See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/273573704

SEISMIC RESPONSE OF R.C. FRAMES WITH CONCENTRIC STEEL BRACING

Conference Paper · April 2006

CITATIONS READS
9 548

3 authors, including:

Ashraf Osman Ahmed Atef Rashed

Cairo University Cairo University
73 PUBLICATIONS 346 CITATIONS 14 PUBLICATIONS 67 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Ashraf Osman on 15 March 2015.

The user has requested enhancement of the downloaded file.

 Proceedings of the 8th U.S. National Conference on Earthquake Engineering
April 18-22, 2006, San Francisco, California, USA
Paper No. 1979

SEISMIC RESPONSE OF R.C. FRAMES WITH

CONCENTRIC STEEL BRACING

A. Osman1, A. Rashed1,2 and M. El-Kady3

ABSTRACT

This paper summarizes the results of an analytical study conducted to investigate

the seismic behavior of concrete frames designed for gravity-load and upgraded
using concentric steel bracing. Effects of different brace design aspects, such as
brace cross-sectional area, slenderness ratio and brace arrangement, on the
response of strengthened frames under both static monotonic and earthquake time
history loads were examined. Guidelines for appropriate design and proper
arrangement of bracing members are provided.

Introduction

In seismic areas, reinforced concrete (R.C.) frames proportioned mainly for gravity and
wind loads with no consideration for seismic forces are usually considered as a great threat to
their occupants due to the possibility of partial or total collapse during earthquakes. Mitigation
of such failure possibility can be achieved by strengthening these frames. Concentric steel
bracing is one of the retrofitting techniques that are widely used for upgrading these gravity-load
designed R.C. frames. They enhance frame capabilities to withstand earthquakes through
improving their lateral strength and stiffness without affecting their weights and architectural
appearance.

Previous research studies conducted in this field (Pincheira and Jirsa, 1995 and Goel and
Lee, 1990) indicated that steel bracing is an economical and feasible scheme for seismic
upgrading of R.C. frames. However, few drawbacks were associated with the utilization of such
scheme. They included: 1) The significant changes in the magnitude of axial forces within the
columns forming the braced bays and the need for jacketing them heavily; 2) The need for
retrofitting the existing foundation underneath these highly stressed columns to sustain the
expected higher compression forces or the possible uplift forces; and 3) The reduction of the
building fundamental period which result in attracting higher seismic forces.

Generally, the first problem can be overcome by strengthening the columns within the

1
Professor, Structural Engineering Dept., Cairo University, Giza, Egypt
2
Vice Chairman, Hamza Associates Consulting Office, Cairo, Egypt
3
Structural Consulting Engineer, Cairo, Egypt
braced bays using steel, concrete or RFP jackets to allow them to sustain the developed high
axial compression/tensile forces. The second problem can be mitigated by proper arrangement of
steel bracing to avoid applying high pressure/uplift forces on the footings and minimizing the
need for retrofitting both footings and columns, while the third problem can be minimized by
appropriate selection of slenderness ratios and cross sectional areas of bracing members.
However, details and guidelines for these remedial measures are not available yet. Therefore, in
this study a detailed analytical investigation is performed to examine the effects of different
design aspects such as brace cross-sectional area, slenderness ratio, and bracing arrangement on
the seismic response of strengthened frames with the aim of achieving safe and economical
design guidelines for this retrofitting scheme.

Analytical Study

Analyzed Structures
For the purpose of this study, two hypothetical skeleton gravity load designed concrete
buildings were selected. One building is a low-rise structure with five stories, while the other is a
medium-rise building with ten stories. The buildings heights as measured from the ground level
were 17 m and 32m, respectively, as shown in Fig. 1. The buildings floor plans were 13.5 m by
18 m, symmetrical about the buildings axes and comprising four equal bays in the X-direction
and three equal bays in the Y-direction (see Fig. 2). The buildings main structural systems were
frames designed mainly for dead, live and wind loads with no consideration for earthquake
loads. The design dead loads consisted of the structure self weight in addition to the weight of
floor finish having an intensity of 200 kg/m2, the loads for exterior masonry cladding walls and
the weight of internal partitions. Live load of intensity 200 kg/m2 was considered as uniformly
distributed over the floor area. Wind loads in accordance with the old provisions of the Egyptian
Code (El Behairy, 1979) were utilized for the design in both X and Y directions. The buildings
were then evaluated for safety against both wind and earthquake loads according to the new
provisions of the Egyptian Code for calculating loads and forces in buildings and structural
works (Housing and Building Research Center, 1992) and found to be unsafe in the X-direction.
Consequently, the two structures had to be strengthened in this direction by adding steel braces
within the third bay of the exterior frames along axes A & D.

