SEISMIC STRENGTHENING OF NINE-STOREY RC BUILDING USING

CONCENTRIC AND BUCKLING-RESTRAINED BRACING

A. Rahai, Amirkabir University of Technology, Iran
M. Lashgari*, Amirkabir University of Technology, Iran

31st Conference on OUR WORLD IN CONCRETE & STRUCTURES: 16 - 17 August 2006,

Singapore

Article Online Id: 100031039

The online version of this article can be found at:
http://cipremier.com/100031039

Thisarticleisbroughttoyouwiththesupportof
SingaporeConcreteInstitute
www.scinst.org.sg

AllRightsreservedforCIPremierPTELTD
YouarenotAllowedtoredistributeorresalethearticleinanyformatwithoutwrittenapprovalof
CIPremierPTELTD
VisitOurWebsiteformoreinformation
www.cipremier.com

31st Conference on OUR WORLD IN CONCRETE & STRUCTURES: 16 17 August 2006, Singapore

SEISMIC STRENGTHENING OF NINE-STOREY RC BUILDING USING

CONCENTRIC AND BUCKLING-RESTRAINED BRACING
A. Rahai, Amirkabir University of Technology, Iran
M. Lashgari*, Amirkabir University of Technology, Iran

Abstract
Steel bracing members are widely used in steel structures to reduce lateral
displacement and dissipate energy during earthquake motions. Concentric steel
bracing provide an excellent approach for strengthening and stiffening existing RC
buildings. Using these braces the designer can hardly adjust the stiffness together
with ductility as needed because of buckling of braces in compression. Encased
bracing (buckling-restrained bracing) can permit designer choosing required stiffness
and strength together with high ductility independent of susceptibility to buckling.
These braces are composed of steel core member enclosed in concrete-filled square
steel tube. In this study the use of steel bracing and buckling-restrained bracing
(BRB) for retrofitting an inadequate reinforced concrete building are investigated. The
effectivness of these two systems in rehabilitating a mid-rise nine-storey reinforced
concrete (RC) building were examined using performance-based design and
nonlinear static analysis according to FEMA-356 seismic rehabilitation guidelines.
Results show that both systems improve the strength and stiffness of the original
structure but due to excellent behavior of BRBs in nonlinear phase and under
compressive forces this system shows much better performance than the
rehabilitation system of concentric bracing.
Keywords: Buckling Restrained Bracing (BRB), Concentric Bracing, RC Building, Retrofitting,
Capacity, Demand, Seismic
1. Introduction
A large number of existing RC buildings in seismically active regions of iran are inadequate by
current code requirements. In many cases these structures have deficiencies in lateral strength and/or
ductility. Common causes of deficiencies are: insufficient design loads, changing seismic codes
criteria, flawed design or construction and changing in building occupancy. Unless these structures are
retrofitted, they may perform poorly during an earthquake.
Different retrofitting schemes may be used to upgrade the seismic performance of exsisting RC
structures. The two main approaches for structural rehabilitation are to add new structural elements
such as wall and steel bracing or to selectively strengthen deficient structural elements of building.
Experimental research on seismic rehabilitation of RC frames using steel systems indicates that in
general these systems significantly improve the strength and the stiffness of the RC frames [1,2].
Using steel bracing systems for seismic rehabilitation of RC frames are also investigated analytically
[3,4,5,6]. A frame fabricated by incorporating in concentric bracing system has the following character

