Elsevier

Engineering Structures

Volume 222, 1 November 2020, 111142
Engineering Structures

Behaviour and design calculations of rectangular CFST beam-columns with slender sections

https://doi.org/10.1016/j.engstruct.2020.111142Get rights and content

Highlights

  • Finite element model for high-strength rectangular CFST beam-columns is presented.

  • Accuracy of Australian, European and American design standards is examined.

  • Australian and European standards overestimate strength of slender CFST beam-columns.

  • Recommendations based on Australian and European design standards are proposed.

Abstract

In this paper, design calculations of Concrete-Filled Steel Tubular (CFST) columns with slender sections subjected to combined axial compression and bending are considered based on the Australian code (AS 5100), American code (AISC 360–16) and European code (EC 4) approaches. A Finite Element (FE) model is developed to predict the strength of rectangular CFST slender section beam-columns. The experimental data is collected and used to examine the accuracy of AS 5100, AISC 360–16 and EC 4 in predicting the strengths of CFST slender section beam-columns, and to validate the FE model. In addition, based on the test results, FE analysis results and limits specified in the codes, design procedures are developed to propose modifications to the current AS 5100 and EC 4 to improve their accuracy in designing CFST slender section columns. The effects of material strengths on the accuracy of the proposed methods are also investigated. The results indicate that, while AISC 360–16 provides conservative strength predictions, the current AS 5100 and EC 4 overestimate the strengths of CFST slender section beam-columns. The comparative study also shows that better strength predictions are achieved using the proposed design procedures.

Introduction

Based on the local slenderness ratio (depth-to-thickness), Concrete-Filled Steel Tubular (CFST) columns can be categorised as compact, non-compact and slender sections [1], [2]. Compact sections have adequate wall thickness to develop yielding of the steel tube, and thus provide confinement for the concrete core, non-compact sections have enough wall thickness to develop yielding of the steel tube, but not to confine the concrete core, and slender sections have not sufficient wall thickness neither to develop yielding of the steel tube nor to confine the concrete core [1]. In recent years, CFST slender section columns have been widely used in structures such as concrete-filled pipe piles and even columns in tall buildings where the diameter of columns is generally large and it would be hard to fabricate steel tubes with large thicknesses to satisfy the compact limit [3]. This is mainly due to the fact that unlike steel columns, which are prone to buckle before yielding, local buckling of the steel tube in CFST slender section columns is delayed by the concrete infill [4], [5]. In fact, local buckling failure of CFST slender section columns occurs at local slenderness ratios higher than those observed for Hollow Structural Sections (HSS). However, since the steel tube in CFST slender section columns does not have a sufficient thickness to confine the concrete infill and thereby develop yielding in the steel, local buckling and overall instability are major concerns in design.

Although many researchers have investigated the behaviour of CFST columns, most of their studies are related to CFST compact and non-compact section columns subjected to combined axial compression and bending (beam-columns), or CFST compact, non-compact and slender section columns subjected to pure axial compression [6], [7], [8], [9], [10], [11]. Liang [12] proposed a performance-based analysis technique for designing CFST compact and non-compact section beam-columns. Lai et al. [4] proposed revisions to AISC 360–10 [13] for designing CFST non-compact and slender section beam-columns. Some of the prominent experimental studies conducted on rectangular CFST slender section beam-columns are presented here. Uy [14] carried out tests on 13 rectangular CFST columns using high-strength steel, where two specimens with slender sections were subjected to an eccentric axial force. It was found that the design provisions of EC 4 were unconservative and would not be applicable for designing CFST columns with a high-strength steel tube. Mursi and Uy [15] performed three tests on stub and slender CFST slender section columns under eccentric axial compression using a high-strength steel tube and normal-strength concrete. Their study showed that the local buckling effects depended on the slenderness of the column plates and this played a significant role in considering the confinement effect of the concrete infill. Han and Yao [16] tested 35 stub and slender CFST columns, where six were slender section beam-columns with a steel yield strength of 340 MPa and a concrete compressive strength of 22 MPa. The main purpose of the study was to consider the effect of concrete compaction on the strength of CFST columns. Their test results indicated that with better concrete compaction, higher member strengths could be achieved due to enhanced microlocking and macrolocking. Zhang and Guo [17] conducted experimental tests on 26 stub and slender CFST columns using high-strength concrete, where 10 specimens had slender sections. It was found that with increasing slenderness and eccentricity, the ultimate strength of the columns decreased rapidly. Also, local buckling occurred for specimens with depth-to-thickness ratios greater than 50. Liu [18] carried out tests on specimens with high-strength concrete under eccentric loading. They observed three types of failure modes. Local buckling occurred either close to the end or near the quarter height for the specimens with a slenderness ratio of 20, and at the mid height for specimens with a slenderness ratio of 50. Yu et al. [19] studied the behaviour of slender CFST columns with very high-strength concrete. Their experimental results showed that the ductility of CFST columns with very high-strength concrete is smaller than that of normal concrete. Tao et al. [20] investigated the experimental behaviour of slender stiffened CFST columns with slender sections. Their results showed that the stiffeners not only effectively delayed the local buckling but also increased the load-carrying capacity of the columns. From the above experimental studies, it can be deduced that test results for CFST beam-columns with slender sections are limited, as CFST slender section columns were not common in practical structures.

