retrofitting of rc frames by steel braced frames utilizing

Pasha Javadi, Tetsuo YamakawaJournal of Advanced Concrete Technology, volume ( ), pp.11 2013 89-107

Seismic Retrofitting Methods Newly developed for Railway Concrete StructuresTadayoshi Ishibashi, Takeshi Tsuyoshi Kaoru Kobayashi,Journal of Advanced Concrete Technology, volume ( ), pp.2 2004 65-76

Investigation of a Hybrid Technique for Seismic Retrofitting of Bare FramesMd. N. Rahman, Tetsuo YamakawaJournal of Advanced Concrete Technology, volume ( ), pp.5 2007 209-222

Seismic Retrofit of Reinforced Concrete Building Structures with Prestressed BracesSusumu Kono, Takeshi KatayamaJournal of Advanced Concrete Technology, volume ( ), pp.7 2009 337-345

Seismic Damage of and Seismic Rehabilitation Technique for Railway Reinforced Concrete StructuresTadayoshi Ishibashi Daisuke TsukishimaJournal of Advanced Concrete Technology, volume ( ), pp.7 2009 287-296

,

Retrofitting of RC Frames by Steel Braced Frames Utilizing aHybrid Connection Technique

https://www.jstage.jst.go.jp/article/jact/2/1/2_1_65/_pdf




http://www.editorialmanager.com/jact/

http://www.j-act.org/index.html

Journal of Advanced Concrete Technology Vol. 11, 89-107, March 2013 / Copyright © 2013 Japan Concrete Institute 89

Scientific paper

Retrofitting of RC Frames by Steel Braced Frames Utilizing a Hybrid Connection Technique Pasha Javadi1 and Tetsuo Yamakawa2

Received 4 May 2012, accepted 21 February 2013 doi:10.3151/jact.11.89

Abstract This study presents a new connection technique to retrofit existing RC frames by steel braced frames. The proposed connection which is called the “hybrid connection” plays two important structural roles. The first role is to provide con-nection between an existing RC frame and a steel braced frame. The second role is to increase shear strength and axial compression capacity of boundary RC columns of retrofitted frames. The hybrid connection consists of steel plates, high-strength bolts, and high-strength grout. Experimental investigations were conducted on four one-bay one-story RC frames and one one-bay two-story RC frame which were retrofitted by the proposed method. Moreover, one one-bay one-story RC frame was tested as a non-retrofitted benchmark frame. Experimental results of the specimens indicated that the proposed hybrid connection transferred relatively high direct shear forces between the RC frames and the installed steel braced frames to obtain lateral capacity of the steel braces. In addition, the proposed hybrid connection prevented possible shear failure of the boundary RC columns with the help of the utilized steel plates. Taking into consideration ex-perimental results and observations, simplified design approaches are presented to estimate lateral strength of fundamental failure mechanisms of retrofitted RC frames.

1. Introduction

In past earthquakes such as the 1994 Northridge-USA, the 1995 Kobe-Japan, and the 1999 Izmit-Turkey, many reinforced concrete (RC) buildings significantly dam-aged or completely collapsed because of formation of the soft-story mechanism. In the soft-story mechanism, considerable lateral deformation occurs in a story that has a lower lateral strength and stiffness compared to its adjacent stories. In Japan, the Building Standard Law (BSL 1981) was revised in 1981 to upgrade required lateral resistances of buildings with irregular stiffness distribution in plan and/or along height of buildings. However, in Japan and other countries located on high seismic zones, many existing reinforced concrete build-ings are still vulnerable to the soft-story mechanism.

Application of steel braced frames is one of the common retrofit techniques for existing soft-story RC frames. Experimental investigations by Sugano and Fu-jimura (1980) and Higashi et al. (1980) indicated effec-tiveness of installed steel braces in increasing lateral strength of retrofitted RC frames. Kawamata and Oh-numa (1980) suggested a practical retrofit scheme for strengthening existing reinforced concrete frames by eccentric steel braces. Ohishi et al. (1988) and Sekigu-chi et al. (1988) presented practical applications for

retrofitting RC frames by steel braced frames with indi-rect connection which has been used as the typical retro-fit scheme in the retrofit guideline by the Japan Building Disaster Prevention Association (JBDPA 2001). In the method used by JBDPA (2001), a steel braced frame is indirectly connected to a RC frame by means of studs and anchors which are interlocked at the connection zone with the help of high-strength grout. Studs are welded to the boundary zone of the steel frame, and anchors are inserted into epoxy-injected holes provided on the boundary zone of the RC frame. Then, high-strength grout is cast in the boundary zone between the steel frame and the RC frame to interlock the studs and anchors. Another approach was presented by Jones and Jirsa (1986) and adopted in the guideline by the Federal Emergency Management Agency (FEMA-547 2006). In this method, a steel braced frame is directly connected to a RC frame by means of anchors inserted into epoxy-injected holes on the boundary zone of the RC frame. However, in both methods by JBDPA (2001) and FEMA-547 (2006), the primary concern is connection between a RC frame and a steel braced frame. In fact, a limited number of anchors cannot provide required shear force between a steel frame and a RC frame to obtain lateral capacity of steel braces. Furthermore, in the conventional methods (JBDPA 2001, FEMA-547 2006), an installed steel braced frame only increases lateral strength and stiffness of a retrofitted frame, while RC columns still suffer from possible shear failure in the case of non-ductile RC columns. So, in the current study, it is aimed at proposing a new connection method to provide required direct shear forces between a RC frame and a steel frame and to prevent possible shear failure of RC columns.

1Former Postdoctoral Researcher, Disaster Prevention Research Center for Island Regions, University of the Ryukyus, Okinawa, Japan. E-mail: [email protected] 2Professor Emeritus, University of the Ryukyus, Okinawa, Japan.

P. Javadi and T. Yamakawa / Journal of Advanced Concrete Technology Vol. 11, 89-107, 2013 90

A practical technique for seismic retrofitting of soft-story RC frames by adding wing-walls or panel-walls was proposed by Yamakawa et al. (2006a, 2006b), and Rahman and Yamakawa (2007). In the proposed method which is called the “thick hybrid wall technique”, chan-nel-shaped steel plates jacket boundary RC columns and are extended to bays of RC frames by additional plain steel plates. The steel plates are connected together by means of high-strength bolts. In fact, the steel plates and high-strength bolts make steel formworks inside bays of RC frames. Then, additional concrete is cast in the pro-vided steel formworks. One of the important technical points of the thick hybrid wall technique is to provide appropriate connections between additional wing-walls and existing RC columns. Considering structural advan-tages of the connection in the thick hybrid wall tech-nique, a new connection method which is called the

“hybrid connection technique” was proposed for retro-fitting existing RC frames by steel braced frames (Ya-makawa 2009; Yamakawa 2010). The proposed hybrid connection consists of steel plates, high-strength bolts, and high-strength grout. In the current study, comple-mentary experimental investigations are conducted to verify all possible failure mechanisms of RC frames retrofitted by steel braced frames with the help of the proposed hybrid connection. The experimental studies were conducted on five one-bay one-story RC frames and one one-bay two-story RC frame. The scale factor of the test specimens is 1/4-1/3, to model a low-rise school building designed according to the pre-1971 Building Standard Law of Japan (BSL 1971). The ex-perimental results indicated that the hybrid connection transferred relatively high direct shear forces between the RC frames and the installed steel braced frames.

High-strengthgrout

High-strength grout,σB = 48.7 and 43.7

for specimens R08B-F60 and

R08B-F75

Retrofitted frame Crack pattern and failure mechanism

100

104

90

90

45

b =

175

D = 175 100

125

Steel column

Steel beam

250

100

Stud(D13)

Anchor(D13)

The web of thebottom steel beamAnchor

(13φ)

Spiral hoop(3.7φ)

Stud (D13)

Spiral hoop(D13)

Spiral hoop(D13)

Stud (D13)

Det

ails

of c

onne

ctio

ns

V-R relationship

Stub

Beam connectionColumn connection Bottom connection

N N

Steelframe

RCframe

Conventionalconnection

Direct shearfailure

Shearfailure

Spec

imen

R08

B-F

60

Flexuralplastichinge

BucklingSheardamage to RC column

l

l

N = 0.2bDσB N

N = 0.2bDσB N

1000

1000

Conventionalconnection

Sizes of the steel frame and braces:

1500

BH-75x60x4.5x4.5

Sizes of the steel frame and braces:

1500BH-75x75x4.5x4.5

RCframe

N N

Sheardamage to RC column

λbrace = 63*

λbrace = 49*

Shearfailure

Spec

imen

R08

B-F

75

V (+)

(-)

V (+)

(-)

V (+)

(-)

V (+)

(-)

σ = 22.6 MPa, for the RC frameBσ = 333 MPa, for the steel braced framey

σ = 19.7 MPa, for the RC frameBσ = 354 MPa, for the steel braced framey

Direct shearfailure

Anchor(D13)

Failure mechanism: Firstly, flexural plastic hinges formed in the RC columns, and the steel braces buckled in compression and yielded in tension. Then, sudden degradation occurred in the V-R relationship due to shear punching failure happened at the top of the left-hand RC column and at the top connection, and shear failure happened at the top of the right-hand RC column .

