New type of shear connector for square concrete-filled tubular columns

Abstract

Concrete-filled tubes (CFT) are economical and easy to fabricate. As the cross-sections of CFTs are small, the interface shear between steel tube and concrete can be transferred by direct bond. As the cross-sectional dimension increases, shear connectors may be required to transfer the interface shear. However, welding shear connectors inside the steel tube is difficult. In this study, a new type of shear connector that could be installed from outside the steel tube was developed. A pilot test on five specimens demonstrated that the corner shear connector, which is a steel plate located at the corner of a square steel tube, at a 45° angle to the tube wall, is much superior to shear studs in terms of transferring the interfacial shear force. Furthermore, push-out tests of 18 CFT specimens were performed to investigate the strength and the behavior of the corner shear connectors. The strength of the specimens is found to be independent of the cross-sectional area of the shear connector. Instead, the jointed bearing area, which is defined as the cross-sectional area of the shear connection combining the area enclosed by the shear connector and the steel tube, is the major parameter that is closely correlated with the strength of the specimens. Furthermore, the ultimate strength of the specimens is positively correlated with the bearing distance, which is the distance between the loading point and the bearing surface of the connector. In addition, the strength of the corner shear connectors, which have a total jointed bearing area equaling 10.8% of the total concrete area in the steel tube, can adequately satisfy the strictest requirement for the shear transfer.

Keywords

Composite structure concrete-filled tubular columns shear connector push-out test steel-reinforced concrete structure

Introduction

Composite columns consist of steel sections and concrete. Most configurations of composite columns include steel covered by concrete, which is designated as an encased composite column (Chen and Lin, 2009a; Chen et al., 2009b, 2016, 2018, 2022; Jamkhaneh et al., 2020), and steel tubes filled with concrete, which is designated as a concrete-filled tubular (CFT) columns (Ahmadi et al., 2017, 2018; Furlong, 1967; Gan et al., 2019; Roeder et al., 1999a, 1999b; Zhou et al., 2019). CFT columns are considered more economical than reinforced concrete columns (Furlong, 1967); therefore, it is essential to promote their usage.

In a CFT column, interfacial shear may exist between the tube and the concrete. For example, near a girder to CFT connection, as shown in Figure 1 (e.g. ANSI/AISC 360-16, 2016), the vertical force in the beam is transferred to the steel tube through bolts and a shear tab. The steel tube then transfers part of this vertical force to the concrete. Consequently, shear force must be transferred between the steel tube and the concrete. Interfacial shear force transfer near the brace-beam-column connection in a CFT may also be required, as shown in Figure 2 (Roeder et al., 1999a). The interfacial shear force between the steel tube and the concrete is parallel to the column axial direction and is therefore called the longitudinal shear force (e.g. ANSI/AISC 360-16, 2016). When the longitudinal shear force is small, direct bond between the concrete and the steel tube may have sufficient strength to transfer it. Otherwise, shear connectors or direct bearing may be needed, particularly for CFTs with a large cross-sectional dimension.

Figure 1.

Schematic of longitudinal shear force transfer near girder to CFT connection (Ref. AISC 360-16, Fig. C-I6.1).

Figure 2.

Schematic of longitudinal shear force transfer near brace-beam-column connection.

Shear studs, which are welded to the inner surface of the tube, are commonly used as shear connectors. However, stud welding, which is used to install shear studs, requires a large working space. In many cases, the space in the tube is insufficient for stud welding. As a solution to this problem, an innovative type of shear connector that can be installed outside the CFT column is proposed in study. Since this new type of shear connector is placed at the corner of square steel tubes, it is called corner shear connector.

Longitudinal shear force also exists when CFT column presumably possesses a certain degree of composite action. However, previous research (Furlong, 1967) has shown that CFT beam columns possess a certain degree of composite action without the use of shear connectors. The ANSI/AISC 360-05 (2005) provision also considers that any amount of flexure may engage the concrete sufficiently to help ensure longitudinal shear force transfer (e.g. ANSI/AISC 360-05, 2005; Leon et al., 2007). Therefore, direct bond has enough strength to ensure a composite action for CFT beams or beam columns. Accordingly, the shear connector developed in this study was mainly for the longitudinal shear force occurring in the similar cases shown in Figures 1 and 2.

