Elsevier

Engineering Structures

Volume 33, Issue 12, December 2011, Pages 3197-3207
Engineering Structures

Design formulations for non-welded and welded aluminium columns using Continuous Strength Method

https://doi.org/10.1016/j.engstruct.2011.07.027Get rights and content

Abstract

Aluminium members are used in structural applications due to their high strength-to-weight ratio, corrosion resistance, attractive appearance, recyclability, ease of production and availability. Thin aluminium sections are susceptible to buckling at a relatively low stress and welding makes it even worse; the design stress i.e. 0.2% proof stress is almost halved in the vicinity of the heat affected zones (HAZs). Currently available design codes have their guidelines both for welded and non-welded aluminium columns, but the predictions for welded aluminium columns are often quite inconsistent. The current research exploits a newly developed strain based design approach the ‘Continuous Strength Method’ (CSM) to predict the behaviour of aluminium members with SHS and RHS cross-sections subjected to compression. A new design curve is proposed herein to predict the cross-sectional resistance in compression; this concept is further extended to propose a new Perry type buckling curve to predict the flexural buckling resistance of aluminium columns. A simplified technique is proposed to include the effect of heat affected zone (HAZ) in CSM formulations. The CSM predictions for aluminium columns are compared against those obtained using available guidelines proposed by the European, American and Australian/ New Zealand standards of aluminium structures. The CSM predictions for non-welded columns are in line with the code predictions, whilst the proposed simple technique for transversely welded columns seems to produce significantly improved predictions.

Highlights

► Aluminium columns have been investigated using the Continuous Strength Method (CSM). ► Modified Perry type curve has been proposed to predict flexural buckling resistance of aluminium columns. ► The proposed approach includes the effects of transverse welding of aluminium columns. ► The CSM provides better and consistent predictions for welded aluminium columns.

Introduction

Aluminium tubular members are used in structures due to their high strength-to-weight ratio, corrosion resistance and ease of production. Aluminium members are generally manufactured by heat treated aluminium alloys to increase the design strength i.e. yield stress. The welding process involved during construction, on the other hand, reduces the material strength significantly in the vicinity of welded regions; this phenomenon is known as the softening of heat-affected zone (HAZ). In the case of 6000 series aluminium alloys, the heat generated from the welding process can reduce the parent metal strength by up to 68% in the HAZ [1]. Hence accurate prediction for the resistance of welded aluminium members is an important aspect in designing aluminium structures.

Currently available design codes for aluminium are based on the effectiveness of cross-sections, which is primarily determined by the width-to-thickness ratio of the constituent elements. Softening effect is accounted for by using an equivalent reduced thickness for cross-sections subjected to welding. The available guidelines for non-welded aluminium columns produce reasonably accurate predictions although the code predictions for welded columns are considerably inconsistent when compared against those obtained from experiments. The current research exploits available test results on aluminium columns with flat plated hollow sections i.e. SHS and RHS reported by Zhu and Young [1], [2], Faella et al. [3] and Langseth and Hopperstad [4] to devise design principles following a recently developed strain based design concept the Continuous Strength Method (CSM). The basic concept of CSM was originally proposed for stainless steel [5], [6], [7], whilst the suitability of this concept for other nonlinear metallic materials was demonstrated elsewhere [8]. The basic design curve in CSM is based on the behaviour of cross-sections, which can be further extended to predict the flexural buckling resistance of columns. The current paper investigates the load–deformation behaviour of aluminium SHS and RHS stub columns to devise a basic design curve to predict the deformation capacity εLB of cross-sections. Perry type buckling curves are proposed herein to predict the buckling resistance of long columns using the cross-sectional resistances obtained by CSM formulations. A new addition to the CSM concept is the incorporation of the weakening effect of transverse weld for aluminium columns.

Section snippets

Test program for aluminium stub columns

The test program presented by Zhu and Young [2] provided experimental ultimate loads and failure modes of aluminium alloy square and rectangular hollow sections compressed between fixed ends. The test specimens were fabricated by extrusion using 6063-T5 and 6061-T6 heat-treated aluminium alloys. The test program included 29 columns with both ends transversely welded to aluminium end plates (welded columns), and 12 columns without welding of end plates (non-welded columns). The non-welded and

The Continuous Strength Method

The traditional concept of section classification for metallic thin-walled cross-sections, based on the width-to-thickness ratio of its constituent plate elements, was originated from the yielding phenomenon observed in ordinary carbon steel. The success of this approach prompted its adoption for other metallic materials i.e. aluminium, high strength steel and stainless steel albeit the absence of any notable yield point in stress–strain response. The continuity in stress–strain behaviour forms

CSM for aluminium cross-sections: design curve formulation

Continuous Strength Method is based on the local buckling characteristics of cross-sections. Stub column test results are used to obtain the basic relationship between the slenderness parameter β and the deformation capacity εLB of cross-sections.

Basic concept

Lai and Nethercot [23] investigated the suitability of Perry-Robertson type design formulations for aluminium columns using theoretical results obtained for the buckling of aluminium columns with longitudinal and local transverse welds. The concept of using modified forms of the Perry-Robertson formula, as adopted in Eurocode 9: EN 1999-1-1 [20], has been used in the present research to devise appropriate flexural buckling curve for aluminium columns. Flexural buckling resistance Nb,Rd of an

Conclusions

The compression resistance of aluminium columns have been investigated in the current research, which led to devising an appropriate design equation following the Continuous Strength Method. To take appropriate account of slender cross-sections i.e. Class 4 sections in EN1999-1-1 [20], a cross-section effectiveness coefficient ϕeff is proposed to include the effects of loss-of-effectiveness for cross-sections with β>1.30. Inverted compound Ramberg–Osgood material model has been used to predict

Acknowledgement

The work described in this paper was supported by the William M.W. Mong Engineering Research Fund provided by The University of Hong Kong.

References (25)

  • C. Faella et al.

    Local buckling of aluminium members: testing and classification

    J Struct Eng ASCE

    (2000)
  • L. Gardner

    The Continuous Strength Method

    Proc Inst Civ Eng Struct Build

    (2008)
  • Cited by (24)

    View all citing articles on Scopus
    View full text