当前位置:首页 >> 建筑/土木 >>

A study on the seismic performance of concrete-filled square steel tube column-to-beam connections r


Journal of Constructional Steel Research 66 (2010) 962–970

Contents lists available at ScienceDirect

Journal of Constructional Steel Research
journal homepage: www.

elsevier.com/locate/jcsr

A study on the seismic performance of concrete-filled square steel tube column-to-beam connections reinforced with asymmetric lower diaphragms
Sung-Mo Choi a, , Su-Hee Park a , Yeo-Sang Yun b , Jin-Ho Kim c
a b c

Department of Architectural Engineering, University of Seoul, Seoul, Republic of Korea Harmony Engineering, Seoul, Republic of Korea Research Institute of Industrial Science & Technology Steel Structure Research Laboratory, Kyungkido, Republic of Korea

article

info

abstract
This paper studies the development of the through-type concrete-filled square steel tube column-tobeam connection, reinforced with an asymmetric lower diaphragm. This type of connection can be used in weak-earthquake regions such as Korea. A simple tension test was performed on the suggested lower diaphragms and the combined cross diaphragm in order to confirm their tensile behavior. Subsequently, four types of concrete-filled square steel tube column-to-beam connection (with combined cross diaphragm, nothing, horizontal T-bar and vertical plate as lower diaphragm) were fabricated in actual size and tested according to the ANSI/AISC SSPEC 2002 cyclic loading program. The horizontal T-bar and stud bolts in the vertical plates were designed to transmit the tensile stress from the bottom flange of the beam to the filled concrete. All the test specimens satisfied the 0.01 radian inelastic rotation capacity requirements for the composite ordinary moment frame (C-OMF) of the AISC seismic provisions. It was concluded that the suggested simplified lower diaphragms have sufficient strength, stiffness and plastic deformation capacity to be used in the field. 2010 Elsevier Ltd. All rights reserved.

Article history: Received 2 December 2008 Accepted 17 January 2010 Keywords: Concrete-filled square steel tube Diaphragm Column-to-beam connection Cyclic loading program Simple tension test Seismic performance

1. Introduction 1.1. Research background and objectives The Concrete Filled Tube (CFT) structure is acknowledged within the industry as a very economic and effective structural system. The CFT structure has been studied by many researchers and has been widely used in the field [1–4]. The connection details for the CFT structures have been developed mainly to resist strong earthquakes. It would therefore be uneconomical to apply these details to structures in weak-earthquake regions such as Korea. For the CFT column-to-beam connections requiring field welding, the AISC Seismic Provisions [5] provide the through-type connection where the beam penetrates the column as the seismic connection detail of the composite special moment frame (C-SMF). This through-type connection is adequate for strong-earthquake regions, such as America and Japan. However, it is uneconomic to apply the through-type connection to structures in weakearthquake regions, such as Korea. In addition, tall buildings in Korea usually adopt a reinforced concrete (RC) core and a moment frame as the lateral load resisting system, such as in the building



Corresponding author. Tel.: +82 2 2210 2396; fax: +82 2 2248 0382. E-mail address: smc@uos.ac.kr (S.-M. Choi).

in Fig. 1. The RC core inside the building is usually designed to resist 80%–90% of the lateral load, while the exterior moment frame is required to resist only a small portion of the lateral load. In this study, the authors suggest the use of asymmetric diaphragms at the bottom of the through-type connections that can obtain an inelastic rotation capacity of 0.01 rad. This would comply with the requirements for the composite ordinary moment frame suggested by the AISC Seismic Provisions [5] and be applied in weak-earthquake regions such as Korea. This lower diaphragm is designed to resist 40% of the tensile force derived from the fullplastic moment of the beam. In existing CFT column-to-beam connections that have different depths of beams at both sides, the lower diaphragm in the connection becomes inclined or two-layered, as shown in Fig. 2. This makes the construction work complicated and deteriorates the concrete-filling condition in the steel tube. Thus, it is necessary to develop a connection detail that can compensate for these weak points. However, a guide and manual on the details for various column-to-beam connections does not exist, and neither the Korean Society of Steel Construction nor the Architectural Institute of Korea has published such material [6]. The objective of this study is to present the base data on CFT column-to-beam connection details that could obtain the performance of fabrication and construction work and exhibit a sufficient ductile behavior. This data is presented as part of the development of a CFT column-

0143-974X/$ – see front matter 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.jcsr.2010.01.004