Pilot tests on shear connectors

Steel plates functioning as longitudinal stiffeners are placed inside square tubes to enhance the local buckling resistance and the interface shear capacity of CFTs (Gan et al., 2019; Zhou et al., 2019). In this study, steel plates were considered for developing a new type of shear connectors, and a pilot test was performed to identify the most effective configuration of the steel plates. Five square CFT specimens, denoted as PLP, PPP, CRP, ST3, and ST5, were designed, fabricated, and tested to determine the most effective steel-plate configuration for interfacial shear force transfer. All specimens were fabricated from hollow structural section (HSS) with dimensions of 300 mm × 300 mm × 9 mm. The PLP, PPP, and CRP specimens used four steel plates, with a thickness of 12 mm and a width of 70 mm, as shear connectors. The steel plates in the PLP specimen were centered along each wall of the square tube and positioned flat against the tube wall, as shown in Figure 3(b). In the PPP specimen, the plates were centered along each tube wall in a perpendicular position, as shown in Figure 3(c). In the CRP specimen, the plates were located on the four corners of the steel tube at a 45° angle to the tube walls, as shown in Figure 3(d). For comparison, two specimens with shear studs, named ST3 and ST5, were fabricated and installed. In these specimens, a shear stud with a diameter d = 13 mm was placed at the center of each tube wall, as shown in Figure 3(a). The lengths of the shear studs in specimens ST3 and ST5 were 3d and 5d, respectively. The cross-sectional area of the steel plates used in the PLP, PPP, and CRP specimens was nearly identical to the projection area of the shear stud used in ST5.

Figure 3.

Cross-sections of the pilot test specimens.

Figure 4 depicts a schematic of the vertical cross-section of the test specimens. The heights of the steel tube and the concrete were 350 mm and 300 mm, respectively. On top of the concrete, a clearance of 50 mm was reserved for slippage during the load test. A load was applied to the steel tube from the top. The reaction force was received by a steel-supporting plate. This supporting plate was in contact only with the concrete at the bottom of the specimen. The applied load was transferred from the loading plate to the steel tube, then to the concrete through the shear connectors, and finally from the concrete to the supporting plate. The bottom of the shear connector was the bearing surface that bears against the concrete. For specimens ST3 and ST5, the center of the shear studs was defined as the bearing surface. The distance between the bearing surface and the top of the supporting plate was the bearing distance L_b, as shown in Figures 3 and 4. The magnitude of L_b was selected as 150 mm, which is equal to one half of the cross-sectional dimension, to reduce stress disturbance near the bearing surface. When pouring the concrete, the bearing surface faced downward.

Figure 4.

Schematic of the vertical cross-section of the pilot test specimens.

Steel tubes meeting the JIS STKR490 (2015) standard requirement with a measured yield stress of 387 MPa were used. Steel plates meeting the ASTM A36/A36M-14 (2016) standard requirement with a measured yield stress of 251 MPa were used for the shear connectors. Shear studs with a nominal yield stress of 446 MPa were used. The measured average compressive strength of the concrete during the loading test period was 22.2 MPa.

The test setup employed for the pilot test is shown in Figure 5. A 600-ton universal testing machine (UTM) was used to apply a quasi-static monotonic load P to the top of the specimen, which was measured using the embedded load cell of the UTM. The downward displacement at the bottom of the steel tube was measured using dial gauges D1 and D2; the average displacement was the relative slippage δ between the steel tube and the concrete.

Figure 5.

Test setup for the pilot test.

Figure 6(a) illustrates the P-δ curves of the five specimens. The CRP and PPP specimens were remarkably stronger than PLP, ST5, and ST3. The strength of CRP started to develop the earliest and therefore reached the highest value. When δ = 5 mm, the specimen strength of CRP (1060 kN) was 3.7 times that of ST5 (290 kN). The shear studs in ST5 and ST3 fail when δ approaches 15 mm. Some studies (Kumar and Chaudhary, 2019; Pallarés and Hajjar, 2010; Shen and Chung, 2017; Tao et al., 2016; Wang et al., 2018; Xue et al., 2012) have reported similar results. However, even when δ reached 35 mm, the strengths of the CRP and PPP specimens was still developing gradually. The PPP specimen was stronger than ST5 and ST3 because the flexural stiffness and strength of the shear connector used in PPP are much greater than those of ST5 and ST3. However, after the loading test, the steel tube wall in the PPP specimen exhibited an out-of-plane deformation of more than 30 mm, as shown in Figure 7, whereas the other specimens exhibited smaller out-of-plane deformations. Overall, the CRP specimen exhibited a high strength and small out-of-plane deformation of the steel tube; thus, it is the optimal shear connector. Because this shear connector was located at the corner of the steel tube, we named it “corner shear connector.”

