demand. The advantages of using a Displacement-Based approach in
lieu of a Force-Based approach during the sizing of the braces are
discussed and illustrated. The examples provided in the paper allow
for an understanding of what the ideal situation is, in terms of
structural performance and efficiency, for contribution of steel
frames to the structural properties of dual systems.
2. Basis for a Displacement-Based design of low-height
braced frames
The Displacement-Based methodology used herein is based on
the explicit control of the lateral displacement of the building, in
such a manner that there is the need to develop design aids that
explicitly relate the structural properties of the brace (inclination
angle, transverse area, and yield stress) to its global mechanical
characteristics (lateral stiffness and lateral displacement at yield).
The expressions that are offered next have been obtained by
neglecting the global flexural deformation of the bracing system
(produced by the axial deformation of the columns that support it);
that is, they only consider the global shear deformation due to the
axial deformation of the braces. This implies that the relations are
not applicable to slender or tall buildings.
479
L
h L

Fig. 2. Notation used for buckling-restrained brace.
The lateral stiffness that a buckling-restrained brace contributes
to a given story (KL) is related to its steel core area (A) through the
following expression [14]:
KL Ecos2 y
¼ ð1Þ
ðA=LÞ g þ Zð1gÞ
where L is the total length of the brace, E its Young’s modulus, and y
its inclination angle. g is the ratio of the length of the brace
core segment (Lc according to Fig. 2) to the total brace length L and Z
the ratio of the average axial stress in the brace outside the
brace core to the stress in the brace core. Eq. (1) can be used to
estimate the required area of braces in a given story as a function of
the geometry of the bracing system and the lateral stiffness that it
should provide to that story.
Regarding the inter-story drift at yield (IDIy):
 
DL fy ½g þ Zð1gÞ
IDIy ¼ ¼ ð2Þ
h y E sin y cos y
where DL and h are the inter-story displacement and story height,
respectively (Fig. 2); and fy the yield stress of the steel core. The sub-
index y denotes yield.
3. Earthquake ground motions
One set of twenty ground motions, corresponding to the Los
Angeles urban area, was considered in this study. The ground motions
were established for the FEMA/SAC Steel Project [15] and correspond
to the design earthquake involving firm soil with 10% exceedance
probability in fifty years. Fig. 3 shows the elastic mean plus one
standard deviation (s) pseudo-acceleration and displacement spectra
corresponding to the set of motions. Note that the strength spectrum
has a corner period close to 0.4 s and that the displacement spectrum
shows a practically linear increase in displacement with period.
4. Existing steel frames
Four regular three-bay frame models having two different
numbers of stories (4 and 8), as shown in Fig. 4, were considered.
The frames are considered to be representative of exterior
moment-resisting steel frames found in typical office buildings
in California. The steel members of the frames were designed by
Santa-Ana and Miranda [16] according to the lateral load distribu-
tion along height specified in the 1994 Uniform Building Code
(UBC), and their flexural stiffness was tuned to obtain two families
of frames: one that comprises stiff frames and a second one that
includes flexible frames. Particularly, the frames were designed to
exhibit fundamental periods of vibration that could be considered
 Author's personal copy

480 A. Terán-Gilmore, J. Ruiz-Garcı́a / Soil Dynamics and Earthquake Engineering 31 (2011) 478–490

2.5 80
70
2 Displacement
60
Strength
1.5 50

Sd (cm)
Sa /g

40
1 30
20
0.5
10
0 0
0 1 2 3 0 1 2 3
T (sec) T (sec)

Fig. 3. Elastic spectra corresponding to the set of motions under consideration.

Fig. 4. Steel moment-resisting frames under consideration [16].

representative of the dynamic properties derived for this type of
frames from actual earthquake records [17]. While the family of stiff
frames was intended to provide a realistic upper bound in terms of
lateral stiffness, the flexible frames were aimed at providing a
realistic lower bound. It should be mentioned that all frames satisfy
the maximum inter-story drift limitations of the 1994 UBC when
subjected to the design lateral loads corresponding to Zone 4. It can
be said that the two families of frames realistically represent the
wide variety of frames that can result from the application of the
1994 UBC and that their consideration allows for a comparison of
the seismic performance of retrofitted buildings with the same
number of stories but a different balance of lateral stiffness provided
by the original frames and the added buckling-restrained braces.
The frames were modeled as two-dimensional centerline non-
linear models using the computer program RUAUMOKO [18]. A
Rayleigh damping equal to 5% of the critical value was assigned to
the first and second modes in all the frame models. While global
P D effects were considered during the analyses, local P d effects
were neglected. Beams and columns were modeled as frame
elements that concentrate their inelastic response in plastic hinges
located at their ends. An elasto-plastic moment–curvature relation-
ship was assigned at each end. In addition, the axial load–flexural
bending interaction was considered to model the hysteretic beha-
vior of the steel columns. To account for the possibility of fracture in
the beam-to-column connections (observed in many steel buildings
as a consequence of the 1994 Northridge earthquake), a flexural
strength-degradation model as implemented in RUAUMOKO was
contemplated for the hysteretic behavior of the steel beams (i.e.,
their moment capacity is progressively reduced as a function of the
number of inelastic cycles and the curvature ductility ratio reached
at the previous cycle). The parameters that characterize the
degradation model used the values calibrated experimentally by
Filiatrault et al. [19]. The flexural moment capacity of beams and
columns was determined using actual yield strengths of 337.8 and
399.9 MPa, respectively.
4.1. Mechanical characteristics
The nonlinear models were used to evaluate the global mechan-
ical characteristics of the steel frames. Particularly, a triangular
load pattern along height was used to establish their capacity
curves. While Vb and droof denote base shear and roof displacement
in Fig. 5, respectively, W denotes the total reactive weight of the
frames. Table 1 summarizes the elastic fundamental periods of
vibration (Ti) and the corresponding seismic coefficients (Vb/W)
estimated from the nonlinear models. In the table, Te denotes the
effective fundamental period of vibration [20], Sa/g the spectral
pseudo-acceleration ordinate corresponding to Te, and C0, C1, C2 and
C3 modification factors that, respectively, relate or represent [20]:
the spectral displacement of an equivalent single-degree-of-free-
dom system to the roof displacement of the frame; expected
maximum inelastic displacement to the displacement calculated
for linear elastic response; effect of degrading hysteretic behavior
 Author's personal copy