S.-M. Choi et al. / Journal of Constructional Steel Research 66 (2010) 962–970

963

to-beam connection to be used in weak-earthquake regions such as Korea. 1.2. Test plan As a base study to detect the behavior of a CFT square columnto-beam connection, a simple tension test was carried out on three types of lower reinforcing. The specimens for the simple tension test were idealized with only a tension flange of the beam in the CFT column-to-beam connection. The applicability of the suggested connections was checked based on these simple tension test results. In addition, four types of CFT columnto-beam connection were fabricated in actual size and tested according to the ANSI (American National Standard Institute)/AISC SSPEC-2002 cyclic loading program [5] in order to discover their structural behavior through the seismic performance evaluation. A parametric study was also carried out so as to understand their behavior. 1.3. Suggested connection details Two connection details of the HT and VP type are suggested in this study. These connection details are developed so that they can be applied in cases where the compressive load interacts only as a part of the development of a CFT column-to-beam connection. Furthermore, they are developed so that they can be used in weakearthquake regions such as Korea [7]. The bottom of the connection details is reinforced with a horizontal T-bar (HT) and vertical plate (VP), in order to improve concrete filling and construction work. Fig. 3 illustrates these two connection details. In the HT type, when compression develops on the bottom flange of the beam, the stress is transferred to the filled concrete of the CFT column, through the vertical plate welded to the end of the trapezoidal horizontal plate. If tension develops on the bottom flange of the beam, tension is transferred by the anchor action of the T-bar in the filled concrete. In the VP type connection detail, the compression of the bottom flange of the beam is transferred to the filled concrete in the CFT column through the stud bolts, which are welded to the vertical plate at regular intervals and confined by the filled concrete. A sleeve-inserted combined cross diaphragm (CDS) is also presented for comparison with the HT and the VP type connection details. The CDS is an existing connection type that can be used in strong-earthquake regions. The detail of the CDS is designed to transfer stress more clearly and obtain a sufficient ductile behavior. The research team in this study has achieved finite element analysis for the connection details of the CDS type from previous research, and has evaluated the load-carrying capacities, load transfer mechanism and stress concentration associated with the CDS type [8,7]. The research team has also suggested the evaluation equation of load-carrying capacities for this connection [9]. The unbalance of load on the beams at both sides of the connection is also considered in this research. More specifically, the diaphragm in the strong axis penetrates the column and is connected to the beam directly, while the diaphragm in the weak axis is inserted into the column. 2. Simple tension test 2.1. Specimen design Three specimens of CDS, HT and VP were fabricated in order to discover the behavior of the suggested lower reinforcing types under tension, and the applicability to the CFT column-to-beam connection. These specimens are illustrated in Figs. 4–6. The CDS specimen had a sleeve-inserted combined cross diaphragm. It was used for comparison with the other types. In the HT specimen, a T-bar was welded to the steel tube horizontally inside the CFT column.
9 @ 8.4m = 75.6m 37@4.05m = 149.8m 2@14.5m + 13m = 42m 2@14.5m + 13m = 42m

(a) Plan (Typ.).

(b) Elevation. Fig. 1. R4 building.

Fig. 2. Diaphragm in connection with different depth of beams at both sides.

Strong Axis Combined cross Diaphragm

Weak Axis

Vertical Flat bar

Horizontal T- bar
Fig. 3. Supposed connection detail.

The T-bar was made by welds of 6 mm and 12 mm-thick plates and had a 25 mm diameter hole for concrete filling. The tension of the beam flange is transferred to the filled concrete in the CFT column through the horizontal T-bar. In the VP specimen, a 9 mm-thick reinforcing plate was welded to the steel tube vertically and 4 stud bolts of 13 mm in diameter and 60 mm in length were welded to this vertical plate at intervals of 100 mm. 2.2. Loading and measurement method A pullout test was carried out on the beam flanges of the specimens by the universal testing machine with a 2940 kN capacity. This test was continued until the ultimate capacities and failure modes were confirmed. The axial displacement of the beam flange was measured by the displacement gauges attached to the beam flange, which were in the range of 600 mm gauge length, starting at a 100 mm distance from the column face. Strain gauges

5@4.6m = 23m

964

S.-M. Choi et al. / Journal of Constructional Steel Research 66 (2010) 962–970

1400 1200 1000 Load (kN) 800 600 400 200 0 0
Failure at welded part of vicinity of Steel tube (441kN, 25mm)

CDS

Failure at Diaphragm in major axis (1284kN, 24mm) Failure at welded part of Steel tube (597.8kN, 23mm)

HT

Pp

VP 30 35 40

5

10

15

20

25

Displacement (mm)
Fig. 7. Load–displacement curve.

were also attached to the flanges of the beam and column in order to detect the strain distribution.
Fig. 4. Sleeve-inserted combined cross diaphragm (CDS) type.