Figure 6.

P-δ curve of the pilot test specimens.

Figure 7.

Out-of-plane deformation of the PPP specimen.

In addition, the local P-δ curve, as illustrated in Figure 6(b), shows that, when the load was 100–225 kN, all specimens exhibited a sudden decrease in load and increase in slippage. Subsequently, as the amount of slippage increased, the load on the specimen increased gradually initially and then rapidly for all the specimens. This phenomenon was basically caused by the concrete bleeding, since concrete bleeding create a void next to the bearing surface, as indicated in Figure 8.

Figure 8.

Concrete bleeding next to the bearing surface.

Pilot tests showed that the corner shear connector is efficient in transferring the interfacial shear force of the CFT. In addition, by introducing slits in the tube, the corner shear connector can be installed from the outside of the tube. Figure 9 illustrates the details of the slits and welds when a steel plate is used as the corner shear connector.

Figure 9.

Detailed structure of the corner shear connector.

The pilot tests also revealed that the concrete bleeding at the bearing surface of the corner shear connector might affect the connector’s bearing behavior. Therefore, subsequent tests were planned with the BSC series of specimens to explore the influence of concrete bleeding. However, to be conservative, the other specimens were fabricated such that the bearing surface experienced concrete bleeding.

Loading test of corner shear connectors

Specimen design and fabrication

Four series, namely BSC, ASB, AJB, and LB, with a total of 18 square CFT specimens were designed and fabricated for the push-out tests. Table 1 lists the designation of the specimens and their primary dimensions. All specimens used HSS 400 mm × 400 mm × 12 mm tubes. All specimens had a corner shear connector at each corner of the steel tube with the welding details as shown in Figure 10(a).

Table 1.

Details of the BSC, ASB, AJB, and LB specimens.

Series	Specimen type	t_s (mm)	w_s (mm)	a_sb (mm²)	a_jb (mm²)	L_b (×B)	γ_jb (%)	Number of specimens
BSC	BSC-B	12	90	936	1990	1.0	5.6	3
BSC	BSC-NB	12	90	936	1990	1.0	5.6	2
ASB	ASB-S36	9	89	720	1950	1.0	5.5	1
	ASB-S47 (same as BSC-B)	12	90	936	1990		5.6	—
	ASB-S58	15	92	1160	2090		5.9	1
	ASB-S100	Solid steel block		1990	1990		5.6	1
AJB	AJB-5.6 (same as BSC-B)	12	90	936	1990	1.0	5.6	—
	AJB-8.2		108	1150	2890		8.2	1
	AJB-10.8		124	1340	3810		10.8	1
	AJB-13.4		138	1510	4730		13.4	1
	AJB-15.9		150	1660	5590		15.9	1
LB	LB-0.5B	12	90	936	1990	0.5	5.6	2
	LB-0.75B					0.75		2
	LB-1.0B (same as BSC-B)					1.0		—
	LB-1.5B					1.5		2

Figure 10.

Definition of a_sb, a_jb and R

The thickness and width of the corner shear connector are denoted as t_s and w_s, respectively. The sum of the cross-sectional areas of the corner steel plate and the welds is the steel bearing area a_sb, which can be obtain by using equation (1). The sum of a_sb and the area enclosed by the corner shear connector and the steel tube corner is the jointed bearing area a_jb, which is illustrated by the shaded area in Figure 10(b). Equation (2) is used to calculate a_jb, where R is the radius of the arc of inner surface at tube corner, as indicated in Figure 10(c). One corner shear connector was placed at each corner of the tube; thus, there were four corner shear connectors at the same elevation in every specimen. The total steel bearing area of all four corner shear connectors is denoted by A_sb, and consequently A_sb = 4 a_sb. The total jointed bearing area of all four corner shear connectors is denoted A_jb, where A_jb = 4 a_jb.

a_{s b} = (w_{s} - t_{s}) t_{s}

(1)

a_{j b} = \frac{w_{s}^{2}}{4} + (\frac{π}{4} - 1) {(R - t)}^{2}

(2)

Figure 11 illustrates a schematic of the vertical cross-section of a specimen. The height of the steel tube is denoted as L. The length of the corner shear connector was 150 mm. A total clearance of 50 mm was reserved for slippage at the bottom of the specimen. Load was applied on a steel-loading plate on top of the specimen and transferred to the concrete. The plate did not touch the steel tube. The applied load is transferred to the concrete through the loading plate, then the load is transferred to the steel tube through the shear connectors, and finally, the load is transferred form steel tube to the support.

