You are on page 1of 12

Journal of Constructional Steel Research 65 (2009) 20872098

Contents lists available at ScienceDirect

Journal of Constructional Steel Research


journal homepage: www.elsevier.com/locate/jcsr

Review

Test of shear transfer enhancement in symmetric cold-formed steelconcrete composite beams


J. M.Irwan a, , A.H. Hanizah b , I. Azmi b
a b

Faculty of Civil And Environmental Engineering (FKAAS), Universiti Tun Hussein Onn Malaysia (UTHM), Parit Raja, 86400 Batu Pahat, Johor, Malaysia Faculty of Civil Engineering (FKA), Universiti Teknologi MARA Malaysia (UiTM), 40450 Shah Alam, Selangor, Malaysia

article

info

abstract
An experimental programme was conducted to study the strength and behaviour of a bent-up tab shear transfer enhancement. Sixty eight push-out test specimens of cold-formed steel lipped channel sections (CFS) embedded in concrete planks which make use of the new proposed shear transfer enhancement called bent-up triangular tab shear transfer (BTTST) were tested in this programme. This paper summarises the results of the experimental study and develops an expression in order to predict the innovative shear transfer enhancement mechanism, BTTST in a new type of precast composite beam comprising CFS embedded in a precast concrete plank. The results show that specimens employed with shear transfer enhancements increase the shear capacities of the specimens as compared to those relying only on a natural bond between cold-formed steel and concrete. In this investigation, BTTST provided better performance in terms of strength. Furthermore the effects of different angles and sizes for BTTST, different thickness for CFS and different strengths for concrete are also discussed. It is concluded that the proposed shear transfer enhancement has sufficient strength and is also feasible. 2009 Elsevier Ltd. All rights reserved.

Article history: Received 6 November 2007 Accepted 10 July 2009 Keywords: Cold-formed steel Composite beams Push-out test Shear transfer mechanisms

Contents 1. 2. Introduction........................................................................................................................................................................................................................2088 Research programme .........................................................................................................................................................................................................2088 2.1. Description of test..................................................................................................................................................................................................2088 2.1.1. Test specimen .........................................................................................................................................................................................2088 2.1.2. Test programme......................................................................................................................................................................................2088 2.2. Instrumentation and test procedures...................................................................................................................................................................2088 Test results and discussion ................................................................................................................................................................................................2089 3.1. Failure mechanisms ...............................................................................................................................................................................................2090 3.1.1. Failure mode 1: Bonding failure ............................................................................................................................................................2091 3.1.2. Failure mode 2: Concrete crushing-splitting ........................................................................................................................................2091 3.2. Analysis of results ..................................................................................................................................................................................................2092 3.2.1. Effect of BTTST ........................................................................................................................................................................................2092 3.2.2. Effect of concrete strength .....................................................................................................................................................................2092 3.2.3. Effect of cold-formed steel section thickness .......................................................................................................................................2092 3.2.4. Effect of BTTST angle ..............................................................................................................................................................................2093 3.2.5. Effect of BTTST size .................................................................................................................................................................................2095 Development of a new equation: BTTST enhancement in CFS-concrete composite beams .........................................................................................2095 4.1. General form ..........................................................................................................................................................................................................2095 4.2. Regression analysis ................................................................................................................................................................................................2096 Conclusions.........................................................................................................................................................................................................................2096 Acknowledgments .............................................................................................................................................................................................................2097 References...........................................................................................................................................................................................................................2097

3.

4.

5.

Corresponding author. Tel.: +60 197758186; fax: +60 355435275. E-mail addresses: irwan_kuittho@yahoo.com, irwan@uthm.edu.my (J. M.Irwan), hanizah_ah@yahoo.com (A.H. Hanizah), azmii716@yahoo.com (I. Azmi). URLs: http://www.uthm.edu.my (J. M.Irwan), http://www.fka.uitm.edu.my (A.H. Hanizah). 0143-974X/$ see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jcsr.2009.07.008