2.3. Material test A material test was carried out in order to confirm the mechanical properties of the steel that was used in the simple tension test specimen. The test was carried out according to KS B 0802, Method of Tensile Test for Metallic Material in Korean Standards, and the results are given in Table 1. Welds of 12 mmthick SM490 steel plates were used to fabricate the columns of each specimen, while 16 mm-thick plates of SS400 steel were used for the diaphragms and beam flanges. The compressive test was also carried out on the concrete that was used to fill the steel tube, in order to detect its 28-day compressive strength. 2.4. Load test results Fig. 7 illustrates the relationship between the load and displacement of three specimens according to the simple tension test results. In Fig. 7, Pp is the full plastic moment of the beam flange, which is a product of Fy and sectional area. Here, the value of Fy is given in the material test results of Table 1. According to the simple tension test, the capacity of the CDS specimen was governed by the tensile strength of the diaphragm in the strong axis (the diaphragm in the strong axis and the beam flange are made of the same steel). In the case of the HT and VP specimens, their capacities were governed by a failure of the connection. 2.5. Ultimate capacity and initial stiffness The ultimate capacity and initial stiffness of specimens are compared in Fig. 7. The ultimate capacity of the HT specimen was 10% larger than that of the VP specimen. However, the ultimate capacities of the HT and VP specimens were only about 50% of that of the CDS specimen. Similarly, the initial stiffness of the CDS specimen was 40%–45% higher than that of the HT and VP specimens. This might be caused by the fact that the drop-off in capacity of the HT and VP specimens occurred earlier than that of the CDS specimen, with their reinforcing diaphragms undergoing a plastic state (See Figs. 8 and 9). 3. Cyclic load test 3.1. Specimen design In all specimens, the top of the connection was designed to transfer the stress more clearly and have a large ductility. In addition, the unbalance of the load on the beams at both sides of the connection was taken into account. Specifically, the upper

Fig. 5. Horizontal T-bar (HT) type.

Fig. 6. Vertical Plate (VP) type.

S.-M. Choi et al. / Journal of Constructional Steel Research 66 (2010) 962–970 Table 1 Material test results on the simple test and the cyclic load test specimens. Member Steel tube Beam flange Diaphragm T-bar (horizontal) T-bar (vertical) Flat-bar Concrete Steel grade SM490 THK (mm) 12 10 16 6 9 12 E (GPa) 2108 2051 2050 2152 2106 2106 Fy (MPa) 462.6 340.1 342 231.3 330.3 262.6 Fu (MPa) 603.7 450.8 478.2 364.6 476.3 355.7 Fy /Fu 0.77 0.75 0.72 0.63 0.69 0.74

965

Elo. (%) 26 28 32 29 26 34

SS400

fck = 49 MPa

Fig. 8. Comparison of ultimate capacity.

Fig. 9. Comparison of initial stiffness.

diaphragm in the strong axis which is under the larger load, penetrated the column and was directly connected to the beam, while the upper diaphragm in the weak axis, which is under a smaller load, was inserted into the column. Four specimens of CDS, CDSN, CDST and CDSV were fabricated for the cyclic load test. The upper diaphragm of all specimens was a combined cross diaphragm and the four specimens were distinguished by the lower diaphragm. Figs. 10–13 show the detail of each specimen. In all specimens, the column was a square tubular section of -400 × 400 × 12, SM490 steel and the beam was a wide flange section of H-500 × 200 × 10 × 16, SS400 steel. The steel tubular column was filled with a high-flexible concrete with a 49 MPa compressive strength. In the top of the connection, a 200 mm-wide diaphragm penetrated into the column in the strong axis and welded with the diaphragm of the weak axis so as to form a combined cross diaphragm. A round sleeve of 114.3 mm in diameter, 6 mm in thickness and 130 mm in length, was inserted into the center of the combined cross diaphragm. The ratio of the opened area to the gross area of the diaphragm for this sleeve was 28.4%. A sleeve was inserted into the upper diaphragm so as to compensate for the reduction in the sectional area due to the hole in the center of the diaphragm and to the increase of the in-plane strength under tension. The CDS specimen included this combined cross diaphragm at both the top and the bottom of the connection [10]. The CDSN