Figure 11.

Schematic of the vertical cross-section of a specimen.

The BSC specimens were employed to explore the effect of concrete bleeding at the bearing surface. As shown in Table 1, these specimens all have identical specifications: t_s = 12 mm, w_s = 90 mm, a_sb = 936 mm², a_jb = 1990 mm², and L_b = 1.0 B, where B is the width of the steel tube. The BSC specimens were designated as BSC-B and BSC-NB. BSC-B consisted of three identical specimens with concrete bleeding at the bearing surface. When pouring the concrete, the bearing surface faced downward. BSC-NB consisted of two identical specimens with no concrete bleeding at the bearing surface. When pouring the concrete, the bearing surface faced upward. Premixed concrete with a maximum coarse aggregate dimension of 19 mm were used for all specimens.

Each corner shear connector for the ASB specimens had the same a_jb as the steel bearing area a_sb was varied. In ASB-S100, solid steel blocks with their a_sb equal to a_jb were used, as shown in Figure 12. On the other specimens, a_sb was between 720 mm² and 1160 mm². The differences originated from varying the straight corner steel plate thickness t_s, as shown in Table 1. For each specimen, the ratio of a_sb to that of ASB-S100 $[i . e ., a_{s b} / {(a_{s b})}_{ASB - S 100}]$ was defined as γ_sb. The γ_sb for specimens in the ASB series ranged from 36-100%. The number following the specimen serial notation ASB-S is the γ_sb percentage of that specimen. All specimens had a similar A_jb, between 7800 and 8360 mm². The L_b for all specimens was 1.0 B. Except for ASB-S47, each specimen designation had only one specimen. ASB-S47 had three identical specimens and is also called BSC-B, AJB-5.6, and LB-1.0B, as explained in the following paragraphs.

Figure 12.

Corner shear connector of ASB-S100 specimen.

The major parameter for the AJB specimens was the jointed bearing area a_jb. a_jb ranged from 1990-5590 mm², whereas A_jb ranged from 7960-22,400 mm². The ratio of A_jb to A_c, which is the total cross-section of the concrete inside the steel tube, was the jointed bearing area ratio γ_jb. γ_jb ranged from 5.6-15.9%, as listed in Table 1. The number following the specimen serial notation AJB- is the γ_jb percentage of that specimen. Except for AJB-5.6, only one specimen was used from each denotation. For the AJB specimens, t_s = 12 mm and L_b = 1.0 B.

As the LB specimens were employed to explore the effect of the bearing distance, L_b was gradually increased from 0.5 B to 1.5 B (Table 1). The number following specimen notation “LB-” is L_b. For example, 0.5 B denotes the L_b that is 0.5 times the steel tube width. LB-1.0B had three identical specimens, and LB-0.5B, LB-0.75B, and LB-1.5B had two identical specimens. For the LB specimens, t_s = 12 mm, w_s = 90 mm, a_sb = 936 mm², and a_jb = 1990 mm².

As mentioned earlier, we used the JIS STKR490 (2015) standard HSS tubes for the specimens. The corner shear connector, including the solid steel block used in ASB-S100, was fabricated from ASTM A572/A572M-15 (2016) standard steel. The measured yield stress F_ya and tensile strength F_ua of all steel materials are listed in Table 2. The original idea is to install the corner shear connectors from outside the tube. However, for reducing budged and easy construction purposes, the corner shear connectors were welded to the steel tube from inside the tube without machining slits. Figure 13(a) illustrates the installed corner shear connectors of the AJB-13.4 specimen. Figure 13(b) shows the welds between a corner shear connector and the steel tube. During the loading test, the mean measured compressive strengths

{f^{'}}_{c}

was 22.4 MPa.