Computer Modeling

To model analytically these hypothetical frames, the computer code Drain-2DX was used
(Parakash and Powell, 1993). Due to buildings symmetry, only half of each building was
considered and represented as shown in Fig. 3 by an exterior braced frame and interior one,
connected at each floor level by fictitious rigid link simulating the effect of the rigid floor
diagram. Masses were lumped at beam column joints. To simulate the buildings inelastic
response, the beams were modeled using a bilinear, non-degrading hysteretic model, while
columns were modeled as elastic-perfectly plastic element with axial force-moment interaction
curve. Cracking was considered in the beam and column model by reducing their initial stiffness
to 40% of their gross inertia. The steel braces responses were simulated using a bi-linear
hysteretic model for axially loaded pin-ended braces. The hysteretic law for the brace included a
non-degrading elastic-perfectly plastic behavior in tension and inelastic buckling in
compression.

Figure 1. Elevations for the Five Story and Ten Story Buildings

Figure 2. Typical Plan for the Five Story and Ten Story Buildings
 Beam-column connections were modeled as springs with high rotational stiffness
estimated using the procedure described by El-Kady, 2000. To detect the failure of the
foundation under the high induced axial compressive forces or possible uplift. The footings were
modeled as spring-like support. The force-displacement behavior for these springs were
presented by an elastic-perfectly plastic relationship, where compression yielding occurs at the
footing when axial forces within the columns bay equate the ultimate bearing capacity, while
tension yielding, which represents the case of uplift, occurs when the tension force exceeds the
weight of the footings and the backfilled soil above it.

FRL

FRL

FRL

FRL

FRL

Kc Kc Kc Kc Kc Kc Kc Kc Kc Kc

Ks Ks Ks
Ks Ks Ks Ks Ks Ks Ks

Exterior Fram e Interior Frame

Symbol Defination
Rotational Spring of Beam-Column Connection

Translational spring in vertical direction

Kc Translational spring for Column Yield model

Ks Translational spring for Soil- Foundation model
(FRL) Fictitious Rigid Link Member

Figure 3. Descretized Model for the Five Story Braced Building

Loads

The considered structures were analyzed for both static and dynamic loads. In addition to
static gravity loads, lateral static loads were applied as increasingly static monotonic load having
a triangular loading pattern simulating the building 1st mode shape. The dynamic loads were
base excitation to building, simulating the earthquake induced motion. Three earthquake records
having different (a/v) ratio were considered, namely the N-S component of the 1985 Mexico
City earthquake with low a/v ratio (a/v = 0.54), the N-S component of the 1940 Imperial Valley
earthquake recorded at El-Centro with intermediate a/v ratio (a/v = 0.97), and the 1971 San
Fernando earthquake with high a/v ratio (a/v = 1.76). The peak ground acceleration for the three
records were scaled to 0.15g to match the maximum expected acceleration in Cairo city.

Analyzed Cases

Parametric Study

The performed parametric study involved examining the effects of bracing design aspects
such as bracing slenderness, bracing cross-sectional area and bracing arrangement on the static
and dynamic response of strengthened buildings. Fifteen brace sections, categorized into three
groups were selected and utilized for both buildings to examine the response of the frames
having steel bracings with different slenderness ratios and cross-sectional areas. Eight different
proposed arrangements for bracing, covering a range of practical applications that satisfy
different architectural demands and structural requirements, were analyzed to assess the effect of
brace arrangement on the frames response. Examined parameters included, maximum story
shear, maximum story displacement, maximum brace ductility demand at each story, maximum
beams and columns rotation demand at each floor, maximum column axial compression at each
story, and uplift induced at foundation.

Effects of Brace Cross-Sectional Area and Slenderness Ratio on Building Response

To evaluate the effects of the brace cross-sectional area and slenderness ratio on the
buildings behavior, static and time-history analyses were performed on upgraded buildings
utilizing braces belonging to three different categories of steel bracing.

Table 1. Bracing Cross-Sectional Area and Slenderness Ratio for Analyzed Cases

Cat. Section Brace Section Area Slender. Comp.