specimen included this combined cross diaphragm only at the top of the connection. It was intended that the compression of the bottom flange of the beam be transferred to the filled concrete in the CFT column directly, without a diaphragm at the bottom. The CDST specimen had the T-shape of a lower diaphragm, which consisted of a vertical rectangular plate of 200 mm in width, 100 mm in height and 12 mm in thickness and a horizontal trapezoidal plate of 200 mm in lower side length, 100 mm in upper side length, 88 mm in height and 6 mm in thickness. The trapezoidal shape was used in the lower diaphragm in order to distribute the stress to the filled concrete under compression, and to increase the confinement effect by the concrete under tension. The CDSV specimen had a vertical plate of 572.5 mm in height, 100 mm in width and 9 mm in thickness. It also had 5 stud bolts of 13 mm in diameter and 60 mm in length as the lower diaphragm. Stud bolts were perpendicularly welded to the vertical diaphragm at intervals of 60 mm. The lower diaphragm of the CDST and CDSV specimens was designed mainly for compression because, in weak-earthquake regions, it is on the bottom of the beam where compression mostly develops. If tension develops on the bottom of the connection under a strong earthquake, tension can be transferred to the filled concrete in the CFT column through the embedded plate and stud bolts. The beam webs of all specimens were connected to the column through the bolts and the 20 mm-thick single plate. This bolted connection was adopted in order to minimize the influence of shear on the connection and to evaluate the ductility of the connection under pure bending. FCAW (Flux-Cored Arc Welding), with E71T-1 1.6 mm welding rod, was adopted for the welding of all specimens [10]. In Korea, this is the most common method for shop welding and it satisfies the AWS (American Welding Society) standards. SMAW(Shielded Metal Arc Welding) with E7018 1.8 mm welding rod was used in the connection of the steel tube and the beam flange. The end of one plate that was to be welded was cut diagonally and welded to the other plate by groove welds. A backing strip was used in case it was difficult to weld inside the column. The purpose of this was also to minimize weld defects such as weld crack and slag inclusion and to prevent the influence on the test results by welds. As shown in Fig. 14, a complete penetration groove weld was used and a scallop with a radius of 35 mm was placed at the top and bottom corners of the beam webs in the connection between the steel tube and the beam flange, taking into consideration the construction work in the field. 3.2. Material test Table 1 gives the material test results for the modulus of elasticity (E), yield strength (Fy, fck ), tensile strength (Fu ), yield ratio (Fy /Fu ) and elongation (Elo.). 3.3. Loading system The loading system is illustrated in Fig. 15. The specimen was placed on the reaction footing and was hinged-connected to the reaction wall. An actuator with 1960 kN capacity was attached at the end of the beam and preclusive supplementary supports were

966

S.-M. Choi et al. / Journal of Constructional Steel Research 66 (2010) 962–970

Fig. 13. CDSV specimen detail (vertical plate). Fig. 10. CDS specimen detail (Sleeve-inserted combined cross diaphragm).

Fig. 14. Scallop detail in connection.

installed at both ends of the loading point so as to prevent lateral buckling of the beam. 3.4. Loading and measurement method
Fig. 11. CDSN specimen detail (no reinforcing in the bottom).

Fig. 12. CDST specimen detail (horizontal T-bar).

As shown in Fig. 16, a load was applied to the specimen according to the ANSI/AISC SSPEC- 2002 cyclic loading program [5]. The load was controlled by the storey drift angle. The displacement of the beam end was defined using the distance between the loading point of the beam and the center of the column, according to the storey drift angle [11]. A displacement-controlled actuator applied the load. Specimens were loaded for 6 cycles at 0.375%, 0.5% and 075% of the storey drift angle, respectively, and 4 cycles at 1%. After 2% of the storey drift angle, they were loaded for 2 cycles at intervals of 1% of the storey drift angle, until failure of the specimen occurred. Displacement gauges and strain gauges were attached to those points where the overall behavior and local deformation of the specimens could be shown sufficiently. The vertical displacement of the actuator controlled the displacement of the specimens. The load was measured through the load cell in the actuator. The inelastic rotation angle [12] for the measurement of ductility was obtained by dividing the plastic displacement at the beam end by the distance between the beam end and the center of the column, in accordance with the Seismic Provision for Structural Steel Building [5]. The plastic displacement at the beam end was also obtained by subtracting the elastic displacement from the displacement at the beam end.