and problems. When braces of high slenderness ratio are used and tensile forces work on the braces
the deformation capability is high. But when the compressive deformation work on these braces they
can not bear compressive forces and lateral deformation occurs easily and as a result they show sliptype hysteresis behavior under cyclic loading. When slenderness ratio of the braces are sufficiently
small they can bear compressive forces and show good dynamic behavior under cyclic loading. When
the slenderness ratio of the braces are intermediate between the two those above mentioned the
dynamic behavior is between the two characteristics explained. Because of embrittling nature of
concentric steel braces under compressive forces its difficult to design a frame in which proper rigidity
together with high ductility capacity incorporated.
Considerable research has been devoted to development of braces that exhibit more ideal elastoplastic behavior. Research led to designing a brace that has a stable force-deformation characteristic
and enables compressive strength to be equal to tensile yield strength. This system called bucklingrestrained bracing (BRB) is commonly composed of a load carrying element (core member) and a
lateral support element (encasing member). The basic principle in the construction of a BRB is to
prevent buckling of central steel core by encasing it over its length in a steel tube filled with concrete or
mortar. As the axial loads are to be taken only by the steel core it needs to provide a slip surface or
unbonding layer between the steel core and the surrounding concrete. The BRB is able to develop
strength and energy dissipation capacity of the steel core used to fabricate the load carrying element
[7,8,9,10].
The concept of using steel bracing in rehabilitating inadequate RC frames can be extended to
buckling-restrained bracing system. The objective of the current study is to evaluate and compare the
rehabilitation results of a nine-storey RC building using concentric bracing and buckling-restrained
bracing. Behavior of the original and retrofitted building were evaluated by conducting Nonlinear static
analysis using the computer program PERFORM-3D. FEMA-356 [11] seismic rehabilitation guidelines
used to evaluate the performance of the exsisting building and rehabilitation cases.
2. Overview of the buckling-restrained brace
The basic principle in the construction of a buckling restrained brace is to prevent euler buckling
of a central steel core by encasing it over its length in a steel tube filled with concrete or mortar. In
order to achieve stable force-deflection relation that allows the brace to yield under compressive forces
it is necessary that the core member encased in a steel tube having euler load at least 1.5 times the
yield load of the core ( Pe / Py 1.5 ) [7]. Figure 1 illustrates an overall view of a BRB and figure 2 and 3
illustrate the end details of a BRB.

Figure 1. Overall view of a BRB

Steel tube

Mortar

Core plate
Splice plate

Stiffner plate

Interior reserve space

Figure 2. Detail of ends

Unbonding material

Figure 3. End section

3. Description of the selected building

The building selected for study was originally a six-storey RC building. The owners decided to add
three more stories and complete the building for commercial purposes. The first storey is used as
parking space while the top eight stories are used as apartments. The building is 11.6 m wide ,14.5 m
long and a floor-to-floor storey height of 3.2 m. Figure 4 shows a plan view of typical floor. Its structural
system is a moment resisting space frame (MRSF), consists of 50 mm thick concrete topping on top of
joist and block slabs (figure 7). RC square columns cross-sections are shown in figure 5.

Y
(c)
X

(b)

(a)

Figure 4. Typical floor plan of building

Figure 5. Reinforcing details of columns :

(a) Storey 1,2 (b) Storey 3,4 (c) Storey 5,6,7,8,9

The exterior and interior typical beam sections in X direction are also shown in figure 6. The
concrete compressive strength is 25 MPa and the steel Yield strength is 300 MPa. The design live
load and the ground snow load are taken according to iranian code of practice for loading.

Figure 6. Detail of beams in X direction of building

Figure 7. Detail of floor slab

4. Analytical evaluation of the existing building

The nine-storey building was modelled as a 3D space frame with rigid diaphragm at each floor
level. As the retrofitting of building in X direction is going to be described in this paper all the
characteristics and results belong to X dirction of building. The ultimate base shear ratio of the
exsisting structure in X direction determined using pushover analysis is 0.074 W while at the base
shear ratio of 0.025W fisrt plastic hinging occurs and the structure passes the linear limit, where W is
weight of the structure. The base shear design value mandated by the new iranian seismic code [12] is
0.091W. In the calculation of the code design base shear, accidental eccentricity provisions are taken
into account. The ultimate lateral load resisting capacity of the building is lower than the base shear
design value. A pushover analysis is conducted to evaluate performance of the structure and

performance curve of existing building is shown in figure 8. Using the displacement coefficient method
With an initial period of 2.17sec the displacement demand is 480mm. The designed building must
achieve life safety (L.S) performance level under the design earthquake with a 10 percent probability
of exceedance in 50-year period in order to meet iranian code minimum requirements (the basic safety
objective). But analysis shows that at a displacement of 289 mm the building passes the life safety
performance level and goes into collapse prevention performance level. The rahabilitation of existing
building was aimed at increasing the stiffness and strength in order to achieve L.S performance level.
1750