Failure mechanism: Sharp degradation occurred in the V-R relationship due to shear punching failure happen at the top of the left-hand RC column and at the top connection, and shear failure happened at the top of the right-hand RC column .

-600

-400

-200

0

200

400

600

-5 -4 -3 -2 -1 0 1 2 3 4 5

V (kN) 0.8Vmax(+)

0.8Vmax(-)

Vmax (+)=+544.0 kN

R (%) = δ (cm)

Vmax (-)=-552.5 kN

Non-retrofitted

Retrofitted

-600

-400

-200

0

200

400

600

-5 -4 -3 -2 -1 0 1 2 3 4 5

R (%) = δ (cm)

V (kN) 0.8Vmax(+)

Vmax (+)=+570.7 kN

Vmax (-)=-580.6 kN

0.8Vmax(-)

Non-retrofitted

Retrofitted

Notes: λ*=kl/r; where k: effective length factor (for this case, k=1) , l: length of steel braces (see the specimens), r: gyration of the steel brace section about its weak axis. Longitudinal and shear reinforcements in the RC columns are 8-D10 (Pg=1.85%) and 3.7φ-@105 (Pw=0.12%), respectively. Longitudinal and shear reinforcements in the RC beam are 4-D13 (Pg=1.63%) and D6-@120 (Pw=0.43%), respectively. In the V-R relationship, V is the lateral resistance and R denotes the drift angle of the first story.

Fig. 1 An overview of the conventional connection method [unit: N, mm].


Furthermore, the hybrid connection prevented shear failure of RC columns. In addition to experimental in-vestigations, corresponding design approaches are sug-gested to calculate lateral strengths of fundamental fail-ure mechanisms of retrofitted frames.

2. An overview of the conventional method

The conventional connection method for retrofitting an existing RC frame by a steel braced frame is suggested in the guidelines by the Japan Building Disaster Preven-tion Association (JBDPA 2001). In order to give an overview of the conventional method, two RC frames retrofitted by steel braced frames utilizing the conven-tional connection are presented in Fig. 1 (Maeda and Yamakawa 2012). As shown in Fig. 1, the steel braced frames are indirectly connected to the RC frames by means of studs and anchors which are interlocked at the connection zones with the help of high-strength grout. In the retrofitting operation, studs are welded to the boundary of the steel frame, and anchors are inserted into the epoxy-injected holes provided on the boundary of the RC frame. Then, high-strength grout is cast in the boundary zones between the steel frame and the RC frame to interlock the installed studs and anchors. The specimens R08B-F60 and R08B-F75 shown in Fig. 1 were tested under constant axial forces and cyclic hori-zontal loading (Maeda and Yamakawa 2012). As shown in Fig. 1, the difference between the two specimens is the sizes of the steel braced frames.

In the specimen R08B-F60, the slenderness ratio of the steel braces is λ = 63 (see Fig. 1). In this specimen at R = 0.7%, steel braces buckled in compression and yielded in tension. Moreover, flexural plastic hinges formed at the top and at the bottom of the RC columns. By proceeding the loading test, at the drift angle of R = 2.5%, the lateral resistance V suddenly dropped due to shear punching failure at the top of the left-hand RC

column and shear failure at the top of the right-hand RC column. Simultaneously, direct shear failure happened at the top connection.

In the specimen R08B-F75, the slenderness ratio of the steel braces is λ = 49 (see Fig. 1). In the V-R rela-tionship of the specimen, as the lateral resistance V reached to its maximum value, rapid degradation hap-pened due to shear punching failure at the top of the left-hand RC column and shear failure at the top of the right-hand RC column. Simultaneously, direct shear failure happened at the top connection.

From the experimental results of the two specimens, it is obvious that serious damages, such as punching failure, shear failure and etc., are strongly likely to hap-pen in a RC frame retrofitted by steel braced frame with the help of the conventional connection. In fact because of low direct shear resistance of the conventional connection, it is difficult to reflect lateral capacity and behavior of steel braces in the global response of the retrofitted frames. Therefore, it is required to propose a connection with rela-tively high direct shear strength to capture lateral capacity of steel braces.

3. Experimental program of the proposed method

3.1 Typical retrofitting procedure of the test specimens Typical procedure of the proposed retrofit method is illustrated in Fig. 2. The retrofitting operation is identi-cal in all test specimens. Details of the retrofit schemes of the test specimens are explained in Section 3.3. The typical retrofitting procedure of the test specimens are expressed in the following items; - Placing the steel braced frame inside the RC frame: The fabricated steel braced frame with the inverted-V shaped braces is placed inside the existing RC frame.

Typical retrofit procedure:

(1) Placing the steel braced frame inside the existing RC frame, and primarily fixing the steel frame by two base plates

(2) Sandwiching the RC wing-wall column and the steel column by two lateral plain steel plates

(3) Jacketing the RC column and the steel column by the channel-shaped steel plate

(4) Sandwiching the top RC beam and the steel beam by two lateral plain steel plates

(5) Installation of bolts in the provided holes

(6) Casting high-strength grout

Note; Details of retrofit schemes are given in Fig. 5.

(1)

(2)

(3)

(4) (5)

(6)

Fig. 2 Typical retrofitting procedure of the proposed technique.


The steel braced frame is primarily fixed inside the RC frame by means of two base plates. Each base plate is connected to the stub through anchor bolts. The anchor bolts are inserted into epoxy-injected holes drilled in the stub. - Providing the steel formworks of the hybrid connec-tions at the boundary columns: Two different retrofit schemes are considered for boundary RC columns. One scheme is for columns with wing walls and another scheme is for columns without wing-walls. In the case of wing-wall columns, two plain steel plates are used (No. (2) in Fig. 2). In the case of columns without wing-walls, channel shaped steel plates are utilized (No. (3) in Fig. 2). After installation of the steel plates, the high-strength bolts are passed through the steel plates into the holes punched on the steel columns (No. (5) in Fig. 2). - Providing the steel formwork of the hybrid connec-tion at the top beam: The top beam of the RC frame is partially connected to the steel beam through the pro-posed hybrid connection. Two lateral plain steel plates sandwich the RC beam and the steel beam (No. (4) in Fig. 2). Two horizontal series of high-strength bolts stitch the two lateral sandwiching steel plates. One se-ries of the high-strength bolts is passed through the top RC beam, and another series is passed through the top steel beam.

- Casting high-strength grout: After installation of the steel plates and the high-strength bolts, high-strength grout is cast in the provided spaces in the formworks (No. (6) in Fig. 2). The steel plates are used as form-works for casting the grout. To ensure that the grout can easily flow and uniformly fill the provided spaces, about 10 mm gap is considered between the outer faces of the RC columns and the steel plates. 3.2 Fundamental failure mechanisms of the ret-rofitted specimens Idealized deformations of the fundamental mechanisms are illustrated in Fig. 3. It should be noted that it is as-sumed the RC subs are fixed to a rigid floor, however, in real buildings uplift of foundations is likely to happen. Specific structural characteristics of the fundamental failure mechanisms are shown in Fig. 3 and are ex-plained as follows; Mechanism (a): In this mechanism, the steel braces yield in tension and buckle (or yield) in compression. The RC frame and the steel frame behave in a flexural manner. Mechanism (b): In this mechanism, direct shear failure happens simultaneously at the top hybrid connection zone and at the top of the RC columns. Mechanism (c): In this mechanism, direct shear failure happens simultaneously at the bottom of the RC col-umns and in the anchor bolts of the base plates. Mechanism (d): In this mechanism, overall-overturning

Flexural plastic hingeN NVa

VbN NDirect shear failure

Buckling of steelbrace

N

Mechanism (a) Mechanism (b)

NVc

VdN N

Mechanism (c) Mechanism (d)

Direct shear failure

Overturning

Steelframe

RC frameConnection

Connection

RC beam Steel beamRC column

RC column

Steelcolumn

RC column

Steelcolumn

Fig. 3 Idealized deformations of the fundamental failure mechanisms.


behavior occurs in the retrofitted frame due to vertical slip of the longitudinal reinforcements at the bottom of the boundary RC column and due to vertical slip of the anchor bolts. 3.3 Reinforcement details and retrofit schemes of the test specimens Many reinforced concrete buildings collapsed during the 1995 Kobe-Japan earthquake due to brittle shear failure in columns. The same failure mode was observed in school buildings after the 1968 Tokachi-oki earthquake in Japan. The Building Standard Law of Japan was re-vised in 1971 (BSL 1971) to require close spacing of lateral (tie) reinforcements in columns (Otani 2004). Details of the original RC frames of the test specimens are given in Fig. 4. The scale factor of the RC frames is about 1/4-1/3 of a low-rise school building designed according to the pre-1971 Building Standard Law of Japan (BSL 1971). As shown in Fig. 4, geometrical di-mensions and reinforcement patterns of all one-bay one-story frames are identical. The clear shear span-to-depth ratio M/(VD) of the RC columns is 2.50, and that of the RC beams is 2.65. Since the RC frames are designed according to the pre-1971 Building Standard Law of Japan (BSL, 1971), the RC columns contain insufficient amount of transverse reinforcements of pw = 0.12%, and consequently shear failure is likely to happen in the RC columns. In the one-bay two-story test specimen, the dimensions and reinforcements of the first story is the same as those in the one-bay one-story specimens. Ret-rofit details of the test specimens are illustrated in Fig. 5. Specific retrofit schemes of the test specimens (in Fig. 5) are explained as follows;

The specimen R05P-P0 is a non-retrofitted one-bay one-story RC frame. This specimen is the benchmark specimen that represents behavior of soft-story RC frames.

The specimen R09B-F is retrofitted by the steel frame without steel bracing members. This specimen is planned to verify structural performance of the RC frame and the steel frame in the absence of steel bracing members.

The specimen R09B-P is retrofitted by the steel braced frame with the help of the hybrid connection technique. The steel braced frame is connected to the RC frame at the boundary RC columns and at the top beam. This specimen is planned in such a way that the mechanism (a) (in Fig. 3) would appear in its experi-mental response.

The specimen R09B-CT is retrofitted by the steel braced frame with the help of the hybrid connection technique. Sandwiching steel plates at the top hybrid connection have the thickness of t = 1.2 mm that are thinner than those in other specimens. The specimen is planned in such a way that the mechanism (b) (in Fig. 3) would appear in its ex-perimental response. In some buildings, the presences of wing-walls beside RC columns make it impossible to use channel-shaped steel plates. Considering this fact, in the

specimen R09B-CT, the RC column and the steel column are sandwiched by two lateral plain steel plates instead of a channel shaped steel plate.

The specimen R10B-CP is retrofitted by the steel braced frame with the help of the hybrid connection technique. The anchor bolts of the base plates have a small diameter of D = 10 mm. Because of relatively low direct shear resistances of anchor bolts at the base, di-rect shear failure is likely to happen at that surface. This specimen is planned in such a way that the mechanism (c) (in Fig. 3) would appear in its experimental response.

The specimen R10B-CO is retrofitted by the steel braced frame with the help of the hybrid connection. This specimen is planned in such a way that the mechanism (d) (in Fig. 3) would appear in its experimental response. To provide a relatively high overturning moment at the base, a two-story frame is decided to be tested instead of a one-story frame. The anchor bolts at the base plates have the small diameter of D = 10 mm, and consequently the total tension resistance of anchor bolts against the overturning moment is relatively low in comparison with other retrofitted specimens.

1000175

1325

875

1000

250

350

3.7φ−@105

1-bay 1-story 1-bay 2-story

CL

sp =0.3%

125

250

3.7φ-@105

D6-@120

4-D13

Beam section

60

gp =1.63%

sp =0.3%

pw=0.43%

175

175

8-D10

3.7φ-@105

Column section

pg =1.85%

pw=0.12%

Fig. 4 Reinforcement details (unit: mm).


Channel-shaped steel plate (t )

Specimen R05P-P0

1.0 m

1.5 m

Specimen R09B-F Specimen R09B-PMechanism (a)

Specimen R09B-CTMechanism (b)

Specimen R10B-CPMechanism (c)

Specimen R10B-COMechanism (d)

1.5 m

1.0 m

1.0 ms

175 140

175 Grout

U-shaped reinf.(D6-@100)

175

65 175 140

Grout

U-shaped reinf.(D6-@100)

Column section Column section

High-strength bolt (13φ)


125Grout

250

140

Steel plate (t )s

Steel plate (t )s


U-shaped reinf. (D6-@100)

855

1305

Holes

Base plate190x175x19

DalaSteel braced frameBeam section

b

htf

Section of steel members

tw

ts

hes

les

Steel plate at the top connection

Specimen

R09B-FR09B-PR09B-CTR10B-CPR10B-CO

ts

3.23.23.23.23.2

ts

-4.51.23.23.2

les

-370370630630

la

182182182120120

Da

13φ13φ13φD10D10

h x b x t x t (brace) f w h x b x t x t (frame) f w_

75 x 75 x 4.5 x 4.575 x 75 x 4.5 x 4.575 x 75 x 6.0 x 6.075 x 75 x 6.0 x 6.0

75 x 75 x 4.5 x 4.575 x 75 x 4.5 x 4.575 x 75 x 4.5 x 4.575 x 75 x 4.5 x 4.575 x 75 x 4.5 x 4.5

Notes; t : thickness of steel plates at the boundary columns, t : thickness of steel plates at the top connection, l & h : effective length and height of steel plates at the top connections respectively, l : anchoring length of anchores, D : diameter of anchors, h x b x t x t : section sizes of steel members.

s s eses

a f wa

hes

-172172172172

Fig. 5 Test plans of the specimens.


3.4 Material properties Properties of concrete and grout materials on the day of the loading tests are given in Table 1. Properties of steel materials are listed in Table 2. 3.5 Test setups and loading programs Test setups and horizontal displacement-controlled pro-grams of the test specimens are given in Fig. 6. The vertical loads and horizontal displacements were simul-taneously applied on the specimens. Two servohydraulic actuators applied the constant vertical load of N = 0.2bDσB (where, bD: section dimensions of the RC col-umn and σB: cylinder strength of concrete) per column. A double-acting hydraulic oil jack pushed and pulled the test specimens under the displacement-controlled pro-grams. In the case of the one-bay two-story specimen, at each horizontal loading stage, 5/11 and 6/11 of the base shear force were applied on the first and second story respectively, in order to provide the lateral distribution of seismic forces according to the Building Standard

Law of Japan (BSL 1981). For the non-retrofitted specimen R05P-P0, the cyclic loading test was con-ducted at drift angles in the range of R = ±0.5%, ±1.0%, ±1.5%, ±2.0%, ±2.5%, and ±3.0% for two successive cycles, and ±0.125, ±0.25, ±4.0%, and ±5.0% for one cycle. In the RC frames retrofitted by the steel braced frames, to precisely observe structural behavior of the retrofitted frames, the horizontal cyclic displacement-controlled program was planned to be performed at drift angles with smaller increment compared to that of the non-retrofitted frame. The displacement-controlled pro-gram of the retrofitted frames is in the range of R = ± 0.1, ±0.2, ..., ±1.9%, ±2.0%, and ±2.5, ±3.0, ±4.0% and ±5.0% for one cycle. It should be emphasized that dis-placement-controlled programs of non-retrofitted frame and the retrofitted frames are different, however, this difference can not considerably influence the compari-son between structural behavior of the non-retrofitted frame and the retrofitted ones, especially from the view-point of maximum lateral resistance.

Table 1 Properties of concrete and grout materials.

Concrete Grout Specimen

σB (MPa) εc (%) Εc (GPa) σBg (MPa) εg (%) Εg (GPa) R05P-P0 28.3 0.22 27.2 - - - R09B-F 25.4 0.24 26.5 59.7 0.51 21.6 R09B-P 23.3 0.22 26.1 60.9 0.59 21.0 R09B-CT 19.7 0.18 24.7 57.2 0.49 21.5 R10B-CP 20.2 0.20 22.8 63.7 0.56 22.6 R10B-CO 25.2 0.20 28.0 63.2 0.57 22.2

Notes; σΒ: cylinder strength of concrete, εc: concrete strain at σΒ, Εc: Young’s modulus of concrete, σΒg: cylinder strength of grout, εg: grout strain at σΒg, and Eg: Young’s modulus of grout.