S.-M. Choi et al. / Journal of Constructional Steel Research 66 (2010) 962–970

967

Both the top and the bottom of the connection had the capacity to surpass the full plastic moment of the beam. 4.1.2. CDSN specimen—No reinforcement The CDSN specimen included the combined cross diaphragm only at the top connection. This made a great deal of difference in the behavior between the top and the bottom of the connection, as shown in Fig. 18(a). At 4 steps (0.01 rad), a crack occurred on the weld zone of the bottom flange of the beam. Furthermore, a minute yielding line was observed in the bottom flange of the beam, while the center of the weld zone, between the bottom flange of the beam and the steel tube, swelled out slightly. At 5 steps (0.015 rad), the thickness of the crack developed to the same thickness as that of the beam flange. The yielding line was apparent mainly in the weld zone between the bottom flange of the beam and the steel tube. At 7 steps (0.03 rad), the bottom of the steel tube swelled out a great deal and it reached its ultimate capacity, with an abrupt failure, at both the steel tube and the bottom flange of the beam, as shown in Fig. 18(b). Under tension in the top of the beam, the capacity of the connection surpassed the full plastic moment of the beam. However, under tension in the bottom of the beam, the capacity of the connection was only 30% of the full plastic moment of the beam. 4.1.3. CDST specimen—Horizontal T-bar The CDST specimen included the horizontal T-bar in the bottom of the connection so as to enhance the confinement effect of concrete and to reduce the concentration of the bearing stress. The specimen exhibited a stable hysteretic curve at an early stage. At 4 steps (0.01 rad), the stiffness of the specimen began to decrease with the introduction of a micro-crack on the weld zone between the bottom flange of the beam and the steel tube. This was similar to the behavior of the CDSN specimen. The bottom of the steel tube also began to swell out. During 5 steps (0.015 rad) and 6 steps (0.02 rad), the preceding change on the bottom of the connection developed gradually. However, there was no appearance of any defect such as a crack or deformation at the top of the connection. At 7 steps (0.03 rad), the specimen reached its ultimate capacity with a fracture at the right edge of the steel tube. As the specimen continued to develop its fracture, it was thought that it had lost its load-carrying capacity and the test was concluded. As shown in Fig. 19(a), there was a significant difference between the capacity of the top and that of the bottom of the connection. While the capacity surpassed the full plastic moment of the beam under tension at the top, the capacity was about 70% of the full plastic moment of the beam under tension at the bottom. 4.1.4. CDSV specimen—Vertical flat-bar In the CDSV specimen, the bottom of the connection was reinforced with vertical stiffeners welded to the inner surface of the steel tube. The specimen exhibited elastic behavior until 4 steps (0.01 rad). At 4 steps, the bottom flange of the beam yielded with a micro-crack at the weld zone between the bottom flange of the beam and the steel tube. After 5–6 steps, there was a significant difference in the behavior between the top and the bottom. The top of the connection increased its capacity with a corresponding increase in crack width. On the other hand, the capacity of the bottom decreased rapidly as the bottom of the connection experienced a plastic state. At 7 steps, it reached its ultimate state. Then, as the lower diaphragm experienced a plastic state, the steel tube could no longer bear the tensile force from the bottom flange of the beam and it consequently failed abruptly. The capacity of the top of the connection surpassed the full plastic moment of the beam under the tension in the top, while that of the bottom of the connection was 70% of the full plastic moment of the beam under the tension in the bottom. This behavior was similar to that of the CDST specimen.

Fig. 15. Loading system.

D

Fig. 16. ANSI/AISC SSPEC-2002.