CP

LS
IO

Base shear (KN)

1500
1250
1000

9-Existing
750

first hinge

500
250
0
0

100

200

300

400

Roof displacement (mm)

Figure 8. performance curve of exsisting building

5. Rehabilitation schemes of building
In figure 9, details of four rehabilitation cases are shown. In all cases exterior frames in X
direction are selected in which braces to be installed. In the cases 9-x and 9-v the effect of steel
concentric bracing with X-shape and inverted V-shape are investigated. Rehabilitation cases 9b-v
and 9b-d are used to investigate the effect of V-shape and diagonal BRBs on performance of
rehabilitated structure. In all rehabilitation cases brace distribution over the height of building is
selected according to seismic demand of each level. Double channel Section is used for concentric
brace members. Sections of the concentric brace members used in 9-x and 9-v cases are listed in
table 1. Rectangular and box sections are used in BRBs as the core and lateral support members.
Table 2 and 3 show the sections of BRBs used in the rehabilitation cases 9b-d and 9b-v. Selection
of these sections in BRBs is based on equation (1).

Pe / Py 1.5

(1)

Where Pe is the euler load of the encasing member and Py is the yield load of the core
member. Using a qualified steel tube compose a BRB having the behavior under compressive
loads the same as the behavior under the tensile loads. As the Stable hysteresis characteristic of
these braces under cyclic loading is proved up to a strain of 2% This allows the brace to be
modelled having modeling parameters and acceptance critera of a tensile brace member.
Since the two top stories of existing building have enough lateral strength capacity, in all
rehabilitation cases no brace is installed in order to better distribute the inelastic deformation along
the height of building.
Table 1. Sections of brace members used in the rehabilitation cases 9-x and 9-v
Storey
level

Brace
section

2C140

2C140

2C160

2C160

2C140

2C120

2C100

(a)

(b)
(c)
Figure 9. Rehabilitation cases :
(a) case 9-x; (b) case 9-v; (c) case 9-bd; (d) case 9-bv

(d)

Table 2. Sections of BRBs used in the case 9-bd

Storey

Core dimensions

Tube dimensions

Pe

Py

level

mm mm

mm mm mm

KN

KN

1
2
3
4
5
6
7
8
9

PL-100*20
PL-100*20
PL-120*30
PL-120*30
PL-120*22
PL-100*16
PL-100*12
-

160*160*6.3
160*160*6.3
150*150*5
150*150*5
150*150*5
150*150*5
150*150*5
-

1269
1127
1339
1339
1339
814
814
-

480
480
864
864
634
384
288
-

Pe / Py
2.64
2.35
1.55
1.55
2.12
2.12
2.83
-

Table 3. Sections of BRBs used in the case 9b-v

Storey

Core dimensions

Tube dimensions

Pe

Py

level

mm mm

mm mm mm

KN

KN

1
2
3
4
5
6
7
8
9

PL-100*16
PL-100*16
PL-120*30
PL-120*30
PL-120*22
PL-100*16
PL-100*12
-

140*140*5.6
140*140*5.6
140*140*5.6
140*140*5.6
140*140*5.6
140*140*5.6
140*140*5.6
-

2530
1780
2350
2350
2350
1670
1670
-

384
384
864
864
634
384
288
-

Pe / Py
6.59
4.64
2.72
2.72
3.71
4.35
5.88
-

6. Analytical evalution of the rehabilitation cases

All rehabilitation cases were modelled as 3D space frames with the braces added. The effects
of masonary infilled walls were ignored.
6.1. The stiffness and the lateral strength capacity
A pushover analysis was conducted to evaluate the stiffness and strength of four rehabilitation
cases. Triangular lateral load distribution was used for the pushover analysis. Figure 10 shows the
Performance curves of the existing and rehabilitation cases. The initial stiffness and lateral strength
capacity and ratios between these values in rehabilitation cases and existing building are given in table
4. The rehabilitation case 9-x had the highest stiffness and strength among the all rehabilitation cases
and the cases 9b-d and 9b-v also showed a great increase in stiffness and strength of the original
building. The case 9-v had an increase in stiffness of original building but due to unbalanced forces of
the braces in compression and tension, the concrete beams to which the braces were attached
experienced shear failure at a very low displacement. If local strengthening of the concrete beams is
done the performance curve of the case 9-v is continuous as shown in figure 10.