No. (cm2) ratio Strength
(ton)
A 1 2 UPN # 140 40.8 92.5 63.4
2 2 angles back to back (120X120X11) 50.8 93.4 78.2
3 2 angles back to back (120X120X13) 59.4 93.4 91.5
4 2 angles back to back (120X120X15) 67.8 93.4 104.4
B 5 2 angles back to back (65X130X10) 37.2 74.6 67.3
6 2 angles back to back (80X120X10) 38.2 87.5 62.1
7 2 angles back to back (100X100X10) 38.4 112 48.8
8 2 angles back to back (90X90X11) 37.4 124.4 40,8
9 2 angles star shape (100X100X10) 38.4 130.9 38.5
10 2 angles star shape (90X90X11) 37.4 145.5 30.6
CA 11 2 angles back to back (65X130X10) 37.2 74.6 67.3
12 2 angles back to back (110X110X12) 50.2 101.8 71.1
13 2 angles star shape (120X120X13) 59.4 109.1 77.8
CB 14 2 angles back to back (100X100X10) 38.4 112 48.8
15 2 angles star shape (100X100X14) 52.4 131 52.5

The first category, designated as group A, included a set of steel bracing members having
varied cross-sectional areas, while their slenderness ratios were kept approximately constant. It
worth to mention here, that brace slenderness in this study is defined as the brace buckling
length divided by the brace radius of gyration. The second category, designated as group B,
contained six bracing members having varied slenderness ratios, while the braces cross-sectional
areas were kept equal. In the third category, designated as CA or CB, both the braces cross-
sectional areas and the slenderness ratios were varied, while the brace compressive strengths
according to AISC specifications (1999) were kept constant for each group. Table 1 illustrates in
details the braces slenderness, cross-sectional areas and compressive strength for the analyzed
groups.
0.2
Sec#3 (A=59.4 cm2)
Sec#4 (A=67.8 cm2)

0.15
Sec#2 (A=50.8 cm2)
Base Shear (V/W)

Sec#1 (A=40.8 cm2)

0.1

0.05

0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
Roof Displt. (m)

Figure 4. Force Displacement Relationship for Group A

0.2
Sec#6 (Kl/r = 87.5)

Sec#5 (Kl/r =76.4)

0.15
Base Shear (V/W)

Sec#7 (Kl/r= 112)

Sec#8 (kl/r = 124.4)

0.1 Sec#9 (kl/r = 131)

Sec#10 (Kl/r =145.5)

0.05

0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
Roof Displt. (m)

Figure 5. Force Displacement Relationship for Group B

Figures 4 and 5 show the base shear-roof displacement relationships for the strengthened
five story buildings utilizing braces belonging to groups A and B, respectively, under the action
of static monotonic loads. Examining these figures indicated that varying the braces cross-
sectional area affects significantly the buildings initial elastic stiffness contrary to varying braces
slenderness ratio. Furthermore, it can be noted that the building ultimate strength, defined as the
maximum attained base shear at which building strength start to degrade, increases with the
 increase in the brace cross-sectional area and with the decrease in brace slenderness. At the same
time, examining the results of braces belonging to group CB indicated that brace cross-sectional
area effect is more pronounced compared to brace slenderness (Fig. 6). With regard to other
parameters, it was concluded that changing the brace cross-sectional area has a negligible
influence on the uplift magnitude.
0.2

Sec#15 (A=52.4cm2, Kl/r=131)

0.15
Base Shear (V/W)

Sec#14 (A=38.4cm2, Kl/r=112)

0.1

0.05
Group CB

0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
Roof Displt. (m)

Figure 6. Force Displacement Relationship for Group CB

To study the effect of brace different design aspects on the dynamic response of the
strengthened frames, time history analyses were carried out using previously indicated three
earthquake records on buildings retrofitted by the fifteen brace sections indicated in Table 1.

5 5

4 4
Floor Level

Floor Level

3 3

2 2

1 1
0 0
-0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2 - 15 - 10 -5 0 5 10 15
Story Shear (V/W)
Displacements (cm)
5
5

4
4
Floor Level

3
Floor Level

3
2
2

1
1
0
0
0 0.5 1 1.5 2
0 0.25 0.5 0.75 1
Beam Rotation Factor (Induced/Capacity) Column Axial Compression Load Factor (Induced/Capacity)

Figure 7. Dynamic Response of the Five Story Frames due Mexico Earthquake
( A=40.8 cm2, A=50.8 cm2, A=59.4 cm2, A=67.8 cm2)
 Figure 7 shows as a sample the response of the five story frames utilizing bracing
members belonging to category A under the action of Mexico earthquake. The response is
represented in terms of the envelopes for story shear and story displacement and the beam and
columns rotational demand at each story. As previously concluded, the larger the bracing cross-
sectional area, the higher the building stiffness and the higher the attracted earthquake forces
and base shears. On the other hand, the larger the brace cross-sectional area is, the smaller the
building lateral displacement. It should be noted here that an increase in the area by 70%
resulted in increasing the base shear by 40% and reducing the lateral displacement by 55%.
Also, increasing the brace area resulted in reducing both beam and column rotational demands.
However, it tends to increase the column compressive axial forces and the uplift. On the other
hand, results obtained from changing brace stiffness supported the conclusions previously
derived from the static analyses related to the same aspect. Also, conclusions derived from the
analyses for the ten story frames were similar to those obtained for the five story frames.