4. Test results and analysis 4.1. Moment–rotation angle relation The relationship between the moment and the rotation angle for each specimen and its failure mode are shown in Figs. 17–20. The moment is the value of the load at the loading point multiplied by the distance between the loading point and the center of the column. The rotation angle is the value of the displacement at the loading point, divided by 3500 mm, which is the length of the beam [11]. In addition, the full plastic moment of the beam (Mp ) is the yielding stress of the beam multiplied by the plastic section modulus of the beam. Also, the hysteretic curves between the moment and rotation angles were converted to the monotonic curves for all specimens, in order to compare the behaviors of specimens. Fig. 21 shows the converted monotonic curves. 4.1.1. CDS specimen—Sleeve-inserted combined cross diaphragm The CDS specimen exhibited the most stable hysteretic curve overall, as shown in Fig. 17(a). At 3 steps (0.00375 rad), the surface of the beam flange fell off and the beam started yielding and undergoing a plastic region without any apparent physical change until 6 steps. At 6 steps (0.02 rad), a crack was found at the lower diaphragm and at the bottom flange of the beam, about 1 cm from the column. Subsequently, the capacity of the connection started to decrease as the crack width increased until 2 cycles of 7 steps (0.03 rad). After it reached its ultimate capacity at 7 steps (0.03 rad), the bottom flange of the beam fractured and the test was concluded.

968

S.-M. Choi et al. / Journal of Constructional Steel Research 66 (2010) 962–970

(a) Moment–rotation angle curve.

(b) Photo of failure at 0.03 rad. Fig. 17. Load test result of CDS specimen.

(a) Moment–rotation angle curve.

(b) Photo of failure at 0.03 rad. Fig. 18. Load test result of CDSN specimen.

(a) Moment–rotation angle curve.

(b) Photo of failure at 0.03 rad. Fig. 19. Load test result of CDST specimen.

(a) Moment–rotation angle curve.

(b) Photo of failure at 0.03 rad. Fig. 20. Load test result of CDSV specimen.

S.-M. Choi et al. / Journal of Constructional Steel Research 66 (2010) 962–970

969

4.2. Initial stiffness and ultimate capacity The capacity and stiffness of each specimen are given in Figs. 22 and 23 and in Table 2. When the top flange of the beam was under tension, all the specimens exhibited almost the same capacity for the upper diaphragm of them that was identically reinforced with a combined cross diaphragm. On the other hand, specimens behaved differently from one another under tension in the bottom flange of the beam. The ratios of the ultimate capacity form the top to the bottom are defined as the ultimate capacity under the tension in the top, over the ultimate capacity under the tension in the bottom. These were 1.04, 0.23, 0.51 and 0.48 for CDS, CDSN, CDST and CDSV, respectively. The CDS specimen exhibited nearly the same capacities at the top as at the bottom. The bottom of the CDSN specimen had much less capacity than the top and it changed into a plastic state earlier than did the other specimens, as shown in Fig. 21. This was because, in the CDSN specimen, only the steel tube carried the tensile force of the bottom flange of the beam. The bottom of the CDST specimen had a smaller capacity than the CDS specimen but twice as much capacity as the CDSN specimen. This was due to the anchoring of the lower diaphragm in concrete. The bottom of the CDST specimen showed a decrease in capacity after yielding, as the stiffness of the steel tube decreased. This was due to the tension transferred from the bottom flange of the beam. The CDSV specimen was similar to the CDST specimen in both capacity and behavior. The bottom of the CDSV specimen showed a decrease in capacity due to the deformation of the steel. This is considered to be due to the lower diaphragm of the CDST specimen being installed vertically and that the confining force of concrete generated on the vertical diaphragm was relatively small. Fig. 23 presents the ratio of the stiffness from the top to the bottom at the initial loading for each specimen. With the exception of the CDSN specimen, there was no significant difference in the ratios of stiffness between the specimens, from the ratios of the capacity in Fig. 22. However, in the CDSN specimen that had no reinforcement in the bottom, the stiffness of the bottom was about half of the stiffness of the top. It is therefore considered necessary to reinforce the bottom of the connection. 4.3. Plastic deformation capacity
1.2

Fig. 21. Monotonic load-displacement curve [13,14]. Table 2 Cyclic load test results in capacity and stiffness. No. Fu (kN/mm) Top CDS CDSN CDST CDSV 196.0 214.6 208.7 216.6 Bottom 203.8 48.4 105.8 104.9 Ki (kN/mm) Top 4.5 3.0 3.7 3.8 Bottom 4.1 1.7 3.7 3.3

1.2 1.04 1 0.8 Fb / Ft 0.6 0.4 0.23 0.2 0

Fb : Bottom Maximum Strength Ft : Top Maximum Strength

0.51

0.48

CDS

CDSN

CDST

CDSV

Fig. 22. Maximum capacity comparison of specimens.