2088

J. M.Irwan et al. / Journal of Constructional Steel Research 65 (2009) 20872098

2.1. Description of test List of symbols Lf Ls Ptab fcu E collar length of BTTST (refer to Fig. 6) (mm) span length of BTTST (refer Fig. 6) (mm), the predicted ultimate load per BTTST (N). cube compressive strength of concrete (N/mm2 ). modulus of elasticity of concrete (N/mm2 ). constant to be referred to Table 3. constant to be referred to Table 3. bearing area of BTTST (refer to Fig. 14) shear area of BTTST (refer to Fig. 14) thickness of CFS (mm). yield strength of the BTTST (N/mm2 ) angle of BTTST (refer Fig. 6) (degree) 2.1.1. Test specimen Fig. 3 shows the details of cross-sections for the push-out test specimen. The basic specimen configuration and construction are based on BS5400 [9] and Eurocode4 (EC4) [10] with slight modifications for the allowance of the parameter. CFS I-section beam formed by back-to-back lipped channels was used with the flanges cast into a 300 mm wide 90 mm depth 550 mm height concrete slab. One layer of 100 mm square welded wire fabric steel reinforcements with diameter 8 mm was provided in the concrete slab. A recess of 50 mm in height was provided between the bottom of the concrete slab and lower end of the cold-formed steel section to allow for slip during testing. Each specimen had eight shear transfer enhancements. However, control specimens were tested without shear transfer enhancement. Both sides of the flanges of the I-beam for each specimen were embedded in the concrete slab to form a composite system. 2.1.2. Test programme The testing was done in two different phases. A total of 48 push out test specimens in Phase 1 were tested to gain a better understanding of the newly proposed type called bent-up triangular tab shear transfer (BTTST) enhancement. These tests were performed to investigate the effects of having different parameters as follows: a. Types of shear transfer enhancement. Control specimen without shear transfer enhancement and specimen with newly proposed type called bent-up triangular tab shear transfer (BTTST) are as shown in Figs. 4 and 2 respectively. The first parameter evaluated was the effect of BTTST in the composite beam. A total of 48 specimens were performed; 24 for control and the other 24 with BTTST. These tests were performed to verify that BTTST would increase the ultimate strength of specimens. A BTTST size of 30 mm 30 mm and angle of 60 was used for all BTTST specimens. b. Strength of concrete. The second parameter investigated was on the effect of BTTST strength with different grades of concrete. The concrete strength used were 25 N/mm2 , 30 N/mm2 , 35 N/mm2 , 40 N/mm2 , 50 N/mm2 and 60 N/mm2 . Based on Table 7.1 and 7.2 in BS8110, Part 2 [11], strengths of normal-weight concrete are ranging from 20 N/mm2 to 60 N/mm2 . c. Thickness of CFS. The third parameter investigated was on the effect of the thickness of CFS. This test is to determine the strength of BTTST by using different thicknesses of CFS. This was accomplished by using two thicknesses of CFS; 1.9 mm and 2.4 mm, available in the steel manufacturing factory in the Malaysian market. The programme in Phase 2 consists of twenty push-out test specimens. Specimens were tested to study the effect of having the parameters below: a. Shear transfer enhancement: BTTST (Fig. 2) or Lakkavalli and Liu bent-up (LYLB) (Fig. 5) [6]. b. Dimensions of BTTST, Lf Ls (Fig. 6): 25 mm 25 mm based on LYLB [6] or two other sizes of 20 mm 20 mm or 30 mm 30 mm. c. Angle of BTTST, (Fig. 6): 30 , 45 or 60 . Angle 45 is based on LYLB [6]. 2.2. Instrumentation and test procedures The vertical slip between the slab and the cold-formed steel beam was measured by two displacement transducers (LVDT). Load and slip measurements were recorded with a computerised

1 2
Ab As t fy

1. Introduction Cold-formed steel (CFS) sections, usually between 1.2 and 3.2 mm thickness [1], have been recognised as an important contributor to environmentally responsible and sustainable structures in developed countries, and CFS framing is considered as a sustainable green construction material for low rise residential and commercial buildings. Recent studies on composite beams with CFS had been reported by others [27]. As illustrated in Fig. 1, these studies focused on the use of CFS for beam structures. The figure shows that all researchers agree on reporting the usage of CFS beams and composites with concrete. However, there is still a lack of data and information on the behaviour and performance of CFS beams in composite construction. The use of CFS has been limited to structural roof trusses and a host of non-structural applications [8]. One limiting feature of CFS is the thinness of its section that makes it susceptible to torsional, distortional, lateraltorsional, lateraldistortional and local buckling. Hence, a reasonable solution is resorting to a composite construction of structural CFS section and reinforced concrete deck slab, which minimises the distance from the neutral-axis to the top of the deck and reduces the compressive bending stress in the CFS sections. Also, by arranging two CFS channel sections back-toback restores symmetricity and suppresses lateraltorsional and to a lesser extent, lateraldistortional buckling. The two-fold advantages promised by the system, promote the use of CFS sections in a wider range of structural applications. An efficient and innovative floor system of built-up cold-formed steel (CFS) sections acting compositely with a concrete deck slab has been developed to provide an alternative composite system for floors and roofs in buildings. The system, called Precast ColdFormed Steel-Concrete Composite System, is designed to rely on composite action between the CFS sections and a reinforced concrete deck where shear forces between them are effectively transmitted via another innovative shear transfer enhancement mechanism called a bent-up triangular tab shear transfer (BTTST) as shown in Fig. 2. 2. Research programme Sixty eight push-out test specimens were tested to study the behaviour and capacity of BTTST. The test parameters include concrete compressive strength, concrete modulus of elasticity, CFS strength, dimension of BTTST (Lf , Ls , angle) and CFS thickness. Based on the results of the parametric study, expression for predicting the capacity of BTTST would be derived, which will then fulfill one of the objectives in this study.

J. M.Irwan et al. / Journal of Constructional Steel Research 65 (2009) 20872098

2089

Fig. 1. Previous studies on cold-formed steelconcrete composite beams.

Fig. 2. Specimen with bent-up triangular tab shear transfer (BTTST) enhancement.