A. Terán-Gilmore, J. Ruiz-Garcı́a / Soil Dynamics and Earthquake Engineering 31 (2011) 478–490 481

1 1
0.8 0.8
4-story stiff 4-story flexible
0.6 0.6
Vb /W

Vb /W
0.4 0.4
0.2 0.2
0 0
0 10 20 30 40 0 10 20 30 40
roof (cm) roof (cm)

1 1
0.8 0.8
8-story stiff 8-story flexible
0.6 0.6
Vb /W

Vb /W
0.4 0.4
0.2 0.2
0 0
0 20 40 60 0 20 40 60
roof (cm) roof (cm)

Fig. 5. Capacity curves of steel frames.

Table 1 the four- and eight-story flexible frames exhibit much larger inter-
Parameters and coefficients involved in the estimation of the roof displacement. story drifts and also develop soft stories. The inter-story drift
demands in the frames plus the fact that the frames form soft
Frame Ti (s) Vb/W Te (s) Sa/g C0 C1 C2 C3
stories indicate the need to rehabilitate them. Within this context,
Four-story stiff 0.71 0.84 0.71 1.29 1.2 1.0 1.0 1.0 it should be considered that static nonlinear analyses of the frames
Four-story flexible 1.24 0.31 1.24 0.90 1.1 1.0 1.0 1.0 indicate that a 1% inter-story drift threshold is associated with their
Eight-story stiff 1.18 0.54 1.18 0.89 1.2 1.0 1.0 1.0 operational performance level (see Fig. 6).
Eight-story flexible 1.95 0.23 1.95 0.47 1.1 1.0 1.0 1.0

5. Retrofit of frames

on maximum displacement response; and the increased displace-
ments due to dynamic second order effects.
Fig. 6 shows in all frames the evolution of the critical plastic
hinge (yp) with respect to the inter-story drift index (IDI) of the
story that accumulates the largest drift demands. Note that all
frames remain elastic with inter-story drifts that do not exceed a
threshold value of 0.01.
4.2. Seismic performance
The lateral roof displacement demand in the frames was
estimated according to the Coefficient Method established in FEMA
356 (Federal Emergency Management Agency [20]). While Table 1
summarizes the values assigned to the different parameters and
coefficients involved in the estimation of the roof displacements,
Fig. 7 summarizes the maximum inter-story drift demands expected
in the frames according to the Coefficient Method (black lines) and
nonlinear step-by-step dynamic analyses (gray lines). While the
Coefficient Method used the strength spectrum, as shown in Fig. 3a,
the results shown for the step-by-step analyses correspond to the
mean plus one standard deviation demands obtained for the ground
motions under consideration. In order to provide a context of the
frames’ performance, FEMA 356 [20] recommendations specify
inter-story drift index thresholds of 0.7%, 2.5% and 5% for the
immediate occupancy, life safety and collapse prevention perfor-
mance levels, respectively.
While the four- and eight-story stiff frames exhibit maximum
inter-story drift index demands close to 2% and develop soft stories,
The frames were retrofitted, either by a Force-Based or Dis-
placement-Based Approach, under the consideration that their
central bay was braced with two braces in a chevron configuration
(as schematically shown in Fig. 8). As for the nonlinear dynamic
analyses of the braced frames, the buckling-restrained braces were
modeled as axial spring elements. The axial stiffness of each brace
was computed as follows:
EA
KAxial ¼ ð3Þ
½g þ Zð1gÞL
where all variables have the same meaning as in Eq. (1). Under the
consideration that Lc is equal to half the value of L (see Fig. 2), and
that the average area in the brace outside the brace core is equal to
three times the area of the brace core, the length factor g þ Zð1gÞ
was assigned a value of 0.667.
In addition, elasto-plastic axial force–displacement behavior was
assumed to model the cyclic behavior of each spring. Since
experimental research has shown that buckling-restrained braces
exhibit larger compressive strength than tensile strength, it was
assumed that the yield compressive strength of each spring is 2%
larger than its corresponding yield tensile strength. In this study,
the yield stress associated with the braces was assumed to be 10%
larger than the corresponding nominal stress of 248.1 MPa.
5.1. Force-Based approach
Under the Force-Based approach, the braces were designed to
fully accommodate the design lateral loads [21], in such a way that
the design process was centered on the definition of core areas in
 Author's personal copy

482 A. Terán-Gilmore, J. Ruiz-Garcı́a / Soil Dynamics and Earthquake Engineering 31 (2011) 478–490

0.01 0.01
0.008 0.008
p 0.006 0.006

p
4-story stiff 4-story flexible
0.004 0.004
0.002 0.002
0 0
0.000 0.005 0.010 0.015 0.020 0.000 0.005 0.010 0.015 0.020
IDI IDI

0.01 0.01
0.008 0.008
0.006 0.006
p

p
8-story stiff 8-story flexible
0.004 0.004
0.002 0.002
0 0
0.000 0.005 0.010 0.015 0.020 0.000 0.005 0.010 0.015 0.020
IDI IDI

Fig. 6. Evolution of plastic rotation with increasing inter-story drift in steel frames.