The aim of this paper is to develop a Finite Element (FE) model using ABAQUS software [21] for modelling rectangular CFST columns subjected to eccentric axial compression, with a width-to-thickness ratio greater than 3Es/fy as specified by AS 5100 [2] and AISC 360–16 [1], and 52235/fy as specified by EC 4 [22]. In this research, 72 rectangular CFST slender section beam-columns are proposed and modelled in ABAQUS and their ultimate axial strengths are reported. In addition, the experimental data on rectangular CFST slender section beam-columns is collected and used for the following purposes: (i) examining the accuracy of AS 5100, AISC 360–16 and EC 4 in predicting the strengths of CFST slender section beam-columns and (ii) validating the FE model by comparing the load–displacement curves and ultimate axial strengths reported from the tests with those obtained from FE models. Finally, recommendations are proposed for designing rectangular CFST slender section columns based on the design provisions of AS 5100 and EC 4.

Section snippets

Experimental database

The experimental database presented in this paper consists of 73 tests conducted on rectangular CFST slender section columns under eccentric axial compression. It should be mentioned that Han and Yao [16] performed repeated tests for each column, denoted by 1 and 2. The parameters and test results are presented in Appendix A (Table A1), which includes the length (L), width (B), depth (H), thickness (t), steel modulus of elasticity (Es), steel yield stress (fy), concrete modulus of elasticity (Ec

FE modelling

The FE model was developed in this study based on the previous models developed by Thai et al. [24] for high strength CFST columns as shown in Fig. 1. However, the magnitude of initial imperfections and the stress–strain models for steel and concrete materials were modified. The magnitude of the initial imperfection was taken as L/1000 [25], [26]. If the concrete was compacted by hand, the initial imperfection was taken as L/500 based on this verification study. The stress–strain relationships

FE model analysis and verification

In order to assess the accuracy of the developed FE models, two comparisons between the experimental and FE results were performed. First, the axial strength of the columns obtained from the tests are compared with those obtained from FE analyses. From Fig. 2, it can be observed that there is good agreement between the experimental and FE results, with a mean value of 1.00, showing the ability of FE analysis to accurately predict the axial strength of rectangular CFST slender section columns.

Parametric study

Using the verified FE model, a total of 72 rectangular CFST slender section columns, namely 36 columns under two load eccentricities of B/10 and B/20, are modelled with their geometrics and material properties summarised in Table 4. The length of all specimens is 1,000 mm and the length-to-width ratio ranges between 2.22 and 4 (i.e. stub columns [32]). According to AS 5100 and AISC 360–16, all specimens have λeAS and λeAISC greater than λey and less than or equal to λed, and based on EC 4, 14

Comparison between experimental results, FE model and code predictions

This section presents a comparison between the strengths obtained from the tests, FE models and design codes. In order to compare the experimental and FE results with code provisions, all resistance factors were taken as unity and all limitations on material strengths given by the codes were ignored, and the effective length factor (Ke) was taken as unity for the pinned–pinned boundary condition. In Fig. 6, the ordinates are the ratios of NExp to Neu, and NFEM to Neu where Neu was calculated

Proposed design procedures

Similar to both AS 5100 and EC 4, a full value of fc'is also adopted in this proposed design due to the fact that the concrete core filled in the steel tube of a CFST section has a better curing condition compared with the concrete in a reinforce concrete section. The AS 5100 design provisions for predicting the strength of CFST columns are generally similar to those in EC 4 [7], [24], [32]. Thus, the form factor (kf) adopted by AS 5100 is proposed to be included in the EC 4 design calculations

Comparison between tests, FE models and proposed design procedures

This section presents comparisons between the strengths obtained from the proposed methods and experimental results. In order to examine the influences of material strengths on the predicted results obtained from the proposed methods, the specimens are divided into two groups: M−S denotes specimens with material strengths lower than the allowed values and M−L represents specimens with a higher fc' or fy than the allowed values given in each design code as shown in Table 6.