Table 2 Properties of steel materials.

Steel element Size a (mm2) σy (MPa) σu (MPa) εy (%) Es (GPa) D13** 127 341 524 0.17 200 D13* 127 403 597 0.18 227 D13 127 403 597 0.18 227 D10** 71 412 577 0.21 195 D10* 71 357 512 0.19 184

Longitudinal rebars and anchors

D10 71 371 524 0.19 195 D6** 32 393 488 0.22 176 D6* 32 468 540 0.25 191 D6 32 432 501 0.26 166 3.7φ** 11 643 688 0.32 199 3.7φ * 11 371 426 0.20 188

Hoops and stirrup

3.7φ 11 593 664 - 199 13φ * 133 1243 - 0.66 200 High-strength bolts 13φ 133 1243 - 0.66 200

t* = 1.2 mm - 281 351 0.14 208 t* = 3.2 mm - 306 418 0.15 203 t = 3.2 mm - 338 447 0.16 208 t* = 4.5 mm - 367 490 0.17 211 t = 4.5 mm - 340 465 0.17 198

Steel plates

t = 6.0 mm - 304 425 0.16 191 Notes: ** and * indicate the material belong to R05P and R09B series of the specimens respectively, non-starred materials be-long to R10B series, a: cross section area, σy: yield strength, σu: ultimate strength, εy: yield strain, Es: Young’s modulus of elasticity, and t: thickness of steel plate.


6

4

δ 7

1.0 m

5

4

3

2

8

1

7

1.0 m

5

3

6

2

δ

84

4

1

9

Details of the test setups: 1. Vertical reaction steel frame; 2. Servohydraulic actuator; 3. Strong floor; 4. Counter balance; 5. Horizontal reaction wall; 6. Double-acting hydraulic oil jack; 7. Test specimen; 8. Long high-strength bar (PC bar); 9. Lateral load distributor; δ: Horizontal displacement at the first story

0 4 8 12 16 20 24-6-4-20246

Number of cycle

For the retrofitted frames

Drif

t ang

le o

f the

firs

t sto

ry

R (%)

0 4 8 12 16-6-4-20246

Drif

t ang

le o

f the

firs

t sto

ry

For the non-retrofitted frame

R (%)

Number of cycle

Fig. 6 Test setups and displacement-controlled programs.


4. Experimental results and observations

The photos of the test specimens after finishing loading tests and removing the steel plates are given in Fig. 7. The crack patterns and the V-R relationships of the test specimens are shown in Fig. 8. The experimental obser-vations of the test specimens are explained one-by-one as follows;

In the specimen R05P-P0, flexural cracks appeared at the end of the columns, and at the end of the beam at the drift angle of about R = 0.5% and R = 1.0%, respectively. Shear cracks at the columns generated at about R = 1.5% and widened progressively with increasing drift angles. In the cycle with the target amplitude of R = -2.0%, the right-hand column collapsed suddenly in shear failure at the drift angle of R = -1.7%.

In the specimen R09B-F, flexural cracks generated at the bottom of the RC columns and at the both ends of the RC beam at the drift angles of R = 0.2% and R = 0.3%, respectively. At the drift angle of R = 1.4%, the lateral resistance reached to the maximum value of Vmax = +256kN which is about 2.4 times of that of the non-retrofitted specimen R05P-P0. The lateral force main-tained greater than 0.8Vmax up to the drift angle of R = 5.0%. It is worthy to note that the jacketing steel plates perfectly prevented the shear failure of the RC columns even in large lateral displacements.

In the specimen R09B-P, the RC frame is retrofitted by the steel braced frame with help of the proposed hy-brid connection. During the cyclic loading test, at the drift angle of R = 0.2% and R = 0.4%, flexural cracks generated at the base of the RC columns and at the ends of the RC beam, respectively. Buckling-shape deforma-tion initiated in the steel braces at the drift angle of R = 0.9%. Flexural cracks appeared at the top of the RC columns at R = 1.2%. The maximum lateral resistance of Vmax = +535kN that is about 5 times of that of the non-retrofitted specimen R05P-P0 obtained at the drift angle of R = 1.3%. By increasing the lateral deformation, significant buckling happened in the compression brace, and consequently the compression resistance of the brace gradually decreased. Regarding the discrepancy between the compression resistance and the tension re-sistance of the braces in the post inelastic-buckling stage, an unbalanced downward force produced at the braces joint. Finally, at the large drift angle of R = 5.0%, shear failure happened at the RC beam because of the afore-mentioned unbalanced downward force. The dominant mechanism of this specimen is the mechanism (a) illus-trated in Fig. 3.

In the specimen R09B-CT, the thickness of the steel plates at the top hybrid connection is ts = 1.2mm that is thinner than those in other specimens. Because of apply-ing relatively thin steel plates at the top hybrid connec-tion, direct shear failure was expected at that zone. In the cyclic loading test, horizontal cracks appeared at the top of the RC columns at the drift angle of R = 0.4%. The thin steel plates at the top connection started buck-

ling at R = 0.6%. The lateral resistance reached to the maximum value of Vmax = +473kN at R = 0.8%. In the V-R relationship, a sharp strength degradation can be observed because of direct shear failure at the top hy-brid connection and shear punching at the top of the boundary RC columns. The dominant mechanism of this specimen is the mechanism (b) illustrated in Fig. 3. The buckling shape of the steel plate after finishing the load-ing test is shown in Fig. 9.

In the specimen R10B-CP, the anchors (4-D10 at each base plate) are relatively weaker than those in the specimens R09B-P and R09B-CT (4-13φ high-strength bars). Because of utilizing weak anchorage system at the base of the steel braced frame, direct shear failure is likely to happen at that zone. In the cyclic loading test, at the drift angle of R = 0.1%, horizontal cracks ap-peared at the base of the RC columns. The anchors of the base plates yielded at the drift angle of R = 0.2% due to the provided shear force. At drift angle of R = 0.3%, the longitudinal reinforcements started yielding because of shear punching at the bottom of the RC columns. The lateral resistance reached to the maximum value of Vmax =+551kN at R = 0.9%. Afterwards, the lateral resistance sharply decreased, because the anchors at the base plates started breaking under shear sliding. The domi-nant mechanism of this specimen is the mechanism (c) illustrated in Fig. 3.

In the specimen R10B-CO, the anchors (4-D10 at each base plate) are relatively weaker than those in the specimens R09B-P and R09B-CT (4-13φ high-strength bars). The retrofit details of this specimen are the same as that in the specimen R10B-CP. However, since the specimen R10B-CO is a two-story frame, the height of the applied lateral force is higher, and consequently the provided overturning moment is greater. In the cyclic loading test, at the drift angle of R = 0.2%, the anchors of the base plates yielded in tension due to the provided overturning moment. At the drift angle of R = 0.3%, flexural cracks appeared at the base of the RC columns and longitudinal reinforcements started yielding. Fine shear cracks generated at the shear wall of the second story at R = 0.8%. The lateral resistance reached to the maximum value of Vmax = +476kN at R = 0.9%. Under overall-overturning behavior of the specimen, the an-chors and the longitudinal reinforcements at the bottom of RC columns started breaking at the drift angle of R = 1.3% and R = 1.9%, respectively. The dominant mecha-nism of this specimen is the mechanism (d) illustrated in Fig. 3.

From the backbone curves shown in Fig. 10 (a), it is obvious that both the lateral strength and stiffness of the retrofitted frames increased compared to those of the non-retrofitted frame R05P-P0. The lateral strength of the retrofitted specimen R09B-P in which the steel braces buckled increased to about 5 times of the lateral strength of the non-retrofitted specimen R05P-P0. Since, in the specimen R09B-P, plastic buckling happened in the stocky steel braces (the slenderness ratio of λ = 35),


Fig. 7 Photos of the specimens after finishing loading test and removing steel plates.

Specimen R09B-P

Plastic buckling

Flexural behavior* of RC frame and steel frame, tension yielding and buckling of steel braces

Mechanism (a) **

Specimen R09B-CT


Direct shear failure at the top hybrid connection and at the RC columns Mechanism (b) **

Specimen R10B-CP


Direct shear failure at the base Mechanism (c) **

Specimen R10B-CO

Overturning*

Overall-overturning Mechanism (d) **

* The pointed plastic hinge locations are under push loading from the left-hand side to the right-hand side. ** The associated mechanism is illustrated in Fig. 3.

Flexural plastic hinge*

Specimen R09B-F

Flexural behavior of RC frame and steel frames

Flexural plastic hinge*

Shear failure

Flexural behavior followed by shear failure

Specimen R05P-P0


Flexural plastic hinge

R

δV(+)

V(-)

Ν Ν

1.0 m

1.5 mFlexural plastic hinge

ShearfailureR

δV(+)

V(-)

Ν Ν

Push Pull

δV(+)

V(-)

R

ΝΝDirect shear failure ΝΝδ

V(+)

V(-)

R

δ

R

ΝΝ

5/11V(+)

5/11V(-)

6/11V(+)

6/11V(-)


Plasticbuckling

R

δV(+)

V(-)

Ν Ν

Direct shearfailure

Overturning

h

Specimen R05P-P0 Specimen R09B-F Specimen R09B-P

Specimen R10B-CO

Specimen R10B-CPSpecimen R09B-CT

Notes: Crack patterns shown cracks after finishing loading tests and removing steel plates, * The mechanisms shown on the crack patterns are under the push loading only,** Associated mechanisms are illustrated in Fig. 3, R = the drift angle of the first story, R = δ/h (in where δ = lateral displacement of the first story, and h = height of the first story), N = 0.2bDσB (where bD = sectional area of RC columns and σB = concrete cylinder strength).

-600-400-200

0200400600

-5 -4 -3 -2 -1 0 1 2 3 4 5

V (kN)

Vmax (+) =107 kN

R (%) = δ (cm)

Vmax (-) =-119 kN

Push

Pull

-600-400-200

0200400600

-5 -4 -3 -2 -1 0 1 2 3 4 5R (%) = δ (cm)

V (kN)0.8Vmax (+)Vmax (+) =

256 kN

Vmax (-) =-278 kN0.8Vmax (-)

-600-400-200

0200400600

-5 -4 -3 -2 -1 0 1 2 3 4 5R (%) = δ (cm)

V (kN) 0.8Vmax (+)

Vmax (+) =535 kN

Vmax (-) =-531 kN

0.8Vmax (-)

-600-400-200

0200400600

-5 -4 -3 -2 -1 0 1 2 3 4 5R (%) = δ (cm)

V (kN) 0.8Vmax (+)

Vmax (+) =551 kN

Vmax (-) =-522 kN

0.8Vmax (-)-600-400-200

0200400600

-5 -4 -3 -2 -1 0 1 2 3 4 5

0.8Vmax (+)

0.8Vmax (-) δ (cm)

V (kN)Vmax (+) =

473 kN

Vmax (-) =-466 kN

-600-400-200

0200400600

-5 -4 -3 -2 -1 0 1 2 3 4 5R (%) = δ (cm)

V (kN) 0.8Vmax (+)

Vmax (+) =476 kN

Vmax (-) =-455 kN

0.8Vmax (-)

Fig. 8 Crack patterns and hysteresis responses of the test specimens.

Flexural behavior followed by shear failure* Flexural behavior* Mechanism (a)**

Mechanism (b)** Mechanism (c)** Mechanism (d)**


moderate strength degradation was observed in the post-peak stage of the lateral resistance, especially up to the drift angle of R = 2.0%.

The accumulated absorbed energies of the specimens are given in Fig. 10 (b). The accumulated absorbed energy is derived from the summation of the amount of the energy dissipated in each cycle of the loading. The amount of the energy dissipated in the specimen R09B-P in which buck-ling happened in the steel braces is the highest. It should be noted the displacement-controlled program of the non-retrofitted specimen R05P-P0 is different with that of the retrofitted specimens (in Fig. 6). Therefore, the difference in the loading programs causes a slight discrepancy in the comparison of the accumulated absorbed energies.

5. Ductility index

Application of the proposed hybrid connection provides superior structural performance for retrofitted RC col-umns especially in view point of ductility. Ductility of the specimens is quantified by an indicator called the “ductility index”. The specimens are categorized as two groups according to their failure types. One group represents the specimens with flexural failure in col-umns, and its specified ductility index (F1) is given in Eq. (1) (JBDPA 2001). Another group represents the specimens with brittle failures such as punching and/or shear failure in boundary RC columns, and its corre-sponding ductility index (F2) is given in Eq. (2) (Yama-moto 1985, and Hattori 2004).

250

250

1

1.0 0.27 in case

2 / 13.2 in case

0.75(1 0.05 / )

mumu y

y

mu ymu y

mu y

R RR R

R RF

R RR R

R R

−⎧ + <⎪ −⎪= ⎨ −⎪ ≤ ≥⎪ +⎩

(1)

2 0.6 100 3.2suF R= + ≤ (2)

where Rmu: the ultimate inter-story drift angle (%) at which lateral resistance decreases to 80% of maximum lateral resistance, Rsu: the ultimate inter-story drift angle (%) at which lateral resistance reaches to the maximum lateral resistance, Ry: the yield deformation in terms of inter-story drift angle (%), which principally shall be taken as Ry=1/150 (JBDPA 2001).

The ductility indices of the specimens are given in Table 3. The comparison between propositionally-detailed specimens R09B-P and conventionally-detailed specimen R08B-F75 with the same steel braces sizes is more considerable. This comparison expresses that even though the sizes of installed steel braced frames ( 75 75 4.5 4.5× × × ) in both specimens are identical, the ductility index (F1=3.1) of the specimen R09B-P with the hybrid connection is higher than the ductility index (F2=1.4) of the specimen R08B-F75 with the conven-tional connection. The comparison quantitatively de-scribes the role of the hybrid connection in improving the ductility of retrofitted RC columns. It should be noted that among different specimens retrofitted by steel braced frames utilizing the hybrid connection, the specimen R09B-P in which the steel braces buckled in compression and yielded in tension (the mechanism (a) in Fig. 3) has the most desired seismic performance with the high ductility index of F1=3.1. The comparison between the conventionally-detailed specimens R08B-F75 and the propositionally-detailed specimen R09B-P clarifies two superior structural performances of the proposed connection to obtain high lateral capacity for retrofitted frame, and to increase shear strength of retro-fitted RC columns with the help of steel plates. To obvi-ously show the variation of the ductility index in the test specimens, the ductility indices versus the ultimate de-formations are plotted in Fig. 11.

hes=172mm

les=370mm

ts=1.2mm

Fig. 9 Buckling of steel plate in R09B-CT.

0

100

200

300

400

500

600

0 1 2 3 4 5Late

ral r

esis

tanc

e fo

rce,

V (k

N)

R09B-P

R09B-CT

R05P-P0

R09B-FR10B-CO

R10B-CP

δ (cm)=R(%)

(a) Backbone curves

0

40

80

120

160

0 0.5 1 1.5Acc

umul

ated

ene

rgy,

W (k

N.m

)

δ (cm)=R(%)

R09B-P

R09B-CT

R09B-F

R05P-P0

R10B-CO

R10B-CP

(b) Accumulated absorbed energies

-400

0

400

-1.5 0 1.5

V(kN)

R (%)

R05P-P0

R09B-P

Cycle R=1.5%

Fig. 10 Backbone curves and accumulated absorbed energies of the specimens.


6. Calculation approaches

Four fundamental mechanisms (in Fig. 3) are taken into consideration for a RC frame retrofitted by a steel braced frame with the help of the proposed hybrid con-nection. The specimens were planned and tested to ver-ify structural performance of each failure mechanism. In this section, the methods of calculating the lateral

strength of the mechanisms are explained, and simpli-fied design procedures are suggested. 6.1 Calculation method of the mechanism (a) The idealized deformation of the mechanism (a) is shown in Fig. 12. In the mechanism (a), the lateral strength Va is mainly governed by tension and compres-sion strengths of steel braces. The RC frame also par-ticipates in the global lateral strength. In the specimen R09B-P in which the mechanism (a) is the dominant mechanism, the strain gauges were attached on the steel frame at potential flexural plastic hinge regions. The gauges’ data did not show the formation of flexural plas-tic hinges in the steel frame at the ultimate lateral strength (R = 1.3%). Therefore, the participation of the steel frame in the global lateral strength of the retrofit-ted frame is conservatively ignored. The global lateral resistance Va of the mechanism (a) is given in Eq. 3.