0.25

Base shear (V/W)

0.2

9-Existing
9-X
9-V
9b-d
9b-V

0.15

0.1

0.05

0
0

0.2

0.4

0.6

0.8

1.2

Roof drift ratio (%)

Figure 10. Performance curves of various rehabilitation cases from pushover analysis

Table 4. The initial stiffness and lateral strength capacity

of existing building and rehabilitation cases
Initial stiffness

Ultimate base shear

Stiffness

Strength

KN / mm

ratio (V/W)

ratio

ratio

9 Existing

11.7

0.072

9 X

35.2

0.206

3.01

2.84

9 V

29.5

0.117

2.52

1.61

9b d

29.7

0.165

2.54

2.16

9b V

33.7

0.185

2.88

2.57

modelled case

6.2. Performance level

The performance point of rehabilitated buildings is calculated using displacement coefficient
method. Table 5 shows the performance point of various rehabilitation cases and roof lateral
displacement at each performance level. In the cases 9-x and 9-v uplift of foundations at a low
displacement under the lateral forces caused an increase in initial stiffness and period of these
cases.
Table 5. Performance point of rehabilitation cases
Roof lateral displacement
Case

Displacement at

at performance levels (mm)

first hinging ( mm )

Roof lateral displacement

At Performance point ( mm )

IO

LS

CP

9 Existing

44

166

289

352

480

9 X

87

144

264

301

242

9 V

73

88

88

95

290

9b d

58

125

360

435

283

9b V

51

67

341

420

246

In the rehabilitation case 9 X hinging starts in the beams of third and fourth floors and spreads
to beams of other floors. Buckling and yielding of compressive and tensile braces of all stories were
observed except those of first and second stories Due to high stiffness of first and second stories
columns and uplift of foundations. Hinging of columns was also observed more in columns
neighbouring the braced bays. In this case the building remained at life safety performance level and it
achieved the rehabilitation objective. In the rehabilitation case 9-V hinging starts in beams attached to
braces. As the lateral displacement increases compressive brace members buckle but no tensil brace
yield due to downward deflection of concrete beam at intersections with brace members. Finally at a
displacement of 95 mm the concrete beams experienced shear failure. Since the bay length of
Building varies from one to another and behavior of CBF member changes as the length of member
changes, using different bays for installing CBF members leads to a system in which balance of rigidity
and strength along the height of structure is difficult and lateral displacement tends to concentrate in a
few stories.
In BRB members choosing the proper rigidity together with high ductility is possible so in the
rehabilitation cases 9b-d and 9b-v selecting every three bays of exterior frames in which the braces to
be installed is possible. In these cases the sections of braces are chosen according to table 2 and 3 in
order to distribute inelastic deformation uniformaly along the height of structure. In these cases hinging
starts in beams and columns together with yielding of all the brace members. As a result of
distribution of brace members among the bays , the uplift forces decreased and the foundation uplift
was not observed and this makes these cases to use full stiffness and strength of BRBs capacity.
These two cases met the requirements for selected rehabilitation objective and they remained at life
safety performance level at higher roof lateral displacement. In the case 9b-v no considerable
unbalanced vertical force induced in the concrete beams As the BRBs dont buckle under compressive
forces.
6.3. Redundancy factor
Structural response modification factor (R) of a building is a product of three factors: R S , a perioddependent strength factor, R a period-dependent ductility factor, and R R , a redundancy factor. An
additional

R S accounts for the increase the lateral strength of the structure from the design strength to
the strength associated to the formation of the first plastic hinge and reductions due to redundancy
( R R ) accounts for the increase the lateral strength of the structure from the strength associated to the
formation of the first plastic hinge to the strength associated to the formation of a mechanism. Thus the
reduction factor to be used in design would be given by:
(2)
R R RS RR

A higher value of R R in the structure means a better plastic mechanism that produce a higher
value of response modification factor. Response modification factors were calculated as 1.7, 1.19,
2.19 and 2.43 for the rehabilitation cases 9-x, 9-v, 9b-d and 9b-v, respectively. This indicates that in
inelastic deformation phase the cases 9b-d and 9b-v have a significant better behavior.
7.