data acquisition system. Transducers were placed vertically at both sides of the cold-formed steel web specimen. The locations of the transducers and instrumentation setup are shown in Fig. 7. The average of the two readings was used in the analysis. Thin plywood with 3 mm thickness was placed beneath the slab and on the upper part of the beam to level the surface. A steel plate of 30 mm thickness was placed on the upper plywood to receive the jack. Steel plates were fastened to the loaded end of the cold-formed steel to prevent local buckling. The specimen was placed vertically in the Universal Testing Machine (UTM IPC 1000). It was loaded by an existing 1000 kN jack. The test procedure was based on EC4 [10] with slight modifications for an allowance of the testing machine software. The load was applied up to 40% of the estimated failure load. The load was cycled 25 times between 5% and 40% of the expected failure load. The loading under slip control mode was applied. The rate of loading was kept constant at 0.0095 mm/s.Testing was discontinued when the specimen failed to take an additional load or when a significant load drop had occurred. All data were automatically

recorded by the Control and Data Acquisition System (CDAS) of the testing machine software (UTS010 Software). Data were scanned initially at every load increment until failure. After the test, all recorded data were transferred to an Excel Programme for analysis. After the test, the specimen was dismantled to investigate the condition of the shear transfer enhancement wherever possible. 3. Test results and discussion The average properties of concrete that correspond to the pushout specimen are listed in Table 1. This includes the concrete compressive strength, fcu and modulus of elasticity, Ec . The tensile properties of CFS and welded wire fabric reinforcement are presented in Table 2. The ultimate load, Pu for each push-out specimen is listed in Table 1. The ultimate loads show that there is no significant difference between the replicate specimens of a test group. The tabulated data from Table 1 shows that the difference between Pu for replicate specimens and mean Pu is less than 10%. Hence there is no further test required. This statement is

2090

J. M.Irwan et al. / Journal of Constructional Steel Research 65 (2009) 20872098

Fig. 3. Layout for push-out test specimen.

Fig. 4. Control specimen without shear transfer enhancement.

in EC4 [10], which states that if there are three tests on nominally identical specimens carried out and the deviation of any individual test result from the mean value obtained from all tests does not exceed 10%, no further test of the same kind should be made. In general, the ductility calculated varies from 0.47 mm to 1.80 mm as shown in Table 1. These specimens exhibited brittle behaviour since all ductility values are less than 6 mm as defined in EC4 [10]. This behaviour agrees with the plotted graphs in experimental programme done by Hanaor [4] for screwed and welded specimens. The calculated ductility values in his study gave

1.2 mm and 2 mm for screwed and welded specimens respectively. Referring to Table 1, the specimen with a thinner CFS shows a slightly more ductile behaviour. Fig. 8(a) shows the typical loadslip curves for push-out specimens while Fig. 8(b) shows idealised curves to represent the general behaviour which is divided into three regions named as Region 1, Region 2 and Region 3. Region 1 indicates that the capacity of the specimen is solely based on bonding between the concrete and the CFS. Region 2 indicates that the capacity of the specimen is due to a combination of the bonding and BTTST capacity, measured from ultimate capacity of the control specimen to the second slope of the stiffness plotted in the BTTST specimen while the third region indicates the capacity of the BTTST after there is no bonding between the concrete and the CFS. Hence application of BTTST can increase the load capacity as well as the ductility of specimens. Based on the graphs, at elastic condition, a linear increment of slip with the increment of load can be seen. The contribution of the capacity from bonding and BTTST in the first stiffness results in a higher load as compared to the control specimen indicated in second region. 3.1. Failure mechanisms The experimental results related to push-out tests are presented, along with a description of observed failure mechanisms. The failure modes observed in all push-out test specimens can broadly be classified into two types, as bonding failure or concrete crushing-splitting.

J. M.Irwan et al. / Journal of Constructional Steel Research 65 (2009) 20872098

2091

Fig. 5. Lakkavalli and Liu Bent-up tabs shear transfer (LYLB).

Fig. 6. Dimensions and angle of BTTST.

3.1.1. Failure mode 1: Bonding failure The characteristic feature of this failure mechanism is the downward slip of the cold-formed steel without crushing of concrete. All specimens without shear transfer enhancement exhibited this kind of failure mode. Fig. 9 shows no concrete cracking at both longitudinal and transverse directions. This phenomenon is due to the free slip between concrete and steel. The shear resistance is only contributed by the bonding between concrete and cold-formed steel. Failure of specimens without shear transfer enhancement depends on the bonding between the interface between the cold-formed section and concrete. Fig. 9(b) shows slips of the CFS after the push-out test. 3.1.2. Failure mode 2: Concrete crushing-splitting In specimens with bent-up tab shear transfer enhancement, failure was initiated by concrete crushing followed by splitting of the concrete slabs. (a) Specimens with Lakkavalli and Liu bent-up tabs (LYLB) Longitudinal cracks in the concrete at all edges were found in the specimen with a LYLB as shown in Fig. 10. This

phenomenon indicated that bent-up shear connectors in the cold-formed steel sections tried to prevent the slip between concrete and cold-formed steel, where a high concentration of stress developed at the shear connectors which induced concrete cracking and crushing. After removal of the damaged concrete, it was observed that the LYLB is still in position without yielding as shown in Fig. 11. This implies that the development of high compressive stress surrounding the bentup tab induces concrete crushing at that region. (b) Specimens with bent-up triangular tab shear transfer (BTTST) During testing, it was observed that all specimens exhibited substantial inelastic deformation before failure. At ultimate load, there was no evidence of sudden failure. With further deformation accompanied by a decrease in the load, failure was evident by crushing of the concrete. Fig. 12 shows a typical cracking pattern in push-out test specimens. It was observed that most of the specimens were experiencing the same crack pattern. Diagonal concrete cracking (herringbone pattern) at the edges were obtained in the specimens with BTTST. This is due to the resistance of BTTST located in the transverse direction. After reaching the ultimate load, the bonding of the interface between the CFS and concrete was lost. Fig. 13 shows the shear transfer enhancement after removal of concrete slab in the tested specimen. Deformation of the BTTST element was noticed near the BTTST collar followed by maximum compressive stress reached at the concrete around the tabs. The shear transfer enhancement failed due to concrete crushing in the bearing zone. The shear transfer enhancement resisted the shear force and these forces caused high tensile stress in the cold-formed steel failure zone (around shear area shown in Fig. 14).

