Variability of structural aluminium alloys mechanical properties Tihomir Dokšanovic ´ a,⇑ , Ivica Dz ˇeba b , Damir Markulak a a Faculty of Civil Engineering Osijek, University of Osijek, Drinska 16a, HR-31000 Osijek, Croatia b Faculty of Civil Engineering, University of Zagreb, Fra Andrije Kac ˇic ´a-Miošic ´a 26, HR-10000 Zagreb, Croatia article info Article history: Received 6 September 2016 Received in revised form 15 March 2017 Accepted 23 March 2017 Keywords: Reliability Design Aluminium Material partial factor abstract Content of alloying elements, type of treatment, welding and testing parameters result in variability of mechanical properties of aluminium alloys, which needs to be taken into account in design procedures. The degree of variability and current nominal values safety level were evaluated using a collection of results from published sources, expanded with experimental work. Processed results of proof and tensile strengths were fitted to distributions and cumulative function relevant fractiles for design purposes determined. It was established that ratios of characteristic to nominal values vary from 1.23 to 0.75, being unsafe or having a built in excess margin of safety, so new nominal values were proposed. These new val- ues provide a more economical material partial factor value and a uniform reliability level. Ó 2017 Elsevier Ltd. All rights reserved. 1. Introduction Production of aluminium alloys is done through modern indus- trial processes for which there are established quality control pro- cedures and defined requirements in relevant specifications. Alloys are manufactured in eight series and hundreds of specific chemical compositions, additionally divided by type of treatment (mechan- ical and thermal), which provides availability in several tempers. All combinations of alloys and tempers have to be made in confor- mity with [1,2] for the European and with [3] for the USA market, and in Europe a manufacturer guaranties given properties with its CE (Conformité Européene – European Conformity) marking. Reliable and safe performance of aluminium structures requires design procedures that depend on adequately accurate evaluation of mechanical properties of all available structural alloys. The 0.2% proof strength is the most important property for the design proce- dure, followed by the ultimate tensile strength and then other properties like Young’s modulus and Poisson’s value. All design cal- culations related to cross sectional resistance and instability require the 0.2% proof strength and the ultimate tensile strength is required in checks related to resistance of connections. Values of both 0.2% proof and ultimate tensile strengths presented in nor- mative documents are not characteristic, but guaranteed (mini- mum), which is supposed to provide additional reliability in terms of characteristic to nominal value relation. It should be noted that the term guaranteed is of more commercial than technical value as it does not offer actual statistical information. This par- tially deviates from a general pattern set in EN 1990 [4] where characteristic values are used in design calculations instead of nominal ones. The choice of partial (safety) factors is based on calibrations exercises, which should ensure that the target failure probability is met. These exercises are in most part carried out directly on the resistance associated with a specific failure mode, and no direct 0.2% proof strength or ultimate tensile strength statistical evalua- tion on coupons is conducted. This is justified as most of the global partial factor is influenced by the uncertainty in the engineering model and by the fact that specifications make manufacturers guarantee for the nominal values of strength. With variability of mechanical properties being induced by chemical composition deviations, heat treatment and welding procedure parameters design values reliability can be influenced. In order to quantify the extent of this influence variability of mechanical properties has to be considered on the material level itself, using uniaxial test results. Another implication regarding variability of material prop- erties is the market area, i.e. the regional context of the use of alloys. Namely, based on the proximity of production, regional areas are supplied by different manufacturers, which due to numerous variables may produce products in conformity with related specifications but with greater or lower variability. This fact is partly introduced and recognized by differences in minimum values of strength for the same alloy and temper (Table 1). Euro- pean norms are valid for all countries of the European Union, and wider, and calibration exercise should therefore be made consider- ing this context, but can hardly do so, which is also recognized by existence of National annexes (NA) of Eurocodes that can introduce partial factor values different from recommended ones. http://dx.doi.org/10.1016/j.strusafe.2017.03.004 0167-4730/Ó 2017 Elsevier Ltd. All rights reserved. ⇑ Corresponding author. E-mail address: tdoksanovic@gfos.hr (T. Dokšanovic ´). Structural Safety 67 (2017) 11–26 Contents lists available at ScienceDirect Structural Safety journal homepage: www.elsevier.com/locate/strusafe