5 5
4-story stiff 4-story flexible
4 4
CoefficientMethod
3 StepbyStep 3 FEMA 356
FEMA 356
FEMA 356 LifeSafety Collapse
2 2
Immediate Prevention
Operation
1 1

0 0
0 0.01 0.02 0.03 0.04 0.05 0.06 0 0.01 0.02 0.03 0.04 0.05 0.06

9 9
8 8-story stiff 8 8-story flexible
7 7
6 6
5 5
4 4
3 3
2 2
1 1
0 0
0 0.01 0.02 0.03 0.04 0.05 0.06 0 0.01 0.02 0.03 0.04 0.05 0.06

Fig. 7. Expected performance of steel frames.

the braces that could meet the equivalent lateral load distribution total number of stories, and k a parameter that accounts for
along height. The following two-step procedure was employed: higher modes:

8
<1
> T1 r0:5s
(1) Compute equivalent lateral loads along height in accordance
k ¼ ðT1 þ 0:5Þ=2 0:5 rT1 o 2:5s ð5Þ
with the requirements of FEMA 356 [20]: >
:
2 T1 Z2:5 s
wi hki
Fi ¼ Vb PN ð4Þ where T1 is the fundamental period of vibration of the braced
k
j ¼ 1 wj hj
frame, which can be estimated using empirical or approximate
where Vb is the base shear, wi and hi the weight and the height methods such as those described in FEMA 356 [20]. To determine
measured from the baseline, respectively, of the ith story, N the the static lateral forces, the mean plus one standard deviation
 Author's personal copy

A. Terán-Gilmore, J. Ruiz-Garcı́a / Soil Dynamics and Earthquake Engineering 31 (2011) 478–490 483

Steel Frame: Operational

Buckling Restrained Braces: Life Safety Pushover analysis of
Steel Frame

O O
Steel Frame: IDImax ≤ IDISF Establish IDISF

IDIy = f (fy)
max = f (IDIy , IDImax)

Fig. 8. Configuration of bracing system of the four-story frames.

max = f (IDI max )

Table 2
Dynamic properties and core area of buckling-restrained braces (Force-Based
approach).
Sd
Frame TB (s) Area (cm2) max  =5 %
  = max
1 2 3 4 5 6 7 8

Four-story stiff 0.48 11 7 6 3 – – – –

Four-story flexible 0.67 16 10 8 4 – – – – T
Tmax
Eight-story stiff 0.80 12 9 9 8 7 5 4 2
Eight-story flexible 1.16 23 18 16 15 12 10 7 4

strength spectrum shown in Fig. 3a was used. A force modifica-
tion factor, R, equal to 8 was employed in the design of the
buckling-restrained braces.
(2) Once the equivalent lateral forces were computed, their height-
wise distribution of shear forces (Vi) was obtained. Then, from
static equilibrium, the axial force acting in each brace at the ith
story was computed as follows:
Vi
Pi ¼
2 cos y
where y is the inclination angle of the braces. The required area
for the braces located at the ith story, Ai, can be established as
follows:
Pi
Ai ¼
jfy
where fy is the yield strength of the steel used to fabricate the
core of the braces (248.1 MPa) and j a strength-reduction
factor equal to 0.9.
Table 2 summarizes the actual dynamic properties of the braced
frames and the sizes of the braces established through the Force-
Based procedure. Note that the area of the braces varies from one
story to the next and that, though the stiff and flexible frames
should exhibit a similar performance once retrofitted, the funda-
mental periods of vibration of the flexible braced frames are larger
than those of their corresponding stiff braced frames.
5.2. Displacement-Based approach
The Displacement-Based methodology introduced herein, applic-
able to the retrofit of standard occupation buildings and schemati-
cally shown in Fig. 9, considers the Life Safety performance level. Its
first step involves establishing a qualitative definition of adequate
performance. This is done through the explicit consideration of the
acceptable levels of damage of the different structural elements
within the braced frame (existing steel frame and braces). The second
step consists of the quantification of adequate performance through
Stiffness-based sizing of
braces
Fig. 9. Preliminary seismic design methodology.
establishing response thresholds with the aid of damage indices.
During the third step, the methodology establishes, through the use
ð6Þ
of an appropriate displacement spectrum, the value of the funda-
mental period of vibration of the building, which quantifies the
design lateral stiffness. The sizing of the braces is established
according to the value of this parameter [22].
Regarding the qualitative performance definitions, Life Safety is
ð7Þ considered to be satisfied if the braced frame guarantees the
physical integrity of the occupants of the building and provides
for easy structural rehabilitation. Under this context, while the
existing steel frame should satisfy the operational performance
level, the bracing system should develop significant plastic beha-
vior. Life Safety of the braced frame is assumed to be satisfied if its
maximum inter-story drift index demand (IDImax) is limited in such
a way as to result in an operational steel frame. First, a nonlinear
static analysis of the steel frame should be carried out in order to
estimate the inter-story drift index threshold associated with its
O
operational performance level (IDISF ). To illustrate the concepts in
this paper, it will be assumed that a fully operational frame is
possible if all its structural elements are kept elastic.
The threshold value of the fundamental period of vibration of the
braced frame (Tmax) is established according to Fig. 9. To make this
possible, IDImax can be used to establish a lateral roof displace-
ment threshold for the frame
IDImax H
dmax ¼ ð8Þ
COD
where H is the total height of the building and COD a coefficient of
distortion that considers that inter-story drift is not constant through-
out the height of the building [23]. Table 3 summarizes values of COD
for the preliminary design of structures that exhibit shear-like global
behavior. Because the deflected shapes of low-rise buckling-
 Author's personal copy

484 A. Terán-Gilmore, J. Ruiz-Garcı́a / Soil Dynamics and Earthquake Engineering 31 (2011) 478–490

Table 3 Table 5
Range of values of COD. Suggested values of b for regular buildings exhibiting large post-yield stiffness.