Table 2.

Mechanical properties of steel specimens.

Specification	Thickness (mm)	F_ya (MPa)	F_ua (MPa)	Specimen
STKR490	12	429	502	Square HSS made of BSC, ASB, AJB, and LB series
A572 Gr.50	9	396	503	Corner shear connector made of ASB-S36
	12	415	505	Corner shear connector made of BSC, AJB, and LB series
	12	415	505	Corner shear connector made of ASB-S47
	15	407	542	Corner shear connector made of ASB-S58
	50	352	529	Solid steel block made of ASB-S100

Figure 13.

Corner shear connectors in the AJB-13.4 specimen.

Test setup and application of load

Figure 14 illustrates the apparatus used for the push-out test. Specimens were placed under a 600-ton UTM. The load was transferred to the concrete section of the specimens directly through a steel-loading plate. The loading plate covered approximately 90% of the concrete area. A 440 mm × 440 mm steel reaction box seat with a steel plate thickness of 50 mm was placed under the specimen. The upper end of the reaction box was closely in contact with the steel tube of the specimen, while the bottom end transferred the load to the base of the testing machine. With this arrangement, dial gauges could then be placed in the space within the box.

Figure 14.

Test setup for BSC, ASB, AJB, and LB specimens.

The load P was measured from the embedded load cell in the UTM. The test machine base was used as the reference point for displacement measurement. Dial gauges D1 and D2 measured the downward displacement of concrete at the bottom of the specimen. The mean displacement was used as slippage δ between the steel tube and the concrete. A quasi-static monotonic load was applied on the specimens with a displacement rate of 0.03 mm/s. As δ reached approximately 35 mm, the loading test terminated. Figures 15(a) and (b) show the appearance of the ASB-S100 specimen before and after the loading test.

Figure 15.

Photographs of the ASB-S100 specimen.

Results and discussion

Effect of concrete bleeding

Figure 16 presents the P-δ curve of the BSC specimens; Figure 16(a) shows the overall curve and Figure 16(b) the locally enlarged partial curve. Three BSC-B specimens, which suffered concrete bleeding next to the bearing surface, were tested, and the P-δ curves of these specimens were found to be similar. In addition, two BSC-NB specimens, which didn’t have concrete bleeding next to the bearing surface, also had similar P-δ curves. Therefore, the specimen variability was reasonably small.

Figure 16.

P-δ curve of the BSC specimens.

As illustrated in Figure 16(b), the P-δ curve of the BSC-B specimens show that the load plunged as P reached 250–300 kN, averaging at 51 kN, while δ increased simultaneously at a mean increment of 0.07 mm. We concluded the reason for this drop to be bond failure between the steel tube and concrete. The P-δ curve of the BSC-NB specimens shows significant stiffness degradation at a load of approximately 420 kN. This, too, was concluded as the effect of bond failure between the steel tube and concrete.

The BSC-NB specimens recovered their strength after bond failure much faster than the BSC-B specimens. As illustrated in Figure 16(a), at a load of 1500 kN, the amount of slippage in the two groups of specimens differed by 3.8 mm on average. Concrete bleeding substantially increased the amount of slippage, thereby delaying the strength development. However, regarding the maximum load on the specimen, the BSC-B specimens subtly outperformed the BSC-NB specimens. The test results revealed that concrete bleeding caused a delay in the strength development but did not influence the maximum strength.

Effect of steel bearing area A_sb

Figure 17(a) illustrates the P-δ curve of the ASB specimens. The BSC-B specimens have been renamed ASB-S47 in the figure. Three specimens were classified under ASB-S47, the individual specimen P-δ curves of which are shown in Figure 16(a). Under the same δ, the mean load was calculated for the three P-δ curves to obtain the mean P-δ curve. The mean curve was then used to represent the P-δ curve of ASB-S47, as shown in Figure 17(a).

Figure 17.

P-δ curve of the ASB specimens.

This curves show that the bearing surface of all specimens undergo concrete bleeding. However, the ASB-S100 specimen was the least affected. To facilitate comparison, we assumed that ASB-S100 was subject to a similar concrete bleeding effect as the other three specimens. Therefore, the P-δ curve of ASB-S100 shifted 0.5 mm to the right upon visual examination. Figure 17(b) shows the curves after the shifts.