Effects of Brace Arrangement on Frame Response

Proper distribution of the bracing system along building height in the elevation can
overcome the problems associated with the need for foundation strengthening against uplift and
the heavy jacketing of columns. To examine this issue, eight different bracing arrangements
were proposed for the five story frames as shown in Fig. 8, and the uplift/axial compression
forces in columns produced by each brace arrangement under both static and dynamic loads
were examined. The results indicated that bracing the exterior bays such as arrangement #2
should be avoided, since it exerts high axial forces on the weak exterior columns requiring heavy
jacketing to them. Furthermore, arrangements such as 1, 3, 4, 5 and 8 produce large foundation
uplift forces requiring extensive retrofitting to foundation system as can be noted from Fig. 9.
On the other hand, arrangements 6 and 7 produced the minimum foundation uplift. With regard
to response of the analyzed frames, it was noted that arrangements 6 and 7 enhanced the frames
lateral stiffness by limiting the frame lateral drift, reduced the beam and columns rotational
demands and resulted in better energy dissipation, especially for arrangement 6. Therefore, it can
be concluded that the use of bracing arrangement similar to arrangement 6 improves the
structural performance of retrofitted building. Similar conclusions were obtained from the
analyses for the ten story frames.

Conclusions
Based on the results of this study, the following conclusions can be drawn:
1. Steel bracing is well suited for seismic upgrading of concrete gravity load designed frames. It
can successfully reduce the collapse potential and improve the energy dissipation capacity,
in addition to reducing the rotational ductility demands on the columns.
2. Optimum design for the bracing retrofitting scheme can be achieved through limiting the
brace slenderness ratio to 130 with the use of the smaller possible cross-sectional area.
3. Reducing the uplift on the foundation can be achieved by proper arrangement of the bracing.
The best arrangement requires avoiding bracing the exterior bays in frames and the need to
one interior bay totally through the height of the frame, in addition to at least two adjacent
bays to the braced bay at the ground floor.
 Arrangem ent N o. (1) Arrangem ent N o. (2)

A rrangem ent No. (3) A rrangem ent N o. (4)

A rrangem ent No. (5) A rrangem ent No. (6)

Arrangem ent No. (7) Arrangement No. (8)

Figure 8. Different Bracing Arrangements for the Five Story Building

 200

Mexico
180

160 El Centro

140
Foundation Uplift Force (tons
San Fernando

120

100

80

60

40

20

0
1 2 3 4 5 6 7 8
BRACING ARRANGEMENT N

Figure 9. Effect of Bracing Arrangement on Foundation Uplift for the Five Story Building due to
Different Earthquakes

References

AISC, 1999. Load and Resistance Factor Design Specification for Structural Steel Buildings, American
Institute of Steel Construction, Chicago, USA.

El-Kady, M., 2000. Seismic Strengthening of R.C. Buildings Using Steel Bracing, Ph.D. Thesis,
Structural Engineering Department, Cairo University, Egypt.
rd
El Behairy, S., 1979. Reinforced Concrete Design Handbook – 3 edition, Ain Shams University Press,
Egypt.

Goel, S.C., and H. Lee, 1990. Seismic Strengthening of RC Structures by Ductile Steel Bracing System,
Proc. Of Fourth U.S National Conference on Earthquake Engineering, California, 323-331.

Housing and Building Research Center, 1992. Egyptian Code for Calculating Loads and Forces in
Buildings Structural Works, Egypt.

Parakash, V., and G.H. Powell, 1993. DRAIN-2DX- A General Purpose Computer Program for Static and
Dynamic Analysis of Inelastic Plane Structures, UCB/SEMM-1993/17, University of California,
Berkeley.

Pincheira, J.A., and J.O. Jirsa, 1995. Seismic Response of RC Structures Frames Retrofitted with Steel
Braces or Walls, Journal of Structural Engineering, ASCE, 121 (8).

View publication stats