Global COD Number of Global shear ductility

ductility stories
Regular Irregular mmax ¼2 mmax ¼ 6

1 1.2 1.5 5 1.0 1.0

2+ 1.5 2.0 10 1.0 1.1
20 1.1 1.2
30 1.2 1.3
40+ 1.3 1.4
Table 4
Values of a for regular buildings.
Table 6
Stories a Value of parameters and coefficients involved in the Displacement-Based approach.

m¼1 m ¼ 2+ Frame COD a b IDImax d (cm) d/a (cm) Tmax (s)

1 1.0 1.0 Four-story stiff 1.2 1.2 1.0 0.01 13.7 11.4 0.45
2 1.2 1.1 Four-story flexible 1.2 1.2 1.0 0.01 13.7 11.4 0.45
3 1.3 1.2 Eight-story stiff 1.2 1.2 1.0 0.01 25.9 21.6 0.70
5+ 1.4 1.2 Eight-story flexible 1.2 1.2 1.0 0.01 25.9 21.6 0.70

restrained bracing systems in whom the areas of braces are varied in
every story tend to exhibit a linear deformed shape even in the case of
plastic behavior [6,24], the values of COD corresponding to a global
ductility of one in Table 3 can be used in this case.
The fundamental period of vibration can be estimated through
the use of the displacement threshold dmax and a design displace-
ment spectrum. For this purpose, dmax should be modified to take
into consideration multi-degree-of-freedom effects. According to
what is shown in Fig. 9, the roof displacement threshold should be
corrected through the use of parameter a. Based on the recom-
mendations of FEMA 306 [25] and Teran-Gilmore [26], Table 4
presents values of a for the preliminary design of regular structures
that exhibit shear-like global behavior.
The design methodology also requires an estimate of the global
maximum ductility demand associated with the bracing system
(mmax) to define the design displacement spectrum. The inter-story
ductility of the braces (mint) is established by normalizing IDImax by
IDIy. Then a global value of mmax is assigned to the entire bracing
system according to
mint
mmax ¼
b 1:1  248:1  0:667
where b is a correction factor that depends on the level of structural
irregularity through height and increases with the increasing
number of stories and the maximum ductility to be developed
by the building [27,28]. Table 5 presents values of b for the
preliminary design of regular buildings having a post-yield stiff-
ness that is equal or larger than ten percent of the elastic stiffness.
According to the acceptable level of damage in the braces for Life
Safety, the displacement spectrum contemplates a percentage of
critical damping (x) of 5%.
Once the value of Tmax is available, the braces are sized
accordingly, i.e., the transverse areas of the braces are deemed
adequate if the actual fundamental period of the building (TA) is
equal to or slightly less than Tmax. Once the braces have been sized
for stiffness, the beams and columns of the existing steel frame that
will provide support to the braces are revised using capacity design
concepts and adjusted in case it is necessary.
Unlike direct Displacement-Based design methodologies that
are based on an equivalent linearization approach (e.g., [5,8]),
the Displacement-Based methodology under consideration in this
paper uses the maximum inter-story ductility to characterize
the nonlinear behavior and plastic hysteretic energy dissipating capa-
city of the buckling-restrained braces. In this sense, the methodology
can be easily adapted to code formats that use behavior or force
reduction factors R to establish their design spectra. In the particular
case under consideration, the frames should remain operational in such
a manner that a single R factor can be used to characterize the seismic
behavior of the dual system. Under these circumstances, the R factor
should reflect the plastic hysteretic energy dissipation capacity of the
braces and the maximum deformation capacity of the frames at the
operational performance level.
With respect to the application of the Displacement-Based
methodology summarized in Fig. 9 to the existing frames, the
nonlinear static analyses indicate that to achieve an operational
performance level with the existing frames, the inter-story drift
should be limited to
O
IDISF ¼ 0:01 ð10Þ
Considering that the geometry of the frame is such that
g þ Zð1gÞ  0:667 and a E451 and that the steel with which the
cores of the buckling-restrained braces are fabricated is a type such
that it exhibits a yield stress (fy) of 248.1 MPa (and an overstrength
of 10%), Eq. (2) yields
ð9Þ
IDIy ¼  0:002 ð11Þ
196133sin 451 cos 451
Note in the above equation that the value of fy should be the actual
yield stress. Considering that IDImax should be controlled within
O
the threshold established by IDISF , the bracing system should be
able to develop a maximum inter-story ductility close to
IDImax 0:010
mint ¼ ¼ ¼ 5:0 ð12Þ
IDIy 0:002
Table 6 summarizes the values of COD, a and b used to estimate
the value of Tmax in each frame. Note that the value of COD
corresponds to bracing systems that, although developing signifi-
cant plastic behavior, vary the sizes of their braces in every story.
According to Table 6, b is equal to one in all frames, in such a
manner that according to Eq. (9), mmax ¼ mint ¼5.0. Fig. 10 shows the
estimation of Tmax of the four- and eight-story frames. Note that the
design displacement spectrum (mean plus one standard deviation)
corresponds to 5% of critical damping and mmax of 5. The values of
Tmax of the four- and eight-story frames are summarized in Table 6.
Once the value of Tmax is established for a particular frame,
its braces are sized according to it. Qualitatively, it can be said that
the braces are sized in such a manner as to control the maximum
inter-story demand in the frame within the 0.01 threshold.
 Author's personal copy

A. Terán-Gilmore, J. Ruiz-Garcı́a / Soil Dynamics and Earthquake Engineering 31 (2011) 478–490 485

80 80

70 70

60 μmax = 5 60 μmax = 5
50 50

S d (cm)
Sd (cm)

40 40

30 30 δ roof /α = 21 cm

20 δ roof /α = 11 cm 20

10 10
Tmax = 0.45 sec Tmax = 0.7 sec
0 0
0 0.5 1 1.5 2 2.5 3 0 0.5 1 1.5 2 2.5 3
T (sec) T (sec)

Fig. 10. Estimation of fundamental period of vibration for design of bracing systems.