Fig. 7. Push-out test arrangement.

2092

J. M.Irwan et al. / Journal of Constructional Steel Research 65 (2009) 20872098

300

250
G25B1-2.4 G25B2-2.4

Load (kN)

200

150

G25B3-2.4 G25C1-2.4

100

G25C2-2.4 G25C3-2.4

be attributed to better interlocking at the cold-formed steel-toconcrete interface which exists in the specimens with BTTST. The bearing area of BTTST is larger than the tabs perpendicular to the flange of the cold-formed steel. In LYLB model, the shear transfer is parallel with the longitudinal of concrete which provides a smaller bearing area for resisting concrete slip compared with the model with BTTST. The ductility values determined are also given in Table 1. As seen in the table, measures of BTTST can increase the members ductility. The ductility increase ranges about 3.5% for CFS 1.9 mm thick specimens and 8.5% for 2.4 mm. It seems that BTTST is more effective in increasing a members ductility. 3.2.2. Effect of concrete strength Concrete strength is one of the variables considered in this experimental program. Table 1 shows ultimate loads of tested specimens up to concrete grade 60 indicted as G25 to G60. The increment of concrete grade has increased the load capacity up to 44%. All specimens exhibit similar patterns of failure which are governed by concrete crushing before the concrete splits into two parts. As shown in Table 1, the linearity of these two variables between the increments of concrete grades to the increment of load capacity is seen as exponentially related. For instance, the increment of 5MPa of grade 25 to 30 increases the load capacity by 6.5% while the increment of 10MPa of grade 25 to grade 35 has increased the load by 19%. The effect of concrete strength shows that the ultimate load has increased with the increment of concrete strength. As the concrete grade is increased, the load carrying capacity of the push out specimen increases. The findings from previous researchers show that shear strength of connectors can be increased by increment of concrete grade and this statement is supported by Hosain and Pashan [12] in their experimental programme using channel shear connectors, Veldanda and Hosain [13] and Oguejiofor and Hosain [14] with their studies on perfobond rib shear connectors, Vianna et al. [15] using T-perfobond shear connectors and Lam [16], Ellobody and Young [17], Lam and Ellobody [18], Shim et al. [19] and Li An and Cederwall [20] using headed stud connectors. According to Mujagic et al. [21], Oguejiofor and Hosain [14] and Ollgaard et al. [22], for concrete related failures, ultimate the capacity of a shear connector is proportional to fc . Fig. 15 presents the mean ultimate loads versus fc for specimens in Phase 1. The increase in load with the square root of compressive strength of concrete in Fig. 15 is seen to be approximately linear. 3.2.3. Effect of cold-formed steel section thickness The load capacities for specimens with CFS thickness 1.9 mm and 2.4 mm are shown in Table 1. The specimens with 2.4 mm CFS thickness carried only 2.8% (specimens with G25, Phase 1), 3.4% (specimens with G35, Phase 1) and 9% (specimens with G35, Phase 2) higher load than that with a CFS thickness of 1.9 mm. This was expected, since the CFS thickness does not have a major influence when failure is concrete related. The mode of failure indicates that the specimens failed due to concrete crushing and splitting. In general, the influence of CFS thickness is only significant when failure occurred due to the triangular tabs connectors bending but was minimal for a concrete crushing type of failure. From the experimental work, the specimen with thickness 2.4 mm achieves a higher ultimate load as compared to a 1.9 mm thickness. This result is also similar to the work done by Lakkavalli and Liu [6]. The shear area (Fig. 14) of the triangular tab increases when the thickness is increased. The shear areas are 57 mm2 and 72 mm2 (Phase 1) and 47.5 mm2 and 60 mm2 (Phase 2) for the thickness of 1.9 mm and 2.4 mm respectively. The shear area acts as a resistance against bending for the triangular tabs. The bending capacity of the shear transfer enhancement increases with increasing shear area.

50

Mean BTTST (G25) Mean Control (G25)

0 0 2 4 Slip (mm) 6 8

300 250 200 150 100 Region 1 50 0


BTTST Control

Load (kN)

Region 3

Region 2

3 4 Slip (mm)

Fig. 8. (a) Typical and (b) Idealised load-slip curves for push-out specimen.

3.2. Analysis of results 3.2.1. Effect of BTTST The capacity, stiffness and slip of the composite beams can be affected by the shape of the shear transfer enhancement. Shear transfer enhancement plays an important role in the function of joining two different materials of a composite system. A typical comparison between load capacity of specimen with and without BTTST is shown in Fig. 8(a). Specimens with shear transfer enhancement illustrate a significant increment in load capacity. As seen in Table 1, load capacities for specimens with BTTST are relatively high as compared to control specimens, which are increased by 91% to 135%. A comparison of the capacities of LYLB and BTTST indicates that BTTST results in a higher capacity. The ultimate capacity of specimen with BTTST is 12% to 16% higher than that of specimen with LYLB. This shows a significant increment of load capacity due to the effect of BTTST. The triangular tabs provide resistance between the concrete and the CFS to prevent slip between the concrete and the CFS. This is due to triangular tabs which provide a higher resistance in the direction of concrete width. The resistance of BTTST is affected by its bearing area and the schematic diagram is shown in Fig. 14. After the specimen reaches the maximum capacity, the existing shear tabs provide additional resistance due to presence of concrete slabs at both sides before the slabs failed simultaneously by crushing and praying of the concrete in front and at back face of BTTST. Overall, specimens with BTTST provide higher ultimate loads as compared to specimens without enhancement or with LYLB. These incomparably more desirable results can

J. M.Irwan et al. / Journal of Constructional Steel Research 65 (2009) 20872098

2093

(a) Before loaded.

