Search

This study is initiated by the increasing use of steel tubular members with octagonal hollow sections (OctHSs) in civil structures. Previous studies primarily focused on the local buckling behaviour of cold-formed steel OctHSs. However, investigations on the global buckling behaviour of OctHS long columns remain scarce. Thus, the main objective of this paper is to numerically investigate the structural performance of the OctHS columns. A finite element method (FEM) was developed for tubular columns and validated against relevant experimental results reported in the literature. Subsequently, a parametric study on 239 OctHS columns was carried out. The applicability of design methods outlined in Eurocode 3, ANSI/AISC 360-16 and ASCE/SEI 48-19 was evaluated. The evaluation shows that the design method in Eurocode 3 provides conservative predictions for the load-bearing capacity of OctHS columns, while the ANSI/AISC 360-16 design method was slightly unconservative for the strength predictions. The design method in ASCE/SEI 48-19 gave inaccurate predictions for the load-bearing capacity of the OctHS columns. Modifications to the design methods in ANSI/AISC 360 and ASCE/SEI 48-19 were proposed to improve the accuracy of strength predictions and to obtain safe structural designs. The reliability of the design method in Eurocode 3 and the modified American design methods was assessed through the reliability analysis and found to satisfy the reliability requirements.

This paper presents an experimental investigation on the compression behaviour of cold-formed octagonal, circular and square tubular steel stub columns. Three octagonal hollow steel tubes were examined under axial compression while three circular steel tubes and two square steel tubes were tested for comparison. Nine existing test data on octagonal hollow steel tubes were also collated for comparison. The hollow tubes were fabricated by welding two cold-formed half-section steel plates. To examine the material characteristics, flat, curved and corner coupons were extracted and tested. Residual stress distribution in longitudinal direction was also measured by sectioning method. Fifty longitudinal strips with a nominal width of 10 mm were machined by a wire cutting machine. The test results obtained from the tensile coupon tests and residual stress measurements were examined to evaluate the effect of the fabrication process on different cross-sectional shapes. The geometric imperfections of each specimen were carefully measured. Cross-sectional capacity, load-deformation relationships and failure modes of the stub columns were presented by the stub column tests. Finite element model was developed and validated against the experimental results. The experimental cross-sectional capacities were accurately predicted by the finite element models. The test data accompanied with the FE results were assessed with reference to current design provisions on cross-section classification and a suggested cross-section slenderness limit and an alternative design approach for octagonal cross-sections were proposed.

Existing studies have shown that fiber-reinforced polymer (FRP) jacketing/wrapping is highly effective in strengthening circular concrete columns but is much less effective for rectangular/...

This paper presents an investigation on axially loaded polygonal-shaped steel-tube columns filled with high-strength concrete. The tubes had triangular, hexagonal and octagonal cross-sections, the concrete had a nominal concrete compressive cylinder strength of 100 MPa and the steel had a nominal yield strength of 285 MPa. For each tube cross-section, two identical specimens of concrete-filled tubes and one specimen of a hollow tube were examined. Load shortening, strain histories and failure modes were recorded and compared. Finite-element analysis with three different existing constitutive models of concrete were also developed and validated against the experimental results. The load-carrying capacities of the concrete-filled tubes were compared with the derived values from BS EN 1994-1-1:2004, ACI 318-11/AISC 360-16 and GB 50936-2014. The feasibility of current design standards for concrete-filled steel-tube columns of different cross-sections is discussed.

It is well known that the effect of confinement in concrete filled steel tube (CFST) can enhance the axial compressive strength of the concrete core. The confining pressure is a key issue to determine the compressive strength of concrete while the development of confining pressure is related to the lateral responses of steel tube and concrete core. This paper firstly presents a review on the lateral behaviour of circular CFST and existing lateral-axial strain models for concrete. These models are assessed by the authors’ previous experimental observations. Assessment results reveal deficiencies of current lateral-axial strain models, especially on the CFST with high strength concrete. A modified lateral-axial strain model for both normal and high strength concrete core in CFST was then proposed to accurately capture the interaction behaviour of the steel tube and the concrete. An analytical model of CFST with uniformly confined concrete core (i.e. with circular section) was then proposed to capture the development of confining pressure in CFST. A reduction factor was considered when the high strength concrete was used. The proposed analytical model can well predict the load bearing capacity, load-shortening curve, axial strain-lateral strain relationship of CFST with normal and high strength concrete.

This paper presents an experimental and numerical investigation into stub column behaviour of cold-formed steel octagonal hollow sections (OctHSs). A total of 16 OctHS stub columns were tested. Tensile coupons were extracted from both flat and corner portions of hollow sections to determine corresponding material properties. Finite element (FE) models were developed using commercially available software ABAQUS to replicate the test results generated in this study. The degree to which the enhanced material properties at corners should be extended was investigated. It was found that FE models with corner material properties extended to a width of material thickness beyond the corner portions offer the best agreement with test observations. The validated FE models were then adopted to conduct parametric studies to supplement test database. Cross-sectional slenderness limits specified in current design codes, including EN 1993-1-1, ANSI/AISC 360-16, ASCE/SEI 48-11 and AISI S100-16 (DSM) were evaluated against the test results in conjunction with the FE results. It was found that current limits are not suitable for the design of OctHSs. New cross-sectional slenderness limits in accordance with EN 1993-1-1, ANSI/AISC 360-16, ASCE/SEI 48-11 and DSM were then proposed based on the test and FE results. Cross-sectional capacity predictions obtained from EN 1993-1-1, ANSI/AISC 360-16, ASCE/SEI 48-11 and DSM were also compared with test and FE results. It is shown that the capacity predictions for slender sections from ASCE/SEI 48-11 are slightly unsafe, and ANSI/AISC 360-16 produces relatively satisfactory capacity predictions.

This paper presents a comparative investigation on the load capacity of octagonal concrete filled steel tubes (CFST) with that of the commonly used circular and square CFST. Existing experimental data of octagonal CFST were collected and based on the cross-sectional properties of the existing octagonal specimens, the corresponding circular and square cross-section were generated under three different control parameters: total cross-sectional area of column, confinement ratio and axial stiffness. Those circular and square cross-sections were used in the numerical analysis of CFST to obtain the load capacity for the comparative investigation. Validated finite element models were built for the modeling of the circular and square CFST. The outcome of comparison shows that the confinement ratio is the crucial parameter to the difference of axial behaviour between octagonal and circular CFST. Under the same confinement ratio, octagonal CFST has a very close axial bearing performance to that in circular CFST and are much better than the square CFST.

An experimental investigation on steel tube confined concrete (STCC) stub columns with circular, square and octagonal cross-section shapes under monotonic axial compression was conducted. Different concrete grades were examined with measured compressive cylinder strengths ranging from 38 MPa to 112 MPa. The nominal yield strength of the steel tube was 355 MPa. The load capacity, load-axial shortening curves and axial-hoop strain behaviour of the STCC stub columns are compared to that of the counterparts, concrete filled steel tubes (CFST), with different cross-section shapes based on the same material and geometric properties. Comparisons indicated that the load capacity of STCC stub columns are very close to that in the CFST with circular and octagonal cross-sections, while the performance of square STCC is not as good as the CFST counterparts due to low effectiveness of confinement. According to the experimental results of lateral behaviours of STCC, CFST and plain concrete columns, it was concluded that with very high strength concrete (cylinder strength over 100 MPa), the axial performance of STCC stub columns is better than that in CFST. A design assessment was also conducted on STCC stub columns and shows that the current design formulae for load capacity of CFST are also applicable to that of STCC stub columns.