Conclusions

In this paper, a test database on CFST slender section beam-columns has been compiled. In addition, an FE model on rectangular CFST slender section columns subjected to combined axial load and bending moment was developed. The test and FE analysis results were compared with the predicted results obtained from AS 5100, AISC 360–16 and EC 4. Design procedures were developed based on the code limits to propose revisions to the current AS 5100 and EC 4 for improving their feasibility in designing

CRediT authorship contribution statement

Mohammadreza Zarringol: Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing - original draft, Visualization. Huu-Tai Thai: Conceptualization, Methodology, Software, Validation, Writing - review & editing, Supervision, Project administration. Tuan Ngo: Validation, Writing - review & editing, Supervision. Vipulkumar Patel: Validation, Writing - review & editing, Supervision.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The research presented in this paper was supported by La Trobe University, School of Engineering and Mathematical Sciences, Australia.

References (48)

  • L.H. Han et al.

    Influence of concrete compaction on the strength of concrete-filled steel RHS columns

    J Constr Steel Res

    (2003)
  • D. Liu

    Behaviour of high strength rectangular concrete-filled steel hollow section columns under eccentric loading

    Thin-Walled Struct

    (2004)
  • Q. Yu et al.

    Experimental behaviour of high performance concrete-filled steel tubular columns

    Thin-Walled Struct

    (2008)
  • Z. Tao et al.

    Experimental behaviour of concrete-filled stiffened thin-walled steel tubular columns

    Thin-Walled Struct

    (2007)
  • H.T. Thai et al.

    Numerical modelling of concrete-filled steel box columns incorporating high strength materials

    J Constr Steel Res

    (2014)
  • Z. Tao et al.

    Finite element modelling of concrete-filled steel stub columns under axial compression

    J Constr Steel Res

    (2013)
  • V.I. Patel et al.

    Biaxially loaded high-strength concrete-filled steel tubular slender beam-columns, part II: Parametric study

    J Constr Steel Res

    (2015)
  • L.H. Han et al.

    Performance of concrete-filled thin-walled steel tubes under pure torsion

    Thin-Walled Struct

    (2007)
  • M.A. Tasdemir et al.

    Evaluation of strains at peak stresses in concrete: A three-phase composite model approach

    Cem Concr Compos

    (1998)
  • B. Binici

    An analytical model for stress–strain behavior of confined concrete

    Eng Struct

    (2005)
  • S. Thai et al.

    Concrete-filled steel tubular columns: Test database, design and calibration

    J Constr Steel Res

    (2019)
  • S.H. Lee et al.

    Behavior of high-strength circular concrete-filled steel tubular (CFST) column under eccentric loading

    J Constr Steel Res

    (2011)
  • L.H. Han et al.

    Experimental behaviour of thin-walled hollow structural steel (HSS) columns filled with self-consolidating concrete (SCC)

    Thin-Walled Struct

    (2004)
  • X. Qu et al.

    Experimental study of rectangular CFST columns subjected to eccentric loading

    Thin-Walled Struct

    (2013)
  • Cited by (16)

    • Prediction of the load-shortening curve of CFST columns using ANN-based models

      2022, Journal of Building Engineering
      Citation Excerpt :

      Concentric axial load was applied as a downward displacement at the top reference point. For each FEM, an eigenvalue buckling analysis was first performed and then linked to the main FEM using the *IMPERFECTION command with the magnitude of L/1000 [28]. Since the influence of the residual stresses on the overall behaviour and ultimate strength of CFST columns is negligible [27,29], they were not included in the FEMs.

    • Optimal allocation of material and slenderness limits for the rectangular concrete-filled columns

      2022, Journal of Constructional Steel Research
      Citation Excerpt :

      Therefore, it is essential to accommodate the design of CFT columns constructed from these high-strength materials in design codes either through the inclusion of specific design equations for such members or by assessing the applicability of current design equations of CFT columns to high strength materials. In recent years, many experimental programs and numerical simulations on high-strength CFT columns have been conducted [13,14,16–30] to evaluate the performance of such members and the applicability of the current design code equations beyond the material and slenderness limits shown in Table 1. In general, these studies agree on the applicability of using the current design codes to predict the cross-section strength of CFT columns using high-strength materials.

    View all citing articles on Scopus
    View full text