The major parameter of the ASB specimens was γ_sb (or A_sb). Although this parameter varied from 36 to 100%, the P-δ curves of the four specimens were not substantially different. According to Figure 17(b), let δ = 5 mm and 30 mm correspond to loads P₅ and P₃₀, respectively; the values of these loads are listed in Table 3. Figure 18 shows the distribution of P₅ and P₃₀ with respect to A_sb. Additionally, P₅ and P₃₀ is independent of A_sb. The test results confirm that A_sb had a minimal influence on the strength and behavior of specimens.

Table 3.

Test results of ASB, AJB, and LB specimens.

Series	Specimen	A_sb (mm²)	A_jb (mm²)	P₅ (kN)	P₃₀ (kN)
ASB	ASB-S36	2880	7800	1020	2340
	ASB-S47	3740	7960	1050	2230
	ASB-S58	4640	8360	1110	2380
	ASB-S100	7960	7960	1140	2380
AJB	AJB-5.6	3740	7960	1050	2230
	AJB-8.2	4600	11,600	1240	2540
	AJB-10.8	5360	15,200	1340	2720
	AJB-13.4	6040	18,900	1450	2950
	AJB-15.9	6640	22,400	1680	3180
LB	LB-0.5B	3740	7960	1040	1830
	LB-0.75B			1070	1970
	LB-1.0B			1050	2230
	LB-1.5B			1170	2870

Figure 18.

Distribution of P₅ and P₃₀ of the ASB specimens.

Effect of jointed bearing area A_jb

Figure 19(a) presents the P-δ curves of the five AJB specimens. Concrete bleeding had nearly the same effect on AJB-5.6, AJB-8.2, AJB-10.8, and AJB-13.4, but much smaller on AJB-15.9. To facilitate comparison, we assumed that the effect of concrete bleeding on AJB-15.9 was like that for the other four specimens. Therefore, the P-δ curve of AJB-15.9 shifted 1.3 mm to the right according to visual assessment. Figure 19(b) illustrates the new curves.

Figure 19.

P-δ curve of the AJB specimens.

The γ_jb of the AJB specimens increased from 5.6 to 15.9%. According to Figure 19(b), specimens with a larger γ_jb or A_jb were stronger. The P₅ and P₃₀ of the specimens were obtained from Figure 19(b) and are listed in Table 3. Figure 20 demonstrates the correlations of A_jb with P₅ and P₃₀; the two loads increase with the increase in A_jb. When A_jb is 7960-22,400 mm², its relationship with the load is linear. Clearly, A_jb is a critical parameter affecting the specimen strength.

Figure 20.

Distribution of P₅ and P₃₀ of the AJB specimens.

We defined the nominal axial compressive strength P_nc as the product of the cross-section of the concrete A_c and 85% of the measured concrete compressive strength (e.g. ACI Committee, 2014). The value of P_nc was 2680 kN for the AJB specimens. P_nc is the maximum interfacial shear force that should be transferred. From Table 3 and P_nc = 2680 kN, the P₃₀/P_nc of the AJB specimens was 0.83–1.19. This reveals that, one corner shear connector at each tube corner, if designed properly, can adequately fulfill the strictest shear transfer requirement. The corner shear connector is highly efficient in transferring the interfacial shear force.

Specimen AJB-15.9 had the highest strength among all the specimens in ASB, AJB and LB series. After the test, part of the concrete in the tube was removed. The corner shear connectors and welds now appear as shown in Figure 21. At the bearing surface, the connector was in close contact with the concrete. The shear connectors and welds did not show any signs of damage.

Figure 21.

Appearance of the corner shear connectors of AJB-15.9 specimen after test.

Effect of bearing distance L_b

Figure 22 illustrates the P-δ curve of the LB specimens, where Figure 22(a) is the overall curve and Figure 22(b) is the locally enlarged partial curve. Three specimens from LB-1.0B and two specimens each from LB-0.5B, LB-0.75B, and LB-1.5B were demonstrated. The P-δ curves show that the difference between specimens was within a reasonable range. The averaged P-δ curve is presented in Figure 23(a). As observed in Figures 22 and 23(a), all specimens to a certain extent were affected by concrete bleeding. We expect concrete bleeding to affect specimens with a larger bearing distance more than their counterparts. However, the LB-1.5B specimen suffered the least from concrete bleeding. We speculate this difference to have risen from the timing or expertise of concrete placement. To facilitate comparison, we assumed that all specimens were affected similarly by concrete bleeding. Therefore, LB-1.0B was used as the basis, and visual inspection revealed that the curve shifted 0.8 mm to the right for LB-0.5B, 1.2 mm to the right for LB-0.75B, and 1.8 mm to the right for LB-1.5B. Figure 23(b) illustrates the new curves.