( ) cosa y bu RCV F F Vθ= + + (3)

where Va: global lateral strength of the mechanism (a), Fy: yield strength of the steel brace, Fbu: buckling strength of the steel brace, θ: inclination of the brace toward the horizontal axis, and VRC: flexural lateral strength of the RC frame.

Table 3 Ductility index of the test specimens.

Specimen(1) Type of connection

Size of steel frame and braces Connection

Failure mechanism of left-hand / right-hand

columns

Ultimate drift angle(2)

(%)

Ductility index(3)

R05P-P0 - - - Flexural / Flexural Rmu =-2.0 F1 =2.6

R09B-F Hybrid 75×75×4.5×4.5 (No brace) No failure Flexural / Flexural Rmu =±5.0 F1 =3.2

R09B-P(4) Hybrid 75×75×4.5×4.5 No failure Flexural / Flexural Rmu =-3.0 F1 =3.1

R09B-CT Hybrid 75×75×4.5×4.5

Shear buckling failure of the steel

plate at the top connection

Punching at the top / Punching at the top Rsu =±0.8 F2 =1.4

R10B-CP Hybrid 75×75×6.0×6.0

Direct shear failure at the

anchors of the base plates

Punching at the base / Punching at the base Rsu =±0.9 F2 =1.5

R10B-CO Hybrid 75×75×6.0×6.0 Tension yielding of bottom anchors at the tension side

Overall flexural Rmu =+1.6 F1 =2.3

R08B-F60 Conventional 75×60×4.5×4.5

Direct shear failure at the top

connection after R=2%

Flexural / Flexural Rmu =-2.0 F1 =2.6

R08B-F75(4) Conventional 75×75×4.5×4.5 Direct shear failure

at the top connection

Punching at the top / shear at the top Rsu =±0.8 F2 = 1.4

Notes: (1) The experimental results of the test specimens are presented in Figs.1 & 8; (2) The ultimate inter-story drift angle (%) (see Eqs (1) & (2)), and (+) and (-) mean push and pull directions respectively; (3) Ductility index is calculated by Eq.(1) or Eq.(2); (4) The comparison between the two specimens R09B-P and R08B-F75 expresses that even though the sizes of the in-stalled steel braced frames (75×75×4.5×4.5) in both specimens are identical, the ductility index (F1=3.1) of the specimen R09B-P with the hybrid connection is higher than the ductility index (F2=1.4) of the specimen R08B-F75 with the conventional con-nection.

0.5

1

1.5

2

2.5

3

3.5

0 1 2 3 4 5

Duc

tility

inde

x, F

Ultimate drift angle, Rmu

or Rsu

(%) (see Eqs. 1 & 2)

Eq. 2

Eq. 1

R09B-FR09B-PR05P-P0 & R08B-F60

R10B-CO

R08B-F75 & R10B-CT

R10B-CP

Fig. 11 Ductility index of the specimens.


6.2 Calculation method of the mechanism (b) In the mechanism (b) shown in Fig. 13, direct shear failure happens at the top of the steel braced frame and at the top of the boundary RC columns. As given in Eq. 4, the direct shear strength Vb consists of the direct shear strength of the hybrid connection Qhyb, and the shear punching strength of the two boundary RC columns 2 puQ× .

2b hyb puV Q Q= + × (4)

6.2.1 Direct shear strength of the hybrid con-nection Qhyb: Horizontal shear force between the top RC beam and the top steel beam is transferred by the hybrid connec-tion. The hybrid connection consists of different con-tributing elements, namely steel plates, high-strength bolts, and infilling grout. So, the direct shear strength of the hybrid connection depends on the minimum value of the strengths of the contributing elements that is formu-lized in Eq. 5.

{ }2 min , , ,hyb ps bs yb bgQ Q Q Q Q= × (5)

where Qps: shear strength of steel plates, Qbs: bearing strength of steel plates’ holes, Qyb: shear yield strength of bolts, and Qbg: bearing strength of grout. In the shear transfer mechanism, the sandwiching steel plates deliver the provided horizontal shear force between the RC beam and the steel beam. The shear strength of steel plates Qps is given in Eq. 6. Regarding the geometry of the sandwiching steel plates, the steel plate may yield or buckle under the produced shear force. The shear yield strength of steel plates Qys is given in Eq. 7, and the

elastic buckling strength of steel plates Qcrs is written in Eq. 8. It was assumed that the steel plates are under a pure shear state. The shear buckling strength of steel plates is calculated based on the formulation by Ti-moshenko (1994). As shown in Fig. 14, the effective area of the sandwiching steel plates is limited to the positions of the bolts. It is assumed that bolts provide a simple support condition at the boundary of the effec-tive area of the steel plates. Influences of the geometry and the boundary condition of the steel plates are taken into consideration in the buckling coefficient of ks. The buckling coefficient ks presented in Eq. 9 was suggested by Galambos (1988) for a rectangular steel plate which is simply supported on its four edges.

{ }min ,ps ys crsQ Q Q= (6)

13ys s es ysQ t l σ= (7)

2

2 212(1 )( )

s s escrs s

es

s

E t lQ k

ht

π

ν=

− (8)

2

4.005.34( )

ses

es

klh

= + (9)

where Qps: shear strength of the steel plate, Qys: shear yield strength of the steel plat, Qcrs: shear elastic buck-ling strength of the steel plate, ts: thickness of the steel

N NDirect shear failure

RC frameSteel frameθ

Qpu

tF cF

Qpu

Vb

Qhyb

Hybridconnection

Fig. 13 Deformation of the mechanism (b).

RC beam

Steel plate

High-strength bolt Grout Steel beam

Effective zone Boundaryles

hes

Steel plate Fig. 14 Hybrid connection at the top.

Flexural plastic hingeN N

Yielding

BucklingRC frame

Steel frame

F Fbuy

θ

Va

Hybridconnection

Fig. 12 Deformation of the mechanism (a).


plat, les: effective length of the steel plate (in Fig. 14), σys: tensile yield strength of the steel plate, ks: buckling coefficient, Es: Young’s modulus of the steel plate, ν: Poisson’s ratio, and hes: effective height of the steel plate (in Fig. 14).

Bolts of the hybrid connection transfer direct shear force from the top RC beam and from the steel beam to the sandwiching steel plates. Direct shear yielding of the bolts is likely to happen under transferring shear force. The direct shear yielding of the bolts can be estimated through Eq. 10 (JBDPA 2001). Another important mechanism in transferring the direct shear force is the fracture of the steel plate around the holes. As given in Eq. 11, the bearing capacity of the steel plates’ holes is calculated according to the provisions by AISC-LRFD (1994) which considers both the tear fracture of steel plate’s material and the deformation of the steel plate around the bolts. Moreover, the grout embedding bolts in the hybrid connection zone should have sufficient strength and rigidity to strongly lock the bolts. On the other hand, the split of the surrounding grout under bearing stress of the bolts should be prevented. Conse-quently, the bearing strength of grout Qbg should be checked through Eq. 12 (JBDPA 2001).

0.7yb b b ybQ n a σ=

(10)

2.4bs b s a usQ n t d σ= (11)

{ }0.3 245 ( N, mm)bg g Bg b b b bQ E n a n a inσ= ≤ (12)

where Qyb: shear yield strength of bolts, nb: number of bolts, ab: cross-section area of a bolt, σyb: tension yield strength of bolts, Qbs: bearing strength of steel plates’ holes, σus: ultimate tension strength of steel plates, ts: thickness of the steel plate, da: diameter of a bolt, Eg: Young’s modulus of grout, and σBg: cylinder strength of grout. 6.2.2 Shear punching resistance of a RC col-umn Qpu: The shear punching (direct shear) resistance of RC col-umns is calculated based on the method presented in the provisions by the Japan Building Disaster Prevention Association (JBDPA 2001). In calculation of the direct shear resistance Qpu, the influence of reinforcements cross the failure plane and external stresses normally act on the failure plane are considered. The calculation value of the direct shear strength of RC columns is given in Eq. 13. The factor of kmin in Eqs 13 & 14 repre-sents the minimum estimation of the direct shear strength. In Eq. 14, a is the shear punching span be-tween the direction of the applied shear force and the potential shear punching plane. According to the geome-try of the proposed retrofit scheme, the shear punching span at the top of the retrofitted RC columns is consid-ered as a = 10 mm.

min 0puQ k b Dτ= (13)

min 0.34 /(0.52 / )k a D= + (14)

0 0.98 0.1 0.85 0 0.33 2.750.22 0.49 0.33 2.75 0.660.66 0.66

B B

B B B

B B

τ σ σ σ σσ σ σ σ σσ σ σ

= + + ≤ ≤ −= + − < <= >

(15)

/( )g yp N bDσ σ= + (16)

where kmin: the minimum value of the punching reduc-tion factor, τ0: shear stress at the punching plane, b: the effective width of the column resisting against the direct shear force considering the connected members in the orthogonal direction, D: the depth of the column, a: the shear punching span, σ: normal stress act on the shear punching plane, σB: cylinder strength of concrete, pg: ratio of longitudinal reinforcements cross the shear punching plane, σy: yield stress of longitudinal rein-forcements, and N: axial force. 6.3 Calculation method of the mechanism (c) As shown in Fig. 15, in the mechanism (c), the direct shear failure happens at the base of the retrofitted frame. The direct shear strength at the base of a retrofitted frame is given in Eq. 17. Strengths of the boundary RC columns Qpu and the anchors na Qa (where na is the number of the anchors and Qa is the direct shear resis-tance of an anchor) provide the required direct shear strength at the base.