(a) After loaded. Fig. 9. Specimen without shear transfer enhancement.

Fig. 10. Typical cracking mode in push-out test specimens with LYLB.

Fig. 11. Shear transfer enhancement of LYLB after removal of concrete slab.

3.2.4. Effect of BTTST angle Referring to Table 1, the ultimate capacity of specimen with angle 60 BTTST is higher than that of specimens with angle 45 and 30 . It is calculated at about 2.6% to 9.6% and 18% to 27.3%

higher, respectively. The ultimate capacities of specimens with an angle of 45 are 8% to 24% higher than specimens with an angle of 30 . The results indicate that a higher Pu can be achieved if the BTTST angle is increased. This is because a large angle of

2094

J. M.Irwan et al. / Journal of Constructional Steel Research 65 (2009) 20872098

Table 1 Average concrete properties and push-out test results. Phase 1 Specimen no. Description of specimen Average concrete properties Ultimate load, Pu (kN) Average of Ptab (kN) (per tab) Average of initial slip, I (mm) Average of Slip capacity, u (mm) Average of ductility, u i (mm)

Compressive Modulus First strength, of 2 fcu (N/mm ) elasticity, Ec (kN/mm2 ) P1C- 1, 2, 3 P1B- 1, 2, 3 P1C- 4, 5, 6 P1B- 4, 5, 6 P1C- 7, 8, 9 P1B- 7, 8, 9 P1C- 10, 11, 12 P1B- 10, 11, 12 P1C- 13, 14, 15 P1B- 13, 14, 15 P1C- 16, 17, 18 P1B- 16, 17, 18 P1C- 19, 20, 21 P1B- 19, 20, 21 P1C- 22, 23, 24 P1B- 22, 23, 24 Phase 2 P2C- 1, 2, 3 P2B- 1, 2, 3 P2C- 4 P2L- 1 P2B- 4 P2B- 5 P2B- 6 P2B- 7 P2B- 8 P2C- 5 P2L- 2 P2B- 9 P2B- 10 P2B- 11 P2B- 12 P2B- 13 G25/C/t2.4 G25/B/t2.4/D25/ 45 G35/C/t1.9 G35/L/t1.9/D25/ 45 G35/B/t1.9/D25/ 45 G35/B/t1.9/D25/ 30 G35/B/t1.9/D25/ 60 G35/B/t1.9/D20/ 45 G35/B/t1.9/D30/ 45 G35/C/t2.4 G35/L/t2.4/D25/ 45 G35/B/t2.4/D25/ 45 G35/B/t2.4/D25/ 30 G35/B/t2.4/D25/ 60 G35/B/t2.4/D20/ 45 G35/B/t2.4/D30/ 45 25.54 25.54 36.62 36.62 36.62 36.62 36.62 36.62 36.62 36.62 36.62 36.62 36.62 36.62 36.62 36.62 24.32 24.32 26.95 26.95 26.95 26.95 26.95 26.95 26.95 26.95 26.95 26.95 26.95 26.95 26.95 26.95 122 248 141 253 295 237 302 188 308 164 290 321 297 352 235 358 G25/C/t1.9 G25/B/t1.9/D30/ 60 G25/C/t2.4 G25/B/t2.4/D30/ 60 G30/C/t1.9 G30/B/t1.9/D30/ 60 G35/C/t1.9 G35/B/t1.9/D30/ 60 G35/C/t2.4 G35/B/t2.4/D30/ 60 G40/C/t1.9 G40/B/t1.9/D30/ 60 G50/C/t1.9 G50/B/t1.9/D30/ 60 G60/C/t1.9 G60/B/t1.9/D30/ 60 26.67 26.67 26.67 26.67 32.83 32.83 36.51 36.51 36.51 36.51 43.36 43.36 51.31 51.31 62.35 62.35 24.72 24.72 24.72 24.72 25.63 25.63 27.16 27.16 27.16 27.16 27.52 27.52 27.33 27.33 26.98 26.98 131 245 123 258 140 262 137 295 145 308 150 318 142 333 144 372

Second

Third

126 248 137 245 134 255 142 289 143 301 144 313 146 338 147 329

123 247 138 260 136 272 143 298 141 302 146 322 147 329 141 367

15.00 15.21 15.79 19.17 20.09 21.38 23.54 26.50

0.89 0.93 0.69 1.32 1.05 0.81 1.14 1.19

2.25 2.26 2.11 2.45 2.04 1.84 2.19 1.98

1.36 1.33 1.42 1.13 0.99 1.03 1.05 0.79

123 253

120 250

16.08 14.08 19.26 12.12 20.23 5.94 20.91 15.74 19.67 16.68 23.52 8.89 24.32

0.84 0.50 0.37 0.73 0.25 0.32 0.32 0.62 0.73 1.39 2.12 0.72 0.95

1.70 1.92 1.84 2.53 1.75 1.56 1.55 1.09 1.55 2.55 3.18 1.36 1.67

0.86 1.42 1.47 1.80 1.75 1.24 1.23 0.47 0.51 1.16 1.06 0.64 0.72

G indicates the grade of concrete 25 N/mm2 , 30 N/mm2 , 35 N/mm2 , 40 N/mm2 , 50 N/mm2 or 60 N/mm2 ; C, B, L indicates the types of shear transfer enhancement: C:control, B: BTTST or L: LYLB ; t indicates the thickness of CFS: 1.9 mm or 2.4 mm; D indicates the dimension of shear transfer enhancement: 20 mm 20 mm, 25 mm 25 mm or 30 mm 30 mm; indicates the angle of shear transfer enhancement: 30 , 45 or 60 .

Fig. 12. Typical cracking mode in push-out test specimens.

J. M.Irwan et al. / Journal of Constructional Steel Research 65 (2009) 20872098

2095

Fig. 13. Shear transfer enhancement of BTTST after removal of concrete slab.

380 360 Mean Load (kN)


Fig. 14. Diagram of bearing and shear area of BTTST.

340 320 300 280 260 240 220

BTTST produces a large bearing area to prevent concrete slip. As discussed earlier, the bearing area functions as a resistance to prevent concrete slip. Hence, a higher resistance is provided by BTTST with larger angle when the specimen is subjected to load. The bearing area of the BTTST angle of 30 , 45 and 60 are 156 mm2 , 221 mm2 and 271 mm2 respectively. Fig. 16 shows the bearing area of BTTST and the formulae for its calculation. 3.2.5. Effect of BTTST size The results of specimens with different sizes of BTTST are shown in Table 1 (Phase 2). These specimens are similar in every aspect except for the size of the shear transfer enhancement either 20 mm 20 mm, 25 mm 25 mm or 30 mm 30 mm and the thickness of CFS either 1.9 mm or 2.4 mm. Referring to that table, it can be seen that the capacity of the BTTST enhancement is increased with an increase of BTTST size. BTTST capacities are 188.05 kN, 294.65 kN and 307.79 kN at slips of 0.62 mm, 0.91 mm and 0.57 mm respectively for the push-out specimens with CFS thickness 1.9 mm. The ultimate loads for specimens with size 30 mm 30 mm increase 4.5% to 11.6% than specimens with BTTST size 25 mm 25 mm and 52.5% to 63.7% than specimens with BTTST size 20 mm 20 mm. Specimens with BTTST size 25 mm 25 mm increase 36.7% to 56.7% than specimens with BTTST size 20 mm 20 mm. An increased size of BTTST would increase the bearing area. As previously discussed, the increased bearing area tends to increase the shear resistance. 4. Development of a new equation: BTTST enhancement in CFSconcrete composite beams In order to obtain a mathematical relationship between the ultimate shear strength, Pu of BTTST and the variables which have been discussed earlier, multiple regression analysis based on the 28 results of BTTST push-out test specimens from Phase1 and 2 experimental programs were made.

200

5.5

6.5 fcu

7.5

8.5

Fig. 15. Graph of load versus

fc .

Fig. 16. Bearing area of BTTST.

4.1. General form The first consideration in developing the equation is the contribution factor from concrete properties. The relationship between the capacity of BTTST and fc in this study has been described in Section 3.2.2. It shows an almost linear relationship. Meanwhile, the main properties of the concrete are the compression strength, fcu and modulus of elasticity, E. By considering the properties of concrete, the general expression is, Ptab = 1 fcu E (1)

where Ptab is the predicted ultimate load per BTTST (N), fcu is compressive strength of concrete (N/mm2 ), E is modulus of elasticity of concrete (N/mm2 ) and 1 is a constant to be referred to in Table 3. The bearing area, Ab of the connector is also as one of the important factors contributing to the shear capacity of the connector. As discussed earlier, the bearing area functions as a resistance to prevent concrete slip. Ab is affected by its size and angle of BTTST.

2096

J. M.Irwan et al. / Journal of Constructional Steel Research 65 (2009) 20872098

Table 2 Material Properties of CFS and welded wire fabric reinforcement. Material 203 76 19 CFS 203 76 21 CFS Welded wire fabric reinforcement Yield strength, py (N/mm2 ) 529 533 607 Ultimate tensile strength, pu (N/mm2 ) 569 564 626

Table 3 Results of regression analysis using SPSS. Model 1 2 3 General equation Ptab = 1 Ab fcu E Ptab = 2 As fy Ptab = 1 Ab fcu E + 2 As fy

1
0.052 0.012

2
0.605 0.466

Correlation 0.981 0.994 0.995

The increment of the angle and size of BTTST increase value of Ab and therefore a higher Ptab should be achieved. Thus, Ab is applied by multiplying itself with expression 1 to produce the following expression,
Ptab Predicted (kN)

30.0

25.0

Ptab = 1 Ab

fcu E .

(2)

20.0

The other factor that affects Ptab is the shear area, As of BTTST. Shear area acts as a resistance for the triangular tabs from bending. The bending capacity of the BTTST increases with increase of shear area. Shear area, As of the BTTST as shown in Fig. 14 is obtained by multiplying collar length, Lf (refer to Fig. 6) and thickness of CFS, t. Meanwhile, the discussion in Section 3.2.3 shows that increasing the CFS thickness would increase the shear capacity of the BTTST. Therefore, it is appropriate to include this variable into expression 2 to produce the following expression, Ptab = 1 Ab fcu E + 2 As (3)

15.0

10.0

5.0

0.0 0

10

where 2 is constant to be referred to Table 3. Another factor to be considered is the strength of CFS, fy . It has to be applied by multiplying with its shear area, As. Thus, the following form of the expression is obtained, Ptab = 1 Ab fcu E + 2 As fy . (4)

15 20 Ptab Test (kN)

25

30

Fig. 17. Comparison between tested values of shear enhancement strength and those predicted by Eq. (7).

Based on the formula for bearing area and shear area of the BTTST discussed earlier, (see Fig. 16), Eq. (6) becomes: Ptab = 0.01Lf Ls Sin fcu E + 0.5Lf tfy (7)

Therefore, there are 3 models of regression that have been tried in order to obtain the most accurate equation for predicting Ptab of the BTTST. Referring to Table 3, Model 1 is depending on the factor from concrete properties. Model 2 is depending on the factor from properties of the CFS, and Model 3 is the combination of Model 1 and Model 2. 4.2. Regression analysis In order to compare the results of push-out test specimens from different series of concrete strength, Ptab values are normalised to reflect the differences in concrete strength using the following expression [13,23]: Normalised v alue = (Measured v alue) fcu,mean fcu
0.5

(5)

where fcu,mean is the average concrete strength of all specimens and fcu is the concrete strength of the specimen in question. Results from regression analysis using SPSS are summarised in Table 3. The linear correlation relationship between Model 1, 2 and 3 and independent variables are shown in that table. The largest correlation is obtained with Model 3, which considers all independent variables. Based on this result, Model 3 is used in this study to develop an equation for predicting Ptab of the BTTST. Based on the results and by substituting values in the equation, the ultimate shear capacity prediction model for the BTTST takes the following equation: Ptab = 0.012Ab fcu E + 0.466As fy . (6)

where Lf is collar length of the BTTST (refer Figs. 6 and 16) (mm), Ls is span length of the BTTST (refer Figs. 6 and 16) (mm), is angle of the BTTST (degree) and t is thickness of the CFS (mm). The results of normalised ultimate shear strength of the BTTST from push-out test and those predicted by Eq. (7) for 28 specimens included in the above analysis are listed in Table 4. The ratios of test value over predicted values are concentrated within 0.86 to 1.16. It is obvious that Eq. (7) is reasonably in agreement with the observed test values. The average absolute difference between the test values and predicted values is found to be 8.07%. The average arithmetic mean of the tested/predicted ratio () of this equation is 0.9954. The standard deviation ( ) and the coefficient of variation (CV) for the proposed equation are 0.09682 and 9.7%, respectively. Fig. 17 is prepared to provide a clearer observation of the performance of Eq. (7). This figure plots the comparison between tested values of BTTST strength and those predicted by Eq. (7) for 28 push-out test specimens. 5. Conclusions In this study, a new type of shear transfer enhancement called BTTST has been proposed. BTTST enhancement was employed on the surface of the flange embedded in the concrete to provide shear transfer capacity. Sixty eight (68) push-out specimens were tested to evaluate the strength and behaviour of the BTTST. Besides that, this investigation has resulted in the derivation of an equation for

J. M.Irwan et al. / Journal of Constructional Steel Research 65 (2009) 20872098 Table 4 Test and predicted ultimate shear strength of BTTST results. Specimen no. P1B- 1 P1B- 2 P1B- 3 P1B- 4 P1B- 7 P1B- 8 P1B- 9 P1B- 10 P1B- 11 P1B- 12 P1B- 13 P1B- 14 P1B- 15 P1B- 16 P1B- 17 P1B- 18 P1B- 19 P1B- 20 P1B- 21 P1B- 22 P1B- 23 P1B- 24 P2B- 1 P2B- 3 P2B- 9 P2B- 5 P2B- 10 P2B- 8 Normalised Ptab From test (kN) 16.9 18.1 18.4 20.0 16.3 16.2 18.2 20.0 18.7 19.7 20.7 20.0 20.4 19.6 19.7 20.5 20.4 20.5 19.5 22.1 17.7 21.9 19.2 19.6 19.9 12.3 16.9 21.2 Ptab Using Eq. (7) (kN) 18.8 18.8 18.8 22.9 18.9 18.9 18.9 19.0 19.0 19.0 23.1 23.1 23.1 19.0 19.0 19.0 19.0 19.0 19.0 19.0 19.0 19.0 18.1 18.1 18.2 14.1 17.6 18.3

2097

Ratio of Test/Predicted 0.90 0.96 0.98 0.87 0.86 0.86 0.96 1.05 0.98 1.03 0.89 0.87 0.88 1.03 1.03 1.08 1.07 1.08 1.02 1.16 0.93 1.15 1.06 1.09 1.09 0.87 0.96 1.16

predicting the shear capacity of the BTTST. Based on the results of this study, the following conclusions are drawn: 1. The bent-up triangular tab shear transfer (BTTST) enhancement is a viable alternative for a shear connector. The results show that specimens employed with shear transfer enhancement increase the shear capacities of the specimens as compared to those relying only on the natural bond between coldformed steel and concrete. From investigation of the two types of shear transfer enhancements, BTTST provides better performance in term of strength. Load capacities for specimens with BTTST are relatively higher as compared to the equivalent control specimen, i.e. by 91% to 135%. When compared to LYLB specimens the increments are 12% to 16%. It is concluded that the proposed shear transfer enhancement has sufficient strength and is feasible. 2. The failure mode for the specimens was controlled by concrete crushing and splitting for those with shear connector. 3. The ductility calculated is smaller than the minimum required slip capacity of 6 mm according to Eurocode 4 [10]. This value exhibited a brittle behavior for all specimens. 4. The compressive strength of concrete influences the BTTST strength significantly. The percentage increase in shear connection capacity was found to be approximately 40% of the corresponding increase in concrete strength. 5. By increasing the CFS thickness, this increases the shear capacity of BTTST. A good agreement is observed between Phase 1 and Phase 2 results in case of CFS thickness variation. This is expected, since CFS thickness does not have a major influence when failure is concrete related, as the mode of failure indicates that the specimens failed due to concrete crushing and splitting. 6. The shear capacity of BTTST can be increased with increment of tab angle and size of BTTST. 7. A new equation (Eq. (7)) has been developed for the design of shear capacity of BTTST in precast CFS-concrete composite beams. This equation includes all important parameters, i.e., compressive strength of concrete, modulus of elasticity of concrete, CFS strength, dimension of BTTST (Lf , Ls , angle) and CFS thickness.

8. The proposed equation is in good agreement with the observed test values and Eq. (7) is recommended for the design of bentup triangular tab shear transfer (BTTST) enhancement in CFSconcrete composite beams. 9. Eq. (7) must, therefore, be validated by full scale CFS-concrete composite beam test before its acceptance as a design equation. Acknowledgments The authors would like to thank the Ministry of Higher Education, Malaysia, Universiti Tun Hussein Onn Malaysia (UTHM) and Universiti Teknologi MARA (UiTM), Malaysia for sponsoring this research. References
[1] Yu WK, Chung KF, Wong MF. Analysis of bolted moment connections in cold-formed steel beam-column sub-frames. Journal of Constructional Steel Research 2005;61:133252. [2] Abdullah R, Tahir MM. MH Performance of cold-formed steel of box-section as composite beam. In: 6th International conference on steel and space structures ; 1999. [3] Hossain KMA. Experimental and theoretical behavior of thin walled composite filled beams. Electronic Journal of Structural Engineering 2003;3:11739. [4] Hanaor A. Test of composite beams with cold-formed section. Journal of Constructional Steel Research 2000;54(2):24564. [5] Nakamura S. Bending behavior of composite girders with cold formed steel U section. Journal of Structural Engineering 2002;128(9):116976. [6] Lakkavalli BS, Liu Y. Experimental study of composite steel C -section floor joists. Journal of Constructional Steel Research 2006;62:9951006. [7] Airil YMY, Nethercot DA. Cross-sectional properties of complex composite beams. Engineering Structures 2007;29:195212. [8] Shaari SN, Ismail E. Promoting the use of industrialised building systems and modular coordination in the malaysian construction industry. Board of Engineers Malaysia. Bulletin Ingenieur March; 2003. 68. [9] BS5400. Steel, concrete and composite bridges: Part 5: Code of practice for design of composite bridges. British Standards Institution (London); 1979. [10] Eurocode 4. Design of composite steel and concrete structures: Part 1.1: General rules and rules for buildings. BS EN 1994-1-1:2004. British Standards Institution (London); 2004. [11] BS8110. Structural use of concrete: Part 2: Code of practice for special circumstances. British Standards Institution (London); 1985.

2098

J. M.Irwan et al. / Journal of Constructional Steel Research 65 (2009) 20872098 [18] Lam D, Ellobody E. Behavior of headed stud shear connectins in composite beam. Journal of Structural Engineering 2005;131(1):96107. [19] Shim CS, Lee PG, Yoon TY. Static behaviour of large stud shear connectors. Engineering Structures 2004;26:185360. [20] An L, Cederwall K. Push-out test on studs in high strength and normal strength concrete. Journal of Constructional Steel Research 1996;36:1529. [21] Mujagic JR, Easterling WS, Murray TM. Drilled standoff screws for shear connection in light composite steelconcrete trusses. Journal of Constructional Steel Research 2007;63:140414. [22] Ollgaard JG, Slutter RG, Fisher JW. Shear strength of stud connectors in lightweight and normal-weight concrete. American Institute of Steel Construction Engineering Journal 1971;8:5564. [23] Oguejiofor EC, Hosain MU. Behavior of perfobond rib shear connectors in composite beam: Full-size tests. Canadian Journal of Civil Engineering 1992; 19:22435.

[12] Hosain MU, Pashan A. Channel shear connectors in composite beams: Pushout tests. Composite construction in steel and concrete V. ASCE 2006;50110. [13] Veldanda MR, Hosain MU. Behaviour of perfobond rib shear connectors: Pushout test. Canadian Journal of Civil Engineering 1992;19:110. [14] Oguejiofor EC, Hosain MU. A parametric study of perfobond rib shear connectors. Canadian Journal of Civil engineering 1994;21:61425. [15] Vianna JC, Costa-Neves LF, Vellasco PS, Andrade S. Structural behavior of T Perfobond shear connectors in composite girders: An experimental approach. Engineering Structures 2008;30(9):238191. [16] Lam D. Capacities of headed stud shear connectors in composite steel beams with precast hollowcore slab. Journal of Constructional Steel Research 2007; 63:116074. [17] Ellobody E, Young B. Performance of shear connection in composite beams with profiled steel sheeting. Journal of Constructional Steel Research 2006; 62:68294.

You might also like