PSI - Issue 7

A. Rotella et al. / Procedia Structural Integrity 7 (2017) 513–520 Antonio Rotella et al. / Structural Integrity Procedia 00 (2017) 000–000

514

2

strength of the cast Aluminum alloys [Iben Houria et al. (2015); Le et al. (2016), Mu et al. (2014), Serrano-Munoz et al. (2017)]. Different parameters can influence the fatigue limit of a material affected by natural defects. Among others, the impact of a defect toward fatigue strength can be characterised by the defect size, the defect morphology and the defect position. Buffière et al. (2001), identified a relationship between the defect size and the sphericity of a defect, the authors found that gas pores are usually smaller and with a higher sphericity ratio with respect to natural shrinkages. Nicoletto et al. (2012) observed (by performing finite elements simulations on the real defect geometry) that the pore morphology can influence the stress distribution around a natural pore. Li et al. (2009) identified that the pore shape has a notable impact on the factor K g (defined as the normalised stress/strain concentration factor). Mainly the authors determined that the stress distribution of a natural pore can be approximated using a sphere with the same projected surface on the transversal plane of the specimen. The difference over the evaluation of the K g factor, between the real shape and the approximated shape, is of about 10%, showing that the complex morphology can be approximated by a simpler shape, but the choice of the equivalent shape can influence the final result. A recent study conducted by Serrano-Munoz et al. (2017) showed that the defect position can engender an environmental effect and influence the fatigue crack growth ratio. The authors identified that fatigue crack growth ratio is higher when a crack initiates and propagates from a surface defect due to the interaction with the external environment, this effect is reduced when the crack initiates from an internal defect, since the propagation phase occurs under an inert environment. The purpose of this work is to investigate the effect of the local and global morphology of natural and artificial defects on the fatigue limit of the A357-T6 cast Aluminum alloy. Two natural defect families have been tested (cavity and sponge shrinkages), the effect of the local morphology have been investigated by performing fatigue tests on specimens with surface artificial defects. By means of a µ-CT analysis and 3D finite elements simulations, the influence on the stress / strain distribution around a natural pore has been studied and the results obtained on the real defect morphology, have been compared with two equivalent geometries: a sphere and an ellipsoid. 2. Material and experimental procedure In this work, experimental fatigue tests have been carried out on six different customized castings of an unmodified A357 Aluminum alloy (there is no addition of Strontium during the melting process). The fatigue specimen is obtained after a machining step and is characterized by a cylindrical gage section with a diameter of 10 mm and a height of 20 mm. The chemical composition of the Aluminum alloy is the following : 7.05 % of Si, 0.57 % of Mg, 0.16 % of Ti, 0.12 % of Fe and less than 0.03% of Mn, Cu, Ni, Sn, Zn and Pb. A T6 heat treatment has been performed before testing the structures: (1) heating to solution at 540°C for 10h; (2) quenching in cold water; (3) aging at room temperature for 24h; (4) aging at 160°C for 8h. The mechanical properties of the “as received” material are identified with a monotonic tensile test (R p 0.2 =275 MPa, R m =335 MPa, E=73 GPa). In the context of this study a special casting configuration has been developed in order to obtain six different castings with natural defects, the defect type and grade are defined using the ASTM standard 2422-11 (2011). For the experimental campaign, a reference alloy (grade < 1) has been tested, and the results are compared with 5 degraded castings with a defect grade that ranges from 2 to 4: cavity shrinkage grade 2 (CS 2), cavity shrinkage grade 3 (CS 3), cavity shrinkage grade 4 (CS 4), sponge shrinkage grade 2 (SS 2) and sponge shrinkage grade 3 (SS 3). The characterization of each casting specimen is performed using X-Ray radiography inspection. This control permits to reveal the internal defects (shrinkages, inclusions, gas porosities …). This inspection technique is the same used for a non-destructive control of an industrial casting component. The defect characterization is based into a comparison of two radiographies that are taken by performing a specimen rotation of 90° between the two images. 2.1. Microstructure The microstructure of the A357-T6 cast Aluminum alloy is composed by two principal phases: the Aluminum matrix α and the eutectic phase (rich in Silicon); a micrograp h of the A357-T6 is shown if Figure 1a. The refinement grade of the microstructure is normally characterized by the secondary dendrite arms spacing (SDAS). This parameter directly impact the yield strength [(Wang (2004); Wang et al. (2010)] and the fatigue limit (in function of the applied load) [(Iben Houria et al. (2015); Roy et al. (2012); Wang et al. (2001)]. The SDAS size is directly related to the cooling rate of the casting. In order to obtain the targeted naturals defects in the gage section of the specimens, during the casting process, the mold is thermally modified using chillers. This practice permits to obtain the targeted defect but can affect the microstructure of the material. In order to be sure that the six castings have the same microstructure,