(1) (2)c pu pu a aV Q Q n Q= + + (17)

Calculation of the direct shear strength Qpu of RC columns is the same as that explained in Section 6.2.2. Direct shear strength of an anchor is given in Eq. 18 according to provisions by JBDPA (2001). The direct shear strength of an anchor Qa depends on its yield shear strength Qa1 (in Eq. 19) and the bearing strength of surrounding concrete Qa2 (in Eq. 20)

{ }1 2min , / 294MPaa a a a aQ Q Q Q a= ≤ (18)

1 0.7a ya aQ aσ= (19)

2 0.4a c B aQ E aσ= (20)

N NV

RC frame

tF cF

Steel frameDirect shear failure

QaQaQpu(1) Qpu(2)

θ

Hybridconnection

c

Fig. 15 Deformation of the mechanism (c).


where σya: tension yield strength of anchors, aa: section area of an anchor, Ec: Young’ modulus of concrete, σB: cylinder strength of concrete. 6.4 Calculation method of the mechanism (d) As shown in Fig. 16, in the mechanism (d), overall-overturning behavior appears in the retrofitted frame. When the retrofitted frame is pushed from the left-hand side to the right-hand side, the left-hand RC column and the anchors yield in tension, and the compression zone falls in the right-hand RC column and at the base of the com-pression brace. By taking moment about point O (shown in Fig. 16), the lateral strength Vd of the mechanism (d) can be obtained that is written in Eq. 21. Based on the provi-sions by JBDPA (2001), the tension strength of an anchor Ta depends on tension yield strength of the anchor Ta1, cone failure strength of the surrounding concrete Ta2, and bond failure strength between the anchor and the surround-ing concrete Ta3 which are given in Eqs 23 -25, respec-tively.

1( sin ) /d g y a a a c bV a L NL n T L F L hσ θ= + + − (21)

1 2 3min{ , , }a a a aT T T T= (22)

1a ya aT aσ= (23)

2 0.23 ( N,mm)a B cT A inσ= (24)

3 ( ) ( N, mm)a a a a aT D l D inπτ= − (25)

10 / 21a Bτ σ= (26)

where Vd: lateral strength of the mechanism (d), ag: sec-tion area of all longitudinal reinforcements of the ten-sion-side column, σy: tension yield strength of longitu-dinal reinforcements, na1: number of anchors at the ten-sion-side base plate, σya: tension yield strength of an anchor, aa: section area of an anchor, σB: concrete strength, Ac: projected area of concrete cone failure sur-face of a single anchor (refer to JBDPA (2001)), τa: bond strength of an anchor against pull-out force (MPa), Da:

diameter of an anchor, la: embedment length of an an-chor, and other geometrical parameters are shown in Fig. 16. 7. Comparisons between the experimental and calculated results

The experimental results and the calculated lateral strengths of the specimens are given in Fig. 17. The calculated lateral strengths are based on the approaches presented in Section 6.

In the non-retrofitted specimen R05P-P0, flexural be-havior of the RC frame is the dominant mechanism. However, shear failure finally happened in the RC col-umns. As shown in Fig. 17, from the calculated lateral strengths, the dominant behavior of the frame can be recognized.

In the specimen R09B-F, flexural behavior of the RC frame and the steel frame is the dominant mechanism. The calculated lateral strength of the mechanism (a) (flexural behavior of the RC frame and steel frame) is near the experimentally-obtained lateral strength. Ac-cording to experimental results, it can be concluded that the jacketing steel plates effectively increased the shear resistance of the RC columns. It should be noted that in the non-retrofitted specimen R05P-P0 shear failure hap-pened in the RC columns, but in this specimen the shear failure was perfectly prevented because of utilizing jacketing steel plates in the boundary columns.

In the specimen R09B-P, flexural behavior of the RC frame, tension yielding and buckling of the steel braces appeared in the experimental observations. As it is ex-pected, the experimentally-obtained lateral strength of the specimen is near the calculated lateral strength of the mechanism (a).

In the specimen R09B-CT, since the thickness of the steel plates at the top hybrid connection is relatively thin (t = 1.2mm), direct shear failure happened at the top of the steel braced frame. The dominant mechanism of this specimen is the mechanism (b). As shown in Fig. 17, the calculated lateral strength of the mechanism (b) is close to the experimentally-obtained lateral strength.

In the specimen R10B-CP, because the anchor bolts of the base plates are relatively weak in direct shear resistance, direct shear failure happened at the base of the steel braced frame and at the bottom of the RC col-umns. As it is expected, the calculated lateral strength of the mechanism (c) is near the experimentally-obtained lateral strength.

In the specimen R10B-CO, since the anchor bolts of the base plates are relatively weak in tension resistance, the overall-overturning behavior appeared in the speci-men. The dominant mechanism of this specimen is the mechanism (d). Because the overturning behavior is mainly generated by vertical slip of the longitudinal reinforcements at the base of the RC column in the ten-sion side, the longitudinal reinforcements experienced high concentrated strain at that zone. Considering this

N N

htF cF

Vd

RC frame Steelframe

θ

Overturning

Oag σy na1Τa La

L

Fc sin θ

Lb

la

Hybridconnection

Fig. 16 Deformation of the mechanism (d).


fact, the effect of strain-hardening of longitudinal rein-forcements was considered in calculating the lateral strength of the mechanism (d). As shown in Fig. 17, the calculated lateral strength of the mechanism (d) and the experimentally-obtained lateral strength agree well.

8. Design procedure

Based on the calculation approaches of the fundamental failure mechanisms discussed in Section 6, the design flowchart to obtain the mechanism (a) as the dominant mechanism is presented in Fig. 18. According to ex-perimental results of four fundamental failure mecha-nisms, the mechanism (a) is the most suitable one. In the retrofitted frame R09B-P in which the mechanism (a) was the dominant mechanism, not only the lateral strength and stiffness of the retrofitted frame increased, but also the ductility and energy absorption improved. 9. Summary and conclusions

In this paper, structural behaviors of the RC frames ret-rofitted by the steel braced frames are verified experi-

mentally. The steel braced frames were installed inside the RC frames by a new connection method called the “hybrid connection technique”. Experimental investiga-tions were performed to obtain four fundamental failure mechanisms of retrofitted RC frames. In addition to the experimental investigations, associated calculation pro-cedures to estimate lateral strength of the fundamental failure mechanisms are suggested based on the provi-sions commonly used. The calculated lateral strengths by the proposed calculation procedures agree with the experimentally-obtained results. According to experi-mental observations and results, following conclusions are briefly expressed; 1) The hybrid connection can effectively transfer a

relatively high direct shear force between an exist-ing RC frame and an installed steel braced frame.

2) The hybrid connection not only plays a role as a con-nection between a RC frame and a steel frame, but also increases shear resistance and axial compression capacity of boundary RC columns.

3) The behavior of the specimen R09B-P in which the steel braces yielded in tension and buckled in com-pression (the mechanism (a)) shows superior struc-

200

400

600

800

0

Mechanism (a) Mechanism (b) Mechanism (c) Mechanism (d)


Buckling



Overturning

Experimental lateral strength

Calculated lateral strength of RC frame

Calculated lateral strength of steel braced frame

(a): Lateral strength of the mechanism (a), Eq. (3)

(b): Lateral strength of the mechanism (b), Eq. (4)

(c): Lateral strength of the mechanism (c), Eq. (17)

(d): Lateral strength of the mechanism (d), Eq. (21): Dominant mechanism of the specimens

Ex.

R09B-F

(a)

(c)

(d)

Ex.

(b)

R09B-P

(a) (a)

(c)

(d)

Ex.(b)

R09B-CT

(a)(c)

(d)Ex.

(b)

R10B-CP

(b)

(a)(c)

(d)Ex.

R10B-CO

Ex.

R05P-P0Non-retrofitted Retrofitted specimens

(kN)

RCframe

Steelframe Hybrid

connection

Fig. 17 Experimental results and calculated lateral strengths.


tural performance by increasing the lateral strength and stiffness, as well as improving the ductility. So, the retrofit scheme of the specimen R09B-P with the dominant mechanism (a) can be suggested as the desired retrofit scheme for practical applications.

Acknowledgements The authors would like to express their deepest gratitude to Prof. Chiaki Matsui, Professor Emeritus at Kyushu University for his valuable advice. The authors are in-debted to Dr Kozo Nakada, Associate Professor at Uni-versity of the Ryukyus, for his valuable cooperation. The investigation reported herein was carried out possi-

ble by the grants supported in 2009 and 2010 from the Ministry of Land, Infrastructure, Transport, and Tourism of Japan. Also, this research was partially supported by the Grant-in-aid for the Scientific Research (A) in 2008 from Japan Society for the Promotion of Science (JSPS). References ACI Committee 318, (2008). “Building code

requirement for structural concrete and commentary.” American Concrete Institute, Michigan DC.

AISC-LRFD, (1994). “Manual of steel construction-load and resistance factor design.” American Institute

The mechanism (a) is obtained as the dominant mechanism

Start

Determining the required lateral strength Q for a soft-story frame

Calculation of the lateral strength ofthe mechanism (a) Va from Eq. 3

Estimation of details of the top connectionincluding; number and diameter of bolts, geometry

of steel plates, and mortar strength

Calculate the lateral strength of the mechanism (b) from Eq. 4

Calculation of the possible direct shear resistance ofsteel plates Qps from Eqs 6 ~ 9, bolts Qyb from Eq.10, steel plates' holes Qbs from Eq. 11 and mortar

Qbg from Eq. 12

Calculate the lateral strength of the top connectionfrom Eq. 5

Is Vb > Va ? (Is the mechanism (a) dominant?)

Step (6): Calculation of the punching strength of RCcolumns Qpu in Eq. 13 from JBDPA (2001)

Is Va > Q ?

Step (3): Estimation of the size of steel bracesaccording to required lateral strength Q

NoYes

Step (12): Estimate the diameterof anchors at the bottom

Design related tothe mechanism (a):

Control related tothe mechanism (b):

Calculate the lateral strength of the mechanism (c) by Eq. 17

Calculate the direct shear strength of anchors by Eqs 18 ~ 20

Is Vc > Va ? (Is the mechanism (a) dominant?)

Control related tothe mechanism (c):

No

Step (16): Estimation of the anchorage length ofanchors

Control related tothe mechanism (d):

Calculation of tension strength of anchors, andfinding out the dominant tension failure type

from Eqs 22 ~ 26

Increase the diameter and /or anchorage lengthof anchors

Calculate the lateral strength of the mechanism (d) by Eq. 21

Is Vd > Va ? (Is the mechanism (a) dominant?)

Select the mechanism (a) as the dominant mechanism

End

*

*If instead of the mechanism (a) another mechanism is selected as the dominant mechanism, the sequence of the flowchart should be changed. For example, since in this flowchart the dominant mechanism is the mechanism (a), the design procedure is started with the procedure related to the mechanism (a), and other mechanisms are controlled.

No

NoYes

No

NoYes

Yes

No

Yes

Fig. 18 The design flowchart to obtain the mechanism (a) as the dominant mechanism.


of Steel Construction Inc., Chicago, USA. BSL, (1971). “Building standard law of Japan.”

Building Center of Japan, Japan. (in Japanese) BSL, (1981). “Building standard law of Japan.”

Building Center of Japan, Japan. (in Japanese) FEMA-547, (2006). “Techniques for seismic

rehabilitation of existing buildings.” Federal Emer-gency Management Agency, Washington D.C., USA.

Galambos, T. V., (1988). “Guide to stability design criteria for metal structures.” 4th Edition, John Wiley and Sons, Inc.

Hattori, Y., Yamamoto, Y. and Koh, T. (2004). “ Seismic strengthening method with exterior concrete member included steel plate.” 13th World Conference on Earthquake Engineering, Vancouver, Canada, Paper No. 741.

Higashi, Y., Endo, T., Ohkubo, M. and Shimuzu, T., (1980). “Experimental study on strengthening rein-forced concrete structures by adding shear wall.” 7th World Conference on Earthquake Engineering, Istan-bul, Turkey, VII, 173-180.

JBDPA, (2001). “Standard, guideline and technical manual for seismic rehabilitation and retrofit of existing RC buildings.” The Japan Building Disaster Prevention Association, Japan.

Jones, E. A. and Jirsa, J. O., (1986). “Seismic strengthening of a reinforced concrete frame using structural steel bracing.” PMFSEL Report, University of Texas at Austin, No. 86-5.

Kawamata, S. and Ohnuma, M., (1980). “Strengthening effect of eccentric steel braces to existing reinforced concrete frames.” 7th World Conference on Earth-quake Engineering, Istanbul, Turkey, IV, 513-520.

Maeda, K. and Yamakawa, T., (2012). “Comparison between seismic performance of hybrid and conventional connections for retrofitting RC frames by steel braced frames.” Journal of Structural and Construction Engineering (AIJ), 678, 1299-1307.

Ohishi, H., Takahashi, M. and Yamazaki, Y., (1988). “A seismic strengthening design and practice of an existing reinforced concrete school building in Shizuku city.” 9th World Conference on Earthquake Engineering, Tokyo-Kyoto, Japan, VII, 415-420.

Otani, S., (2004). “Earthquake resistant design of reinforced concrete buildings - Past and future.”

Journal of Advanced Concrete Technology, 2(1), 3-24. Rahman, M. N. and Yamakawa, T., (2007).

“Investigation of hybrid technique for seismic retrofitting of bare frames.” Journal of Advanced Concrete Technology, 5(2), 209-222.

Sekiguchi, I., Okada, T., Murakami, M., Kumazawa, F., Horie, F. and Seki, M., (1988). “Seismic strengthening of an existing steel reinforced concrete city office building in Shizuka prefecture.” 9th World Conference on Earthquake Engineering, Tokyo-Kyoto, Japan, 439-444.

Sugano, S. and Fujimura, M., (1980). “Aseismic strengthening of existing reinforced concrete build-ings.” 7th World Conference on Earthquake Engineer-ing, Turkey, IV, 449-456.

Timoshenko, S., (1994). “Guide to stability design criteria for metal structures.” Two Parts, 2nd Edition, D. Van Nostrand Company, Inc.

Yamakawa, T., Rahman, M. N. and Morishita, Y., (2006a). “Experimental investigation and analytical approach for seismic retrofit of RC columns with wing-wall.” Journal of Structural and Construction Engineering (AIJ), 608, 109-117.

Yamakawa, T., Rahman, M. N., Nakada, K. and Morishita, Y., (2006b). “Experimental and analytical investigation of seismic retrofit technique for a bare frame utilizing thick hybrid walls.” Journal of Structural and Construction Engineering (AIJ), 610, 131-138.

Yamakawa, T., Javadi, P. and Kobayashi, M., (2009). “Cyclic loading tests on retrofitted RC frames -A new hybrid connection for installation of a steel braced frame inside a RC frame (Part 1)-.” Journal of Structural and Construction Engineering (AIJ), 642, 1487-1494.

Yamakawa, T., Javadi, P., Maeda, K. and Kobayashi, M., (2010). “Capacity-based analytical evaluations of retrofitted RC frames -A new hybrid connection for installation of a steel braced frame inside a RC frame (Part 2)-.” Journal of Structural and Construction Engineering (AIJ), 651, 987-996.

Yamamoto, Y. and Aoyama, H., (1985). “Flexure failure tests of existing RC frames retrofitted by steel panel frames.” Proceedings of the Japan Concrete Institute, 457-460. (in Japanese)

retrofitting of rc frames by steel braced frames utilizing

Documents