In an earthquake, the capacity of plastic deformation to absorb large amounts of energy and shear force is essential to the stability of the column-to-beam connection of the structure. Table 3 presents the inelastic rotation angle of connection required for the moment frame, according to the AISC Seismic Provisions for Structural Steel Buildings. Fig. 24 presents the inelastic rotation angle of each specimen under tension in the top and the bottom. Under tension in the top, all specimens exhibited an inelastic rotation angle of more than 0.02 rad and could be classified as a composite intermediate moment frame. Under tension in the bottom, the CDST and CDSV specimens had an inelastic rotation angle of more than 0.01 rad and could be classified as a composite ordinary moment frame. The inelastic rotation angle of the CDSN was greater than that of the CDST and CDSV specimens. It is thought that this was because the deformation of the CDSN specimen continued on without an increase in capacity, after yielding earlier than other specimens. All specimens satisfied the requirements for the composite ordinary moment frame in any case. 4.4. Energy absorbing capacity The energy absorbing capacity of a structure is one of the most important factors in its seismic performance. The energy absorbed by the deformation of the structure can be measured as the area enclosed by the load-displacement hysteretic curve. This is the

1 0.8 Kb / Kt 0.6 0.4 0.2 0

Kb : Bottom Initial Stiffness Kt : Top Initial Stiffness 0.91

1.00 0.87

0.55

CDS

CDSN

CDST

CDSV

Fig. 23. Initial stiffness comparison of specimens.

total plastic work performed by the structure. Fig. 25 compares the total work performed by each specimen until a rotation angle of 0.02 rad is reached. The value of 0.02 rad was selected because fracture of all specimens occurred at an early stage, although they could potentially reach an inelastic rotation angle of 0.03 rad. The total work performed by the CDS specimen was the largest and total work performed by the other specimens was only 31%–68% of that of the CDS specimen. The test results demonstrated that the suggested connection types, with a simplified lower diaphragm, could obtain a sufficient energy absorbing capacity as that required to be used in weak-

970

S.-M. Choi et al. / Journal of Constructional Steel Research 66 (2010) 962–970

Table 3 Connection inelastic rotation capacity requirements. Frame classifications Composite ordinary moment frame (C-OMF) Composite intermediate moment frame (C-IMF) Composite special moment frame (C-SMF) Connection inelastic rotation capacity (rad) 0.01 0.02 0.03

another according to the type of lower diaphragm. It is considered that the suggested types can obtain sufficient energy absorbing capacity by reinforcing only the top of the connection with a combined cross diaphragm. These cyclic load test results demonstrated that the suggested connection types, which were reinforced with the asymmetric lower diaphragm, could obtain more than 0.01 rad of the inelastic rotation capacity, and so could be classified as a composite ordinary moment frame and have sufficient seismic performance. Therefore, it is concluded that the CFT column-to-beam connection with asymmetric lower diaphragm can be safely used in weakearthquake regions.
Fig. 24. Comparison of the inelastic rotation angle.

Acknowledgements This study was conducted with the support of the Construction Technology Innovation Program of the R&D Project (Grant 05 Construction Consequence C 26) funded by the Ministry of Construction & Transportation of Korean government and the National Research Laboratory (Grant R0A-2007-000-10047-0). This financial support is greatly appreciated. References
[1] Choi SM, Kim DK, Kang DA, Jeong KS. A study on development of concrete filled circular tube column-to-beam connection. Research report. Korea: University of Seoul; 1996. [2] Choi SM, Kim DK. A study on the seismic performance evaluation and design method of the CFT square column. Research report. Korean Society of Steel Construction. 1999. [3] Elremaily A, Azizinamini A. Experimental behavior of steel beam to CFT column connections. Journal of Constructional Steel Research 2001;57(10): 1099–119. [4] Ricles JM, Peng SW, Lu LW. Seismic behavior of composite concrete filled steel tube column-wide flange beam moment connections. Journal of Structural Engineering 2004;130(2):223–32. [5] American Institute of Steel Construction. Inc. AISC. Supplement 1 and 2 to seismic provisions for structural steel buildings. Chicago (IL): AISC. May 21; 2002. [6] Choi SM, Kim DK, Kim JH. Design & construction guide on concrete filled tubular structures. Research report. Korean Society of Steel Construction. 2000. [7] Choi SM, Hong SD, Kim YS, Kim JH. Simple tension testing for CFT columnto-beam connections at tension side with new diaphragm. In: International symposium on steel structure: Second symposium. Korean Society of Steel Structure 2002. 2002. p. 405–16. [8] Choi SM, Cha EJ, Kim YS, Kim JH. Reliability analysis for CFT column-to-beam connections with new diaphragm. In: Second international symposium on steel structures. 2002. p. 103–14. [9] Choi SM, Hong SD, Kim DG, Kim YS, Kim JH. Structural capacities of tension side for CFT square column-to-beam connections with combined-crossdiaphragm. PSSC. Mar. 2004. 2004. p. 24–7. [10] Choi SM, Yun YS, Kim JH. Experimental study on seismic performance of concrete filled tubular square column-to-beam connections with combined cross diaphragm. An International Journal of Steel and Composite Structures 2006;6(4):303–18. [11] SAC. SAC/BD-97/02 Version 1.1. Protocol for fabrication, inspection, testing, and documentation of beam–column connection tests and other specimens. Sacramento (CA): SAC Joint Venture; 1997. [12] Giton CS, Uang CM. Cyclic response and design recommendations of weak-axis reduced beam section moment connections. Journal of Structural Engineering 2002. [13] Cheng Chin-Tung, Chung Lap-Loi. Seismic performance of steel beams to concrete-filled steel tubular column connections. Journal of Constructional Steel Research 2003;59(3):405–26. [14] Chou Chung-Che, Uang Chia-Ming. Cyclic performance of a type of steel beam to steel-encased reinforced concrete column moment connection. Journal of Constructional Steel Research 2002;58(5–8):637–63.