Figure 22.

P-δ curve of the LB specimens.

Figure 23.

Mean P-δ curve of the LB specimens.

The P₅ and P₃₀ of the specimens were obtained using Figure 23(b) and are listed in Table 3. Figure 24 illustrates the P₅-L_b and P₃₀-L_b relationships. L_b had little influence on P₅ but a strong and positive influence on P₃₀. Figure 20 shows that A_jb has a strong and positive influence on P₅. Therefore, the jointed bearing area A_jb is the most critical parameter affecting P₅.

Figure 24.

Distribution of P₅ and P₃₀ of the LB specimens.

After the loading test, the out-of-plane deformation at the center of the steel tube wall was measured, which ranged from 2.0-3.3 mm. The deformation trend suggests the exertion of an outward normal force on the inner wall of the steel tube, which is essential for the development of friction force. In addition, Figure 24 reveals that a wider L_b results in a heavier P₃₀. The outward normal force and the wider L_b, which leads to a larger contact area, result in a larger friction force. Therefore, the difference between P₃₀ and P₅ is likely due to the friction force between the steel tube and the concrete.

Conclusions

In a CFT column, interfacial shear may exist between the tube and the concrete. When the interfacial shear force is small, direct bond between the concrete and the steel tube may have sufficient strength to transfer it. Otherwise, shear connectors or direct bearing devices may be needed, particularly for CFTs with larger cross-sectional dimension. However, due to the limited space inside the steel tube, welding shear connectors to the steel tube is difficult. In this study, a pilot test on five square CFT specimens were carried out to identify the most effective of placing a steel plate as a shear connector. It was found that a steel plate located at the corner of a square steel tube with a 45° angle to the tube wall had the highest load carrying capacity, and, this type of steel plate was adopted as a new type of shear connector for further development. Since the shear connector was placed at the corner of the tube, it is called corner shear connector. After the configuration of the corner shear connector was identified, push-out tests on 18 square CFT specimens were carried out to investigate the behavior of the corner shear connectors. The strength of the specimens was found to be independent of the cross-sectional area of the shear connector. Instead, the jointed bearing area, defined as the cross-sectional area of the shear connector combining the area enclosed by the shear connector and the steel tube, is the major parameter that is closely correlated with the strength of the specimens. Furthermore, the ultimate strength of the specimens is positively correlated with the bearing distance, which is the distance between the loading point and the bearing surface of the connector. However, the bearing distance has an insignificant effect on the strength of the specimen when the slippage between the concrete and steel tube is less than 5 mm. The test results also show that the ultimate strength of the corner shear connectors, which have a total jointed bearing area equaling 10.8% of the total concrete area in the steel tube, is higher than the nominal compressive strength of the concrete in the steel tube.

Footnotes

Acknowledgements

We would like to thank the Structural Laboratory of the Department of Civil and Construction Engineering, National Taiwan University of Science and Technology, for providing us with the venue and equipment for the tests.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Taiwan Building Technology Center via The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education in Taiwan. We also received a research grant from the Ministry of Science and Technology-subsidized special topic research project [MOST 106-2221-E-011-030] for producing the specimens.

ORCID iDs

Pin-Da Wu

Cheng-Cheng Chen

References

ACI Committee (2014) Building Code Requirements for Reinforced Concrete (ACI 318-14) and Commentary (318R-14). Farmington Hills, MI: American Concrete Institute.

Ahmadi

Naderpour

Kheyroddin

(2018) A proposed model for axial strength estimation of non-compact and slender square CFT Columns. Iranian Journal of Science and Technology, Transactions of Civil Engineering 43: 131–147.

Ahmadi

Naderpour

Kheyroddin

, et al. (2017) Seismic failure probability and vulnerability assessment of steel-concrete composite structures. Periodica Polytechnica Civil Engineering 61(4): 939–950.