Under the assumption that the lateral response of the building is Table 7
dominated by global shear effects, it is possible to establish that the Dynamic properties and core area of buckling-restrained braces (Displacement-
Based approach).
braces and the steel frame will work as two parallel systems, in
such a manner that [22]: Frame TAR (s) TA (s) Area (cm2)
1 1 1 1 1 1
þ 2 ¼ 2 ) 2 ¼ 2  2 ð13Þ 1 2 3 4 5 6 7 8
2
TBR TSF Tmax TBR Tmax TSF
Four-story stiff 0.44 0.45 21 12 9 5 – – – –
where TSF is the fundamental period of vibration of the existing Four-story flexible 0.45 0.48 29 19 14 8 – – – –
frame, and TBR is the period that establishes the stiffness require- Eight-story stiff 0.70 0.74 36 22 20 18 16 13 9 5
ments for the braces. Eight-story flexible 0.70 0.77 43 30 28 25 22 18 13 7
The stiffness-based sizing of the braces should result in the
actual fundamental period of vibration of the braced frame (TA)
being close to Tmax. In this context, an alternative for the sizing of
braces starts by establishing with Eq. (4) a preliminary distribution columns that provide support to the braces (i.e., the internal columns,
along height of lateral forces in the braced frame. as shown in Fig. 8) was restricted. Note, by comparing the values
Initial areas are assigned to braces (Aini
BR ). Initially, the only included in Tables 6 and 7, that the proposed methodology yields
condition that should be satisfied by these areas is that they exhibit practically equal values of TAR and Tmax. Nevertheless, the columns of
a variation along height that is proportional to the lateral shear the steel frames undergo axial deformations due to the axial stresses
distribution derived from the force distribution established accord- induced within them by the braces, in such a manner that the frames
ing to Eq. (4). Once the braces have been preliminary sized, the exhibit an overall flexural deformation that is not explicitly con-
period associated with the bracing system can be estimated. For sidered by the methodology. Because of this, the values of TA of the
this purpose, it is reasonable to estimate the lateral drift induced in braced frames are slightly larger than their corresponding values of
the braces with Eq. (1). Once the global drifts in the braces have Tmax. It was considered herein that the difference between the values
been computed, an initial estimate of the fundamental period of of TA and Tmax did not merit a resizing of the braces. Attention should
vibration can be established: be drawn to the eight-story flexible frame. Because of the axial
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Pn flexibility of the frame’s internal columns, a value of 0.86 s was
2
ini i ¼ 1 Wi di obtained for TA when applying directly to it the Displacement-Based
T ¼ 2p P n ð14Þ
g i ¼ 1 Fi di methodology. This value was considered unacceptably high in terms
of the seismic performance of this frame. To reduce the overall flexural
where di is the lateral displacement in the ith story of the braced
deformations in the eight-story flexible frame, the areas of the
frame; Wi and Fi the reactive weight and lateral force corresponding
columns that provide direct support to the buckling-restrained braces
to that story, respectively; and g the acceleration due to gravity.
were increased. Particularly, these areas were increased by 150%
Note that di are estimated from the lateral force distribution
(their final areas are equal to 2.5 times their original areas) and the
established according to Eq. (4), and that for such a purpose an
end result of this was to reduce TA to the value of 0.77 considered for
arbitrary value can be assigned to the base shear. Once T ini is
this frame in Table 7. This clearly exhibits the limitations of the
estimated, the definitive areas for braces (ABR) can be established:
methodology used herein. Teran-Gilmore and Coeto [29] discuss the
 ini 2 adaptations that should be made to the methodology to incorporate
T
ABR ¼ Aini
BR ð15Þ the global flexural deformations into the design of a braced frame.
TBR

Once the braces are sized for stiffness, the columns and beams of
the existing frames need to be revised using capacity design
concepts in such a manner that plastic behavior is concentrated
in the braces. It should be mentioned that in terms of strength, the
existing columns of all the steel frames under consideration
provided adequate support to their respective braces.
Table 7 summarizes the actual dynamic properties of the braced
frames and the sizes of their braces. TAR is the fundamental period of
vibration that the frames would have if the axial deformation of the
6. Performance of retrofitted frames
6.1. Frames braced through a Force-Based approach
In order to evaluate the seismic performance of the retrofitted
frames, a series of nonlinear time-history analyses were carried out
using the selected set of earthquake ground motions. Fig. 11 shows
the height-wise distribution of the mean plus one standard
 Author's personal copy

486 A. Terán-Gilmore, J. Ruiz-Garcı́a / Soil Dynamics and Earthquake Engineering 31 (2011) 478–490

5 5
4-story stiff 4-story flexible
4 4

3 3
Story

Story
2 2

1 1

0 0
0 0.01 0.02 0.03 0.04 0.05 0.06 0 0.01 0.02 0.03 0.04 0.05 0.06
IDI IDI

8 8
7 8-story stiff 7 8-story flexible
6 6
5 5
4 4
Story

3 Story 3
2 2
1 1
0 0
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.00 0.01 0.02 0.03 0.04 0.05 0.06
IDI IDI

Fig. 11. Expected performance of retrofitted frames using a Force-Based approach.

deviation maximum inter-story drift index demands in the original
(gray lines) and braced frames (black lines).
As expected, the braces significantly reduce the drift demands in
such a manner that the maximum inter-story drift index demands
in the four- and eight-story braced frames are controlled well
within their corresponding life safety threshold. Particularly, there
is a reduction of about 98% and 138% in terms of the peak inter-
story drift demand in the four-story stiff and flexible frames,
respectively. In addition, it can be observed that the retrofitted
frames exhibit, with the exception of the four-story flexible frame, a
more uniform height-wise distribution of inter-story drift index
demand. This is particularly important in the eight-story frames,
which originally developed soft stories in the ground and fifth
stories. In spite of the significant reduction in inter-story drift
demands, it should be noted that the Force-Based design of the
buckling-restrained braces resulted in inconsistent performance in
the braced frames. This is clearly illustrated by the significantly
different peak inter-story drift index demands in the different
frames (ranging from 0.01 to 0.02), which imply performance levels
that, in general terms, go from operational in the braced stiff frames
to life safety in the braced flexible frames.
6.2. Frames braced through a Displacement-Based approach
Nonlinear step-by-step dynamic analyses were also carried out
in the braced frames designed according to the Displacement-
Based approach. As shown in Fig. 12 and unlike the frames
retrofitted using a Force-Based approach, the drift demands in
all stories of the frames are controlled efficiently by the bracing
systems within the design threshold.
The variability measured in terms of the standard deviation of
the natural log and involved in the estimation of the maximum IDI
for the original (black dashed-line) as well as for the braced frames
(continuous black line for the Force-Based approach and contin-
uous gray line for the Displacement-Based approach) is shown in
Fig. 13. Note that the distribution along height of the variability
becomes more uniform when the frames are braced and that there
are no significant differences in the values corresponding to the
buildings braced by the Force-Based and Displacement-Based
approaches.
6.3. Evaluation of permanent inter-story drift demands
Previous studies have highlighted the fact that steel buildings
can experience significant lateral residual (permanent) deforma-
tions after an earthquake ground motion [30,31]. In this respect,
it is of interest to examine the level of permanent inter-story drifts
in the frames under consideration. Fig. 14 shows the height-wise
distribution of permanent inter-story drift index in each unbraced
steel frame (gray lines). As a reference, FEMA 356 [20] specifies a
permanent inter-story drift threshold of 0.7% for life safety. Note
that the residual inter-story drift demands in the frames are close to
or in excess of this threshold. Not only does this represent a risk to
the occupants of the buildings during aftershocks, but would cause
serious difficulties in case an attempt is made to retrofit the frames.
Residual inter-story drift demands over height in the braced
frames are also shown in Fig. 14. While solid black lines are used in
the case where the braces were sized according to a Force-Baced
procedure, dashed black lines correspond to the Displacement-
Based approach. In general, it can be observed that the addition of
braces reduces significantly the residual drifts in the frames.
Nevertheless, in the case of the four-story frames braced according
to a Force-Based approach, the residual inter-story drift index
demands exhibit large variations along height and exceed the
threshold specified by FEMA 356 for life safety. Although the
residual drift demands corresponding to the eight-story frames
retrofitted with a Force-Based approach are significantly smaller,
the peak residual inter-story drift demand of 0.0045 in the eight-
story flexible braced frame indicates an inability of the Force-Based
approach to adequately control residual drifts on the braced frames.
 Author's personal copy

A. Terán-Gilmore, J. Ruiz-Garcı́a / Soil Dynamics and Earthquake Engineering 31 (2011) 478–490 487

5 5
4-story stiff
4-story flexible
4 4

3 Design threshold 3

2 2

1 1

0 0
0 0.01 0.02 0.03 0.04 0.05 0.06 0 0.01 0.02 0.03 0.04 0.05 0.06

9 9
8 8-story stiff 8 8-story flexible
7 7
6 6
5 5
4 4
3 3
2 2
1 1
0 0
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.00 0.01 0.02 0.03 0.04 0.05 0.06

Fig. 12. Expected performance of retrofitted frames using a Displacement-Based approach.

5 5
4-story stiff 4-story flexible
4 4

3 3

2 2

1 1

0 0
0.0 0.2 0.4 0.6 0.8 0.0 0.2 0.4 0.6 0.8
LnIDI LnIDI

8 8
7 8-story stiff 7 8-story flexible
6 6
5 5
4 4
3 3
2 2
1 1
0 0
0.0 0.2 0.4 0.6 0.8 0.0 0.2 0.4 0.6 0.8
LnIDI LnIDI

Fig. 13. Record-to-record variability involved in the estimation of the maximum IDI.

The Displacement-Based approach has given way to braced their residual deformations. A clear tendency can be observed in
frames that exhibit a more uniform distribution along height of the peak residual inter-story drift demand of the frames retrofitted
permanent drifts and that have the ability to adequately control according to the Displacement-Based approach; while a value of
 Author's personal copy

488 A. Terán-Gilmore, J. Ruiz-Garcı́a / Soil Dynamics and Earthquake Engineering 31 (2011) 478–490

5 5
4-story stiff
4 4

3 3

Story
Story

2 2

1 1

0 0
0 0.005 0.010 0.015 0.020 0.025 0.030 0 0.005 0.010 0.015 0.020 0.025 0.030
IDI IDI

8 8
7 7
6 6
5 5

Story
Story

4 4
3 3
2 2
1 1
0 0
0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.000 0.005 0.010 0.015 0.020 0.025 0.030
IDI IDI

Fig. 14. Permanent drift demands on original and braced frames.

0.001 or less is observed in the stiff braced frames, demands around
0.002 are observed in the flexible braced frames. In any case, all
frames braced according to the Displacement-Based approach are
in good shape to undertake any possible aftershock. Note that in
spite of developing inter-story ductility demands close to five, the
braced frames are able to adequately control their permanent
deformation. This can be explained by the fact that the steel frames
remain fully operational (elastic), thus providing an elastic com-
ponent that mitigates residual displacements [32]. Although the
larger elastic component provided by the stiff frames results in
smaller permanent drifts, the flexible frames provide adequate
control of permanent drifts. Finally it should be mentioned that any
permanent drift in the frames braced according to the Displace-
ment-Based approach is related to plastic behavior that concen-
trates exclusively in the buckling-restrained braces. Under these
circumstances, any permanent drift should be fully recovered by
the existing steel frames once the braces are replaced.
7. Discussion
It is of interest to compare the weight and seismic performance
of similar buildings in which the buckling-restrained braces provide
a different percentage of their lateral stiffness. Take the case of the
bracing systems of the four-story stiff and flexible frames designed
according to the Displacement-Based approach. In the former case,
the design approach requires the braces to provide about 60% of the
total lateral stiffness of the braced frame; in the latter case, this
percentage is close to 85%. In terms of the weight of the frames, the
beams and columns of the stiff and flexible frames weigh 25.8 and
11.2 ton, respectively. The estimated weight of their braces and
connections is 1.3 and 1.9 ton, respectively, in such a manner that
the total weight of the structural elements of the stiff and flexible
braced frames is 27.1 and 13.1 ton, respectively. Even though the
stiff and flexible braced frames exhibit a 2 to 1 ratio in terms of
weight, both result in adequate performance. Similar results can be
obtained with the eight-story frames. This clearly outlines the
efficiency of the use of braces within a Displacement-Based format
to control the lateral response of existing structures, and indicates
that their efficiency increases significantly as the flexibility of the
existing structure increases.
It is interesting to note that due to their large deformation
capacity, it has been suggested that dual systems composed of steel
frames and buckling-restrained braces can and should undergo
significant nonlinear behavior during severe ground motion. Under
these circumstances, the residual deformations in the dual system
need to be controlled either by hinging the ends of the structural
members of the frame [5,7], or by using of strong frames to provide
adequate self-centering capacity. For example, Maley et al. [8]
considered dual systems where the steel frames provide 40–50% of
the lateral strength and estimated residual inter-story drifts of
0.004 and 0.003 when these systems undergo maximum inter-story
drifts close to 0.02 during ground motion. The results summarized
in this paper allow for the formulation of a design approach that
differs from that offered in previous publications. To understand
this, the weight and inter-story drift demands obtained in the
different four-story buildings should be considered. In the first
place, the original (unbraced) four-story stiff frame weighs 25.8 ton
and exhibits maximum and residual inter-story drifts of 0.02 and
0.006, respectively. Second, the braced version of this building
weighs 27.1 ton and has drift demands of 0.01 and 0.001. A first
comparison between these two cases allows for the following
conclusion: while the buckling-restrained braces adequately and
efficiently control the maximum lateral response of the dual system,
the stiff frames (which contribute close to 40% of the lateral stiffness
of the system) allow for efficient control of the residual displace-
ments. Nevertheless, note that no benefits are obtained in terms of
structural efficiency by using stiff frames, because no significant
reduction in weight can be achieved. Consider now the case of the
braced four-story flexible frame that weighs 13.1 ton and exhibits
inter-story drifts of 0.01 and 0.002. Adequate performance is
achieved in terms of maximum and residual displacements with
 Author's personal copy
A. Terán-Gilmore, J. Ruiz-Garcı́a / Soil Dynamics and Earthquake Engineering 31 (2011) 478–490
a frame that contributes about 15% of the lateral stiffness of the dual
system. The fact to be outlined within this context is that the braced
four-story flexible frame is able to achieve this with about half the
structural steel required by the two cases in which the stiff frame
was considered. It can be concluded that in terms of performance
and structural efficiency, a dual system with a flexible gravitational
system can outperform similar systems in which the steel frames
are assigned a larger role in terms of seismic resistance. A key
element within this context is the need to carefully control the
maximum inter-story drift in such a way as to achieve a serviceable
gravitational system.
8. Conclusions
An analytical study aimed at evaluating the feasibility of using
buckling-restrained braces as a retrofit scheme in existing multi-
bay multi-story steel buildings has been presented. The braces
were designed according to Force-Based and Displacement-Based
approaches. From the comparison of the seismic performance of
different versions of the frames, the following conclusions can be
drawn:
 Buckling-restrained braces sized according to Force and Dis-
placement-Based approaches represent an attractive option for
the retrofitting of steel buildings. Nevertheless, a Displacement-
Based approach provides advantages over a traditional Force-
Based approach because the former approach allows for explicit
damage control in the existing frame through the explicit
control of the inter-story drift demands.
 The use of buckling-restrained braces to retrofit existing steel
frames allows for a significant mitigation of residual deforma-
tions in the frames. In particular using a Displacement-Based
approach not only represents an advantage in terms of better
control of the maximum residual inter-story drift demand, but
results in fairly uniform distribution along height of permanent
drift. Also, any permanent drift in a frame braced through the
Displacement-Based approach used herein should disappear
once the damaged braces are replaced.
Because of their large maximum and cumulative plastic defor-
mation capacities, it has been suggested that dual systems com-
posed of steel frames and buckling-restrained braces can and
should undergo significant nonlinear behavior during severe
ground motion. Within this context, the residual deformations in
the dual system need to be controlled either by hinging the ends of
the structural members of the frame or by using strong frames to
provide adequate self-centering capacity. The results obtained in
this paper strongly suggest that although this approach leads to
adequate seismic performance, the end result of its application is
an inefficient use of structural steel. A dual system composed of a
flexible gravitational system and buckling-restrained braces can
outperform, particularly in terms of structural efficiency, similar
systems in which the steel frames are assigned a larger role in terms
of seismic resistance. A key element within this context is the need
to carefully control the maximum inter-story drift in such a way as
to achieve a serviceable gravitational system.
Acknowledgements
While the first author would like to express his gratitude to
Universidad Autónoma Metropolitana, the second author wishes to
thank Universidad Michoacana de San Nicolás de Hidalgo in México
for the support provided in developing the research reported in
this paper. The authors would also like to thank two anonymous
489
reviewers for their comments and suggestions, which helped
improve the final version of the paper.
References
[1] Bertero VV, Anderson JC, Krawinkler H. Performance of steel building struc-
tures during the Northridge earthquake. Report EERC 94/09, University of
California at Berkeley, 1994.
[2] Uriz P, Mahin SA. Towards earthquake-resistant design of concentrically
braced steel-frame structures. Report PEER 2008/08, University of California
at Berkeley, 2008.
[3] Black C, Makris N, Aiken I. Component testing, stability analysis and char-
acterization of buckling-restrained unbounded braces. Report PEER 2002/08,
University of California at Berkeley, 2002.
[4] Uang C-M, Nakashima M. Steel buckling restrained braced frames. Earthquake
engineering: from engineering seismology to performance-based design.
CRC Press; 2003.
[5] Kim J, Seo Y. Seismic design of low-rise steel frames with buckling-restrained
braces. Engineering Structures 2004;26(5):543–51.
[6] Kim J, Choi H. Behavior and design of structures with buckling-restrained
braces. Engineering Structures 2004;26(6):693–706.
[7] Choi H, Kim J. Evaluation of seismic energy demand and its application on
design of buckling-restrained braced frames. Structural Engineering and
Mechanics 2009;31(1):93–112.
[8] Maley TJ, Sullivan TJ, Della Corte G. Development of a displacement-based
design method for steel dual systems with buckling-restrained braces and
moment-resisting frames. Journal of Earthquake Engineering 2010;14(S1):
106–40.
[9] Miranda E, Ruiz-Garcia J. Evaluation of approximate methods to estimate
maximum inelastic displacement demands. Earthquake Engineering and
Structural Dynamics 2002;31:539–60.
[10] Krawinkler H. New trends in seismic design methodology. In: Proceedings
of the tenth European conference on earthquake engineering, vol. 2, 1994.
p. 821–30.
[11] Priestley MJN, Calvi GM, Kowalsky MJ. Displacement-based seismic design of
structures. IUSS Press; 2007 (721pp.).
[12] Ash C, Bartoletti S. Seismic rehabilitation of an existing braced frame hospital
building by direct replacement with buckling-restrained braces. In: Proceed-
ings of the ATC & SEI 2009 conference on improving the seismic performance of
existing buildings and other structures, Paper 68, 2009.
[13] Di Sarno L, Manfredi G. Seismic retrofitting of existing RC frames with buckling
restrained braces. In: Proceedings of the ATC & SEI 2009 conference on
improving the seismic performance of existing buildings and other structures.
Paper 741, 2009.
[14] Tremblay R, Bolduc P, Neville R, DeVall R. Seismic testing and performance of
buckling-restrained bracing systems. Canadian Journal of Civil Engineering
2006;33:183–98.
[15] Somerville PG, Smith N, Punyamurthula S, Sun J. Development of ground
motion time histories for phase 2 of the FEMA/SAC Steel Project. Report SAC/
BD-97/04, SAC Joint Venture, 1997.
[16] Santa-Ana PR, Miranda E. Strength-reduction factors for multi-degree-of-
freedom systems. In: Proceedings of the 12th world conference on earth
engineering, Paper 1446, 2000.
[17] Chopra AK, Goel RK. Building period formulas for estimating seismic displace-
ments. Earthquake Spectra 2000;16(2):533–6.
[18] Carr AJ. RUAUMOKO—Inelastic Dynamic Analysis Program. User’s manual.
Christchurch, New Zealand: Department of Civil Engineering, University of
Canterbury; 2004.
[19] Filiatrault A, Tremblay R, Wanitkorkul A. Performance evaluation of passive
damping systems for the seismic retrofit of steel moment-resisting frames
subjected to near-field ground motions. Earthquake Spectra 2001;17(3):
427–56.
[20] Federal Emergency Management Agency. Prestandard and commentary for the
seismic rehabilitation of buildings. Report FEMA 356, Washington, DC, 2000.
[21] Clark P, Aiken I, Kasai K, Ko E, Kimura I. Design procedures for buildings
incorporating hysteretic damping devices. In: Proceedings of the 68th
annual convention of the Structural Engineers Association of California,
1999, p. 355–71.
[22] Teran-Gilmore A, Virto Cambray N. Preliminary design of low-rise buildings
stiffened with buckling restrained braces by a displacement-based approach.
Earthquake Spectra 2009;25(1):185–211.
[23] Qi X, Moehle JP. Displacement design approach for reinforced concrete
structures subjected to earthquakes. Report UCB/EERC-91/02, University of
California at Berkeley, 1991.
[24] Sabelli R, Mahin S, Chang C. Seismic design on steel braced frame buildings
with buckling-restrained braces. Engineering Structures 2003;25(5):655–66.
[25] Applied Technology Council. FEMA 306, Evaluation of earthquake damaged
concrete and masonry wall buildings. Report no. ATC-43, Redwood City, CA,
1998.
[26] Teran-Gilmore A. On the use of spectra to establish damage control in regular
frames during global predesign. Earthquake Spectra 2004;20(3):1–26.
[27] Nassar AA, Krawinkler H. Seismic demands for SDOF and MDOF systems. Report no.
95, The John A. Blume Earthquake Engineering Center, Stanford University, 1991.
 Author's personal copy

490 A. Terán-Gilmore, J. Ruiz-Garcı́a / Soil Dynamics and Earthquake Engineering 31 (2011) 478–490

[28] Bertero R, Bertero VV. Tall reinforced concrete buildings: conceptual earth-
quake-resistant design methodology. Report UCB/EERC-92/16, University of
California at Berkeley, 1992.
[29] Teran-Gilmore A, Coeto G. Displacement-based preliminary design of tall buildings
stiffened with a system of buckling-restrained braces. Earthquake Spectra, in press.
[30] Pampanin S, Christopoulos C, Priestley MJN. Performance-based seismic
response of framed structures including residual deformations. Part II:
Multiple-degree-of-freedom systems. Journal of Earthquake Engineering
2003;7(1):119–47.
[31] Ruiz-Garcia J, Miranda E. Evaluation of residual drift demands in regular
multi-story frame buildings for performance-based assessment. Earthquake
Engineering & Structural Dynamics 2006;35(13):1609–29.
[32] Kiggins S, Uang CM. Reducing residual drift of buckling-restrained braced
frames as a dual system. Engineering Structures 2006;28(11):1525–32.