Fig. 25. Comparison of total plastic work until 0.02 rad of the rotation angle.

earthquake regions, though they had less energy absorbing capacity than did the connection with the combined cross diaphragm at both the top and bottom, which can be used in strong-earthquake regions. 5. Conclusions This study suggested that the CFT column-to-bottom connection details be used with an asymmetric lower diaphragm. To this end, the study carried out a simple tension test and cyclic loading test on these connections. The test results are as follows: The simple tension test results showed that the ultimate capacities of the HT and VP specimens were only about one half of that of the CDS specimen. Similarly, the initial stiffness of the CDS specimen was 40%–45% higher than that of the HT and VP specimens. As the cyclic load test results demonstrate, all specimens showed a similar ultimate capacity under tension in the top. Under tension in the bottom, the CDSN, CDST and CDSV specimens had a 23%–43% capacity of that of the CDS specimen. The top of the connection reached 0.02 rad of the inelastic rotation capacity required for the composite intermediate moment frame and, in all specimens, the bottom of the connection reached 0.01 rad for the composite ordinary moment frame. In terms of the energy absorbing capacity, the CDSN, CDST and CDSV specimens had a 31%–68% capacity of the CDS specimen. The energy absorbing capacities of suggested connections were different from one


相关文章:
英语基础测试试题69
on the steel structure and reinforced concrete ... seismic response caused by earthquake action will... filled in the circular or square steel tube. ...
福大专业英语复习
seismic performance evaluation procedure for steel ...study on the structural behavior of segmental ...of concrete-filled steel tube 钢管混凝土界面抗剪...
2010.7.26土木建筑学师资与品牌(1)
concrete-filled steel tube bending and axial ...performance of concrete, C30 - C80 concrete are ... and studied the seismic behavior of seismic ...
建筑结构多维地震作用研究综述
Key words: Composite structure; Seismic performance; Composite joints of RCS;Joints of concrete-filled steel tube;Composite joints of SRC 1 1.引言地震动是...
土木工程专业英语论文
Compared with other structure of concrete filled steel tube structure, having the advantages of high bearing capacity, ductility and seismic performance, ...
广西大学 建筑工程毕业合计 外文翻译
Compared with other structure of concrete filled steel tube structure, having the advantages of high bearing capacity, ductility and seismic performance, ...
方钢管混凝土框架结构的研究及展望
This paper each other the concrete-filled steel tubular frame structures, the seismic performance summarized. At the same time, to recent research part of...
参考文献
of connections to concrete-filled steel tubes[J]...square tube column connection with vertical ...Imanpour A.Seismic performance of the accordion-...
上承式大跨度钢管混凝土拱桥地震反应分析
In order to study the seismic response of deck-type long span CFST(concrete filled steel tube) arch bridge, the deck-type CFST arch bridge—Zhijing...
卢教授2007—2010论文
Concrete-filled Steel Tubular Tower》 》来源出版物 : ICIC 2009: SECOND...《TEST AND STUDY ON SEISMIC PERFORMANCE OF BONDED PRESTRESSED 《 CONCRETE ...
更多相关标签: