Skip to content Skip to sidebar Skip to footer

Shear Studs Distribution on Continuous Composite Beams

1 Introduction

Composite construction is a common theme in buildings and bridges. Composite members can be formed by connecting different materials together to create a single member benefiting from the good properties of these materials [1,2,3,4]. There are two methods in creating a composite section. The first is by mixing different materials having suitable properties. The second is the arrangement of different sections with different materials to obtain the best properties. Shear connectors are widely used between steel and concrete to produce composite steel–concrete beams to reduce or prevent the relative displacement between concrete and steel [5,6].

Figure 1(a) shows the obvious slip between concrete and steel due to the lack of interaction between concrete and steel, while (b) shows the composite action due to the bond created by the shear connector which causes a reduction in both deflection and strain between the sections (concrete and steel). Shear connectors cannot achieve a perfect rigid connection between materials, but it widely eliminates the interface slip. Using a proper connection leads the two components to work as one unit, and this connection is known as full or complete interaction [8,9,10,11,12,13,14].

Figure 1                  (a) Non-composite and (b) composite beam [7].

Figure 1

(a) Non-composite and (b) composite beam [7].

However, all shear connectors are flexible to a certain degree and allow a certain amount of slip in the interface. As a result, this problem may occur when less connectors than the required number is used.

The key to composite work is the force transfer in the interface. This mechanism occurs by using shear connectors with different sections. Shear connectors should resist the horizontal forces developed between the composite materials [15,16].

To investigate the behavior of the composite concrete–steel beam joined by headed stud shear connectors, a finite element model is created [17,18]. The software program ABAQUS was utilized. The model's results were compared with experimental data and Practice Codes. The differences in concrete strength and the diameter of the shear connectors were investigated in parametric tests. The results reveal that the shear capacity of the headed studs could be overestimated when using finite element analysis.

To determine the shear and flexural strengths of composite simply supported beams, the finite element method is used [19,20,21,22]. These composite beams are made of steel and concrete and are subjected to a combination of shear and bending loads. A finite element model was built to compensate for the geometric nonlinearity of the beams, and the results were compared with the experimental results. The concrete slab's contributions to the composite beam's shear and moment capacity are calculated using a finite element model. For composite beams that are simply supported, the proposed design models offer a dependable and cost-effective method of design. The finite element results showed that the shear strength increase with the increase in the shear connection contribution.

Using multiple push-out tests, Rambo-Roddenberry et al. [22] studied the influence of steel plate thickness and shear connector location. According to their research, the thickness of the steel plate has an effect on the strength of shear studs in unfavorable locations. The strength difference between favorable and unfavorable positions is approximately 30%. Furthermore, the shear connector's tensile strength has a larger effect on the shear stud strength than the concrete's compressive strength.

Qureshi et al. [23] used 3D finite element models to investigate the spacing and layout of shear connectors in composite beams. The results showed that when the transverse spacing between the studs is 200 mm or more, shear resistance of shear connector pairs positioned in favorable positions is 94% of the strength of a single shear stud on average. A staggered pair of studs only has 86% of the strength of a single stud with the same spacing. Staggered pairs of shear connections have less strength than double shear studs in a favorable position.

Hosseini et al. [24] investigated the behavior of composite beams with trapezoidal profiled sheeting laid transverse to the beam axis. Four parameters were investigated using experimental findings from 24 full-scaled push test specimens, one of which was the shear stud arrangement. When compared to a layout with studs in the first four ribs, using studs just in the middle three ribs improved strength by 23%. Eurocode 4 and Johnson and Yuan [25] equations accurately predicted the stud strength for single stud/rib tests without normal load, with estimations within 10% of the characteristic test load. These equations underestimated the stud capacity by 40–50% when tested under normal load. AISC 360-16 generally overestimated the stud capacity, with the exception of single stud/rib push tests under normal load [26].

In this research, due to the importance of the shear connectors in reducing or preventing the relative displacement between concrete and steel, non-linear finite element analysis until failure is conducted on 24 continuous 2-span composite beams to investigate the effect of the arrangement and the number of shear connectors on the overall behavior of composite beams.

2 Description of samples

In this research, to study the effect of shear connectors, 3D nonlinear finite element analysis is conducted on 24 continuous 2-span composite beams with 2-point loads (one point load in each span). All beams have the same span details, length of 1 m for each span and the concrete slab was of width 250 mm and depth 8 mm. The steel I-section (IPE-140) had a depth of 140 mm, flange width of 72 mm, and thickness of 6 mm, while the web depth was 128 mm, thickness was 5 mm, and the total depth of the test samples was 220 mm as demonstrated in Figure 2.

Figure 2                  Details of section specimens.

Figure 2

Details of section specimens.

The strengthening of the concrete slab followed the criteria of the ACI construction code. Steel reinforcement in the longitudinal and transverse directions were based on shrinkage and temperature requirements [27,28]. Figure 2 demonstrates the negative and positive cross section of the beam.

The concrete slab was connected to a steel beam using the stud method. The shear connectors were assumed to be fully bonded to the I-steel beam's top flange and embedded in the concrete slab. The length and diameter of all studs were the same, but the number of studs varied depending on the study parameters from one sample to another. The variable parameters of the current study were divided into four categories:

2.1 Group A

The first group consisted of 6 steel–concrete beams (BC1, BC2, BC3, BC4, BC5, and BC6) with shear connectors divided into 2 rows with different distances 65, 85, 105, 150, 200, and 250 mm, respectively, as shown in Figure 3.

Figure 3                     Shear connectors distribution for Group A.

Figure 3

Shear connectors distribution for Group A.

2.2 Group B

The second group consisted of 6 steel–concrete beams (BC7, BC8, BC9, BC10, BC11 and BC12) with 1 of row shear connectors distributed along the longitudinal axis with distances of 65, 85, 105, 150, 200, and 250 mm, respectively, as shown in Figure 4.

Figure 4                     Shear connectors distribution for Group B.

Figure 4

Shear connectors distribution for Group B.

2.3 Group C

The third group included 6 steel–concrete beams (BC13, BC14, BC15, BC16, BC17 and BC18) single- and double-shear connectors distributed along the longitudinal axis with different distances 65, 85, 105, 150, 200, and 250 mm, respectively, as shown in Figure 5.

Figure 5                     Shear connectors distribution for Group C.

Figure 5

Shear connectors distribution for Group C.

2.4 Group D

The fourth group included 6 steel–concrete beams (BC19, BC20, BC21, BC22, BC23 and BC24) with one row shear connectors arranged staggered along the longitudinal axis with different distances 65, 85, 105, 150, 200, and 250 mm, respectively, as shown in Figure 6. The properties of materials used is shown in Table 1.

Figure 6                     Shear connectors distribution for Group D.

Figure 6

Shear connectors distribution for Group D.

Table 1

Properties of materials

Steel reinforcement Reactive powder concrete
Modulus of elasticity (MPa) 2 × 105 Compressive strength (MPa) 100.29
Yield strength (MPa) 520 Modulus of elasticity (MPa) 44.5 × 103
Element type T3D2 Dilation angle 40
Element size (mm) 20 Eccentricity 0.1
I-section steel fbo/fco 1.16
Modulus of elasticity (MPa) 2 × 105 K 0.667
Yield strength (MPa) 300 Viscosity 0
Element type C3D8R Element type C3D8R
Element size (mm) 20 Element size (mm) 20

3 Numerical model validation

To ensure that the current model in the ABAQUS software is appropriate, two samples are chosen and compared to Aggar's experimental results [29]. As demonstrated in Figure 7, the results of the experimental and the numerical model are in good agreement. Figure 8 shows the stress distribution in the experimental and numerical models of the sample (BC2).

Figure 7                  Load–deflection curve of the experimental and numerical sample (BC2).

Figure 7

Load–deflection curve of the experimental and numerical sample (BC2).

Figure 8                  Comparison between the experimental and numerical stress distribution at failure for the sample (BC2) [31].

Figure 8

Comparison between the experimental and numerical stress distribution at failure for the sample (BC2) [31].

4 Discussion and results

The results of analyzing the specimens with ABAQUS software demonstrate that the number of shear connections and their arrangement have an effect on the ultimate load failure for composite beams.

4.1 Group A

The results of the analysis indicate that changing the shear connection spacing from 65 to 85 mm and 105 mm increased the ultimate load capacity; however, increasing the spacing from 105 to 150, 200, and 250 mm decreased the ultimate load capacity, as shown in Table 2. Figure 9 shows the load and deflection for the specimens of group A.

Table 2

Ultimate strength and deflection of midspan of group A

Group no. Specimen Ultimate strength (kN) Deflection (mm) Percentage
A BC65 505.52 3.96
BC85 582.33 4.77 15.19
BC105 609.37 6.41 20.54
BC150 508.78 5.18 0.64
BC200 501.47 5.85 −0.80
BC250 483.95 6.42 −4.27

Figure 9                     Load–deflection for the specimens of Group A.

Figure 9

Load–deflection for the specimens of Group A.

Figure 10 shows stress distribution for specimens of Group A.

Figure 10                     Stress distribution for specimens of Group A.

Figure 10

Stress distribution for specimens of Group A.

4.2 Group B

The results of analysis of the second group also showed that the change in the spacing of shear connection from 65 to 85 and 105 mm led to an increase in the ultimate load capacity, while increasing the spacing from 105 to 150, 200, and 250 mm led to a decrease in the ultimate load capacity as shown in Table 3. Figure 11 shows the load and deflection for the specimens of Group B.

Table 3

Ultimate strength and deflection of midspan of Group B

Group no. Specimen Ultimate strength (kN) Deflection (mm) Percentage
B BC65 464.67 3.82
BC85 505.21 4.99 8.85
BC105 520.75 5 12.20
BC150 473.13 5.3 1.94
BC200 455.16 6.54 −1.93
BC250 435.22 8.24 −6.23

Figure 11                     Load–deflection for the specimens of Group B.

Figure 11

Load–deflection for the specimens of Group B.

Figure 12 shows stress distribution for specimens of Group B.

Figure 12                     Stress distribution for specimens of Group B.

Figure 12

Stress distribution for specimens of Group B.

4.3 Group C

The results of analysis of the third group showed that the shear connections with a spacing of 105 mm gave the highest ultimate load capacity than all the other spacings of the same group but with less difference percent than Group A and Group B, as shown in Table 4. Figure 13 shows the load and deflection curves of group C.

Table 4

Ultimate strength and deflection of midspan of Group C

Group no. Specimen Ultimate strength (kN) Deflection (mm) Percentage
C BC65 440.53 4.25
BC85 450.23 4.57 2.20
BC105 460.67 4.38 4.57
BC150 442.09 5.01 0.35
BC200 440.4 6.4 −0.03
BC250 420.19 6.81 −4.62

Figure 13                     Load–deflection for the specimens of Group C.

Figure 13

Load–deflection for the specimens of Group C.

Figure 14 shows stress distribution for specimens of Group C.

Figure 14                     Stress distribution for specimens of Group C.

Figure 14

Stress distribution for specimens of Group C.

4.4 Group D

The results of analysis of the fourth group showed that the shear connections with a spacing of 150 mm gave the highest ultimate load capacity than all the other spacings of the same group and the arrangement of shear connection led to the buckling in the flange of the steel I-section, Table 5 shows deflection and ultimate load for Group D. Figure 15 shows the load and deflection curves of Group D.

Table 5

Ultimate strength and deflection of midspan of Group D

Group no. Specimen Ultimate strength (kN) Deflection (mm) Percentage
D BC65 386.64 3.71
BC85 392.97 4.74 1.64
BC105 410.53 3.12 6.18
BC150 415.2 16.65 7.39
BC200 410.06 11.19 6.06
BC250 365.2 13.75 −5.55

Figure 15                     Load–deflection for the specimens of Group D.

Figure 15

Load–deflection for the specimens of Group D.

Figure 16 shows stress distribution for Group D.

Figure 16                     Stress distribution for Group D.

Figure 16

Stress distribution for Group D.

As shown in Figure 17, it can be noticed that the effect of changing the spacing of shear connectors for Group A is more effective than the other groups.

Figure 17                     The curve of ultimate load–deflection for all groups.

Figure 17

The curve of ultimate load–deflection for all groups.

5 Conclusion

This study presented a numerical investigation of composite steel–concrete members using ABAQUS software, and the outcomes of this study were as follows:

  • The composite steel concrete beam with 2 symmetrical shear stud rows has higher bending strength capacity by 17, 32, and 48% compared with the groups B, C, and D, respectively.

  • For the fourth group (D), the deflection was fluctuating up and down, which resulted from nonsymmetrical distribution of the shear connectors. And this case was considered the worst condition compared with the symmetrical distribution.

  • The optimum spacing was 105 mm in all groups (A, B, C, and D) compared with all other spacings of 65, 85, 200, and 250 mm.

  • When comparing the optimum spacing of shear connector (105 mm) with other spacings such as 150, 200, and 250 mm, the bending strength capacity of the composite beam was reduced by 20, 20, and 26%, respectively.

Acknowledgments

The authors express their gratitude to Al-Mustaqbal University college, Babylon, Iraq for their contribution in completing this work.

  1. Funding information: The authors state no funding involved.

  2. Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission.

  3. Conflict of interest: Authors state no conflict of interest.

References

[1] Shubbar A, Nasr M, Falah M, Al-Khafaji Z. Towards net zero carbon economy: Improving the sustainability of existing industrial infrastructures in the UK. Energies. 2021;14(18):5896. 10.3390/en14185896Search in Google Scholar

[2] Falah MW, Hafedh AA, Hussein SA, Al-Khafaji ZS, Shubbar AA, Nasr MS. The combined effect of CKD and silica fume on the mechanical and durability performance of cement mortar. Key Eng Mater. 2021;895:59–67. 10.4028/www.scientific.net/KEM.895.59Search in Google Scholar

[3] Shubbar A, Al-khafaji Z, Nasr M, Falah M. Using non-destructive tests for evaluating flyover footbridge: case study. Knowl Eng Sci. 2020;1(1):23–39. 10.51526/kbes.2020.1.01.23-39Search in Google Scholar

[4] Al-Khafaji ZS, Falah MW. Applications of high density concrete in preventing the impact of radiation on human health. J Adv Res Dyn Control Syst. 2020;12(1):105961. 10.5373/JARDCS/V12SP1/20201115Search in Google Scholar

[5] Ellobody E. Finite element analysis and design of steel and steel–concrete composite bridges. Oxford, UK: Butterworth-Heinemann; 2014. 10.1016/B978-0-12-417247-0.00005-3Search in Google Scholar

[6] Mazoz A, Benanane A, Titoum M. Push-out tests on a new shear connector of I-shape. Int J Steel Struct. 2013;13(3):519–28. 10.1007/s13296-013-3011-4Search in Google Scholar

[7] Zeng X, Jiang SF, Zhou D. Effect of shear connector layout on the behavior of steel-concrete composite beams with interface slip. Appl Sci. 2019;9(1):207. 10.3390/app9010207Search in Google Scholar

[8] Zona A, Ranzi G. Finite element models for nonlinear analysis of steel–concrete composite beams with partial interaction in combined bending and shear. Finite Elem Anal Des. 2011;47(2):98–118. 10.1016/j.finel.2010.09.006Search in Google Scholar

[9] Esfandiari J, Latifi MK. Numerical study of progressive collapse in reinforced concrete frames with FRP under column removal. Adv Concr Constr. 2019;8(3):165–72. Search in Google Scholar

[10] Esfandiari J, Soleimani E. Laboratory investigation on the buckling restrained braces with an optimal percentage of microstructure, polypropylene and hybrid fibers under cyclic loads. Composite Struct. 2018;203:585–98. 10.1016/j.compstruct.2018.07.035Search in Google Scholar

[11] Esfandiari S, Esfandiari J. Simulation of the behaviour of RC columns strengthen with CFRP under rapid loading. Adv Concr Constr. 2016;4(4):319–32. 10.12989/acc.2016.4.4.319Search in Google Scholar

[12] Ali YA, Hashim TM, Ali AH. Finite element analysis for the behavior of reinforced concrete T-section deep beams strengthened with CFRP sheets. Key Eng Mater. 2020;857:153–61. 10.4028/www.scientific.net/KEM.857.153Search in Google Scholar

[13] Agioutantis Z, Kourkoulis SK, Maurigiannakis S, Papatheodorou E. Notched marble specimens under bending: experimental study and numerical analysis using an elastoplastic model with contact elements. J Mech Behav Mater. 2005;16(1–2):1–8. 10.1515/JMBM.2005.16.1-2.1Search in Google Scholar

[14] Yang J, Zhou J, Wang Z, Zhou Y, Zhang H. Structural behavior of ultrahigh-performance fiber-reinforced concrete thin-walled arch subjected to asymmetric load. Adv Civ Eng. 2019;2019:2019-12. 10.1155/2019/9276839Search in Google Scholar

[15] Zona A, Leoni G, Dall'Asta A. Influence of shear connection distributions on the behaviour of continuous steel-concrete composite beams. Open Civ Eng J. 2017;11(1):384–95. 10.2174/1874149501711010384Search in Google Scholar

[16] Zhai C, Lu B, Wen W, Ji D, Xie L. Experimental study on shear behavior of studs under monotonic and cyclic loadings. J Constr Steel Res. 2018;151:1–11. 10.1016/j.jcsr.2018.07.029Search in Google Scholar

[17] Lam D, El-Lobody E. Finite element modelling of headed stud shear connectors in steel-concrete composite beam. Structural engineering, mechanics and computation. Cape Town, South Africa: Elsevier Science; 2001. p. 401–8. 10.1016/B978-008043948-8/50041-2Search in Google Scholar

[18] Pathirana SW, Uy B, Mirza O, Zhu X. Flexural behaviour of composite steel–concrete beams utilising blind bolt shear connectors. Eng Struct. 2016;114:181–94. 10.1016/j.engstruct.2016.01.057Search in Google Scholar

[19] Liang QQ, Uy B, Bradford MA, Ronagh HR. Strength analysis of steel–concrete composite beams in combined bending and shear. J Struct Eng. 2005;131(10):1593–600. 10.1061/(ASCE)0733-9445(2005)131:10(1593)Search in Google Scholar

[20] Reginato LH, Tamayo JL, Morsch IB. Finite element study of effective width in steel-concrete composite beams under long-term service loads. Lat Am J Solids Struct. 2018;15:15. 10.1590/1679-78254599Search in Google Scholar

[21] Al-Khafaji ZS, Majdi A, Shubbar AA, Nasr MS, Al-Mamoori SF, Alkhulaifi A, et al. Impact of high volume GGBS replacement and steel bar length on flexural behaviour of reinforced concrete beams. IOP Conf Ser: Mater Sci Eng. 2021;1090:012015. 10.1088/1757-899X/1090/1/012015Search in Google Scholar

[22] Rambo-Roddenberry M, Lyons JC, Easterling WS, Murray TM. Performance and strength of welded shear studs. Composite Constr Steel Concr IV. 2002;458–69. 10.1061/40616(281)40Search in Google Scholar

[23] Qureshi J, Lam D, Ye J. Effect of shear connector spacing and layout on the shear connector capacity in composite beams. J Constr Steel Res. 2011;67(4):706–19. 10.1016/j.jcsr.2010.11.009Search in Google Scholar

[24] Hosseini SM, Mamun MS, Mirza O, Mashiri F. Behaviour of blind bolt shear connectors subjected to static and fatigue loading. Eng Struct. 2020;214:110584. 10.1016/j.engstruct.2020.110584Search in Google Scholar

[25] Johnson RP, Yuan H. Models and design rules for stud shear connectors in troughs of profiled sheeting. Proc Inst Civ Eng-Struct Build. 1998;128(3):252–63. 10.1680/istbu.1998.30459Search in Google Scholar

[26] Marshdi QSR, Hussien SA, Mareai BM, Al-Khafaji ZS, Shubbar AA. Applying of No-fines concretes as a porous concrete in different construction application. Periodicals Eng Nat Sci. 2021;9(4):999–1012. 10.21533/pen.v9i4.2476Search in Google Scholar

[27] ACI Committee. Building code requirements for structural concrete (ACI 318-08) and commentary. American Concrete Institute; 2008. Search in Google Scholar

[28] Standardization EC for Eurocode 4. Design of composite steel and concrete structures – Part 1-1: General rules and rules for buildings – EN 1994-1-1; 2004. Search in Google Scholar

[29] Aggar R. Structural behaviour of continuous steel-reactive powder concrete composite beams under repeated loads [dissertation]. Al-Hilla: Babylon University; 2018. p. 174. Search in Google Scholar

This work is licensed under the Creative Commons Attribution 4.0 International License.

whighamwitho1991.blogspot.com

Source: https://www.degruyter.com/document/doi/10.1515/jmbm-2022-0046/html

Post a Comment for "Shear Studs Distribution on Continuous Composite Beams"