Conclusion
The analysis performed on the rehabilitation case 9 X indicated that X bracing system
significantly improves the stiffness and the strength of structure. It also showed an upgrade in
structural performance level. In the case 9-v Due to unbalanced forces of compressive an tensile
braces, premature shear failure occured in beams and caused a downgrade in performance level of
original structure. Due to excellent behavior of buckling restrained bracing system under compressive
and tensile forces and adequate stiffness and strength of them, in rehabilitation cases in which BRB
system is used a great increase in the stiffness and the strength was observed. Furthermore a
significant upgrade in structural performance level was made and these rehabilitation cases achieved
the life safety performance objective for the design earthquake.
As it mentioned before the value of RR in the structure depends on plastic mechanism of
structure. Among all rehabilitation cases the two in which BRB system were used indicated a
considerable higher redundancy factor. All results prove better behavior of BRB system by comparison
to concentric bracing system in rehabilitating a nine-storey inadequate RC building.
Another advantage of buckling restrained bracing over concentric bracing system is that in BRB
members as the forces can be fully controlled and the buckling is inhibited they induce controlled
forces in end connections and have no considerable shocking impact on them, therefore transfer of the
forces between the RC frame and the steel system is safer.
References

[1] Goel, S.C., Lee, H.-S., "Seismic strengthening of RC structures by ductile steel bracing system",
Proceedings of Fourth U.S National Conference on Eartquake Engineering, Earthquake Engineering
Reasearch Inst., El Cerrito, California, Vol.3 , 1990, pp. 323-331
[2] Bosh, T.D, Jones, E.A., Jirsa J.O., " Behavior of RC Frame Strengthened using Structural Steel Bracing",
Journal of Structural Engineering, Vol.117, No.4, April 1991, pp. 1115-1126
[3] Pincheira, J.A., Jirsa, J.O., "Seismic Response of RC Frames Retrofitted with Steel Braces or Walls",
Journal of Structural Engineering, Vol.121, No.8, August 1995, pp. 1225-1235
[4] Nateghi-A, F., "Seismic Strengthening of Eight-Story RC Apartment Building using Steel Braces",
Engineering Structures, Vol.17, No.6, 1995, pp. 455-461
[5] Abouelfath, H., Ghobarah, A., "Behavior of Reinforced Concrete Frames Rehabilitated With Concentric
Steel Bracing", Can.J.Civ.Eng., Vol.27, 2000, pp. 433-444
[6] Ghobarah, A., Abouelfath , H., "Rehabilation of Reinforced Concrete Frame using Eccentric Steel bracing",
Engineering Structures, Vol.23, 2001,
pp. 79-104
[7] Atsushi, W., Yasuyoshi, H., Eiichiro, S.,Akira , W., Morihisa, F., "Properties of Brace Encased in BucklingRestraining Concrete and Steel Tube", Proceedings of 9th World Conference on Earthquake Engineering,
Tokyo-Kyoto, Japan, Vol.4, 1988, pp. 719-24
[8] Watanebe, A., et al., "Properties of Brace Encased in Buckling-Restraining Concrete and Steel Tube,
Proceedings, 9WCEE Organizing Committee, Jappan Assn. for Earthquake Disaster Prevention, Tokyo,
Vol.4, 1989, pp. 719-724
[9] Nagao, T., Takahashi, S., "A Study on the Elasto-Plastic Behavior of Unbonded Composite Bracing (Part
1:Experiments on isolated Members under cyclic loading) (in Japanese) ", Journal of Structural and
Construction Engineering (Transactions of AIJ), Vol.415, 1990, pp. 105-115
[10] Nagao, T., Takahashi, S., "A Study on the Elasto-Plastic Behavior of Unbonded Composite Bracing- Part
2:Analytical Studies(in japanese), "Journal of Structural and Construction Engineering (Transactions of AIJ),
Vol.422, 1991, pp. 45-56
[11] FEMA-356, Prestandard and Commentary for the Seismic Rehabilitation of Buildings", ASCE, 2000
[12] Iranian seismic building code, No.2800, BHRC, Tehran, Iran