ANSI/AISC 360-05 (2005) Specification for Structural Steel Buildings. Chicago, IL: American Institute of Steel Construction.

ANSI/AISC 360-16 (2016) Specification for Structural Steel Buildings. Chicago, IL: American Institute of Steel Construction.

ASTM A36/A36M-14 (2016) Standard Specification for Carbon Structural Steel. New York, NY: American National Standards Institute.

ASTM A572/A572M-15 (2016) Standard Specification for High-Strength Low-Alloy Columbium-Vanadium Structural Steel. New York, NY: American National Standards Institute.

Chen

Lin

(2009a) Behavior and strength of steel reinforced concrete beam-column joints with two-side force inputs. Journal of Constructional Steel Research 65: 641–649.

Chen

Hoang

(2016) Role of concrete confinement of wide-flange structural steel shape in steel reinforced concrete columns under cyclic loading. Engineering Structures 110: 79–87.

10.

Chen

Shen

(2018) Effects of steel-to-member depth ratio and axial load on flexural ductility of concrete-encased steel composite columns. Engineering Structures 155: 157–166.

11.

Chen

Zhan

(2022) Design of confinement reinforcement for encased composite columns. Engineering Structures 250(1–11): 113342.

12.

Chen

Suswanto

Lin

(2009b) Behavior and strength of steel reinforced concrete beam-column joints with single-side force inputs. Journal of Constructional Steel Research 65: 1569–1581.

13.

Furlong

(1967) Strength of Steel-Encased Concrete Beam Columns. Journal of the Structural Division 93(5): 113–124.

14.

Gan

Zhou

(2019) Axially loaded thin-walled square concrete-filled steel tubes stiffened with diagonal binding ribs. ACI Structural Journal 116(6): 265–280.

15.

Jamkhaneh

Ahmadi

Sadeghian

(2020) Simplified relations for confinement factors of partially and highly confined areas of concrete in partially encased composite columns. Engineering Structures 208(1–10): 110303.

16.

JIS STKR490 (2015) Carbon Steel Square and Rectangular Tubes for General Structure. China: China Good Quality Precision Steel Tube Supplier.

17.

Kumar

Chaudhary

(2019) Effect of reinforcement detailing on performance of composite connections with headed studs. Engineering Structures 179: 476–492.

18.

Leon

Kim

Hajjar

(2007) Limit state response of composite columns and beam-columns Part I: Formulation of Design Provisions for the 2005 AISC Specification. Engineering Journal 44: 341–358.

19.

Pallarés

Hajjar

(2010) Headed steel stud anchors in composite structures, Part I: Shear. Journal of Constructional Steel Research 66(2): 198–212.

20.

Roeder

Cameron

Brown

(1999a) Composite action in concrete filled tubes. Journal of Structural Engineering 125(5): 477–484.

21.

Roeder

Chmielowski

Brown

(1999b) Shear connector requirements for embedded steel sections. Journal of Structural Engineering 125(2): 142–151.

22.

Shen

Chung

(2017) Experimental investigation into stud shear connections under combined shear and tension forces. Journal of Constructional Steel Research 133: 434–447.

23.

Tao

Song

, et al. (2016) Bond behavior in concrete-filled steel tubes. Journal of Constructional Steel Research 120: 81–93.

24.

Wang

Yao

, et al. (2018) Static behavior of grouped large headed stud-UHPC shear connectors in composite structures. Composite Structures 206: 202–214.

25.

Xue

Liu

, et al. (2012) Static behavior of multi-stud shear connectors for steel-concrete composite bridge. Journal of Constructional Steel Research 74: 1–7.

26.

Zhou

Gan

Zhou

(2019) Improved Composite Effect of Square Concrete-Filled Steel Tubes with Diagonal Binding Ribs. Journal of Structural Engineering 145(101–12): 04019112.

New type of shear connector for square concrete-filled tubular columns

Abstract

Keywords

Introduction

Pilot tests on shear connectors

Loading test of corner shear connectors

Specimen design and fabrication

Test setup and application of load

Results and discussion

Effect of concrete bleeding

Effect of steel bearing area Asb

Effect of jointed bearing area Ajb

Effect of bearing distance Lb

Conclusions

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

ORCID iDs

References

Effect of steel bearing area A_sb

Effect of jointed bearing area A_jb

Effect of bearing distance L_b