PSI - Issue 7

Ana D. Brandão et al. / Procedia Structural Integrity 7 (2017) 58–66

59

2

Author name / Structural Integrity Procedia 00 (2017) 000–000

manufacturing methods. Specifically, in Space applications, AM is seen as a game changing manufacturing process, enabling increased performance and reduced costs of space missions (Orme et al. (2017a); Wohlers Associates (2017); Orme et al. (2017b)). In fact, the applicability of this technology has already been demonstrated for space products, manufactured on earth (Inovar Communications (2014)), on-orbit (Hubscher (2014)) and even to be used on other planets (Pambaguian et al. (2016); Meurisse et al. (2016)). Space hardware produced through Additive Manufacturing is being developed with increasing frequency for ESA’s space missions and will be launched in the very near future, including secondary structures or other non-critical applications. However, for future missions, many more components are envisioned to be manufactured using AM, including primary structures or other mission-critical parts (Ghidini et al. (2015); Lopatriello et al. (2015)). To ensure the desired gain in performances and reduction in costs, it is necessary to improve the design of AM flight hardware, namely by adapting the safety margins to the inherent imperfections of this technology. Hence, it is crucial to understand the impact of these defects on the mechanical properties, taking into account the final application of the part and the material selected. In this quest for increased performance, aluminium alloys are considered greatly attractive for the space industry due to its high specific strength (specific to weight) and relatively low cost (Romano et al. (2017); Buchbinder et al. (2011); Read et al. (2015); Kempen et al. (2012); Hitzler et al. (2017); Brandl et al. (2012); Maskery et al. (2015); Tang and Pistorius (2017); Aboulkhair et al. (2016)). The AlSi10Mg casting alloy is a commonly used aluminium alloy in additive manufacturing, as it shows good weldability. However, its manufacturability via AM has been proven to be more challenging when compared to titanium alloys or stainless steels. Often, this is hypothesised to be mainly due to the high reflectivity of this alloy to the laser beam and the formation of an Aluminium oxide layer (Buchbinder et al. (2011); Read et al. (2015)). As consequence, a stringent control of the process parameters is necessary to produce high quality AlSi10Mg parts. Several studies in literature establish a correlation between the AM process parameters (e.g laser power, scan speed, build orientation) and the mechanical properties of this alloy, namely tensile strength (Buchbinder et al. (2011); Kempen et al. (2012)), hardness (Buchbinder et al. (2011); Hitzler et al. (2017)) and fatigue (Brandl et al. (2012); Maskery et al. (2015); Tang and Pistorius (2017); Aboulkhair et al. (2016); Mower and Long (2016)). Regarding the latter, only a rather low number of studies was found. Brandl et al. (2012), Maskery et al. (2015) and Aboulkhair et al. (2016) reported that the heat treatment has a strong effect on the mechanical properties of this alloy, improving the fatigue performance, for both machined (Brandl et al. (2012)) and as-built samples (Brandl et al. (2012); Maskery et al. (2015); Aboulkhair et al. (2016)). Interestingly, Aboulkhair et al. (2016) showed that the machining of the samples did not improve the fatigue properties for higher stress levels (above maximum stress of 157MPa), showing a pronounced effect only at lower stress values. Similar indications on the effect of surface finishing on the fatigue properties of the AM AlSi10Mg alloy, were given by Mower and Long (2016). These authors reported that neither mechanical nor electro polishing improved significantly the fatigue life, measured in rotating bending mode. Mower and Long (2016) justified this observation with the presence of defects in the bulk of the specimens. Furthermore, the build orientation was seen to affect the fatigue properties, showing an impaired behaviour for the samples built in z direction (Brandl et al. (2012); Tang and Pistorius (2017)). Following the work reported in literature and with the aim of enlarging the understanding of fatigue properties in AM AlSi10Mg alloy a cooperative activity was designed by EOS, RUAG and ESA. This work comprised studying the influence of building direction, platform temperature, powder layer thickness, surface finish and heat treatment on the fatigue properties of the AM specimens. As a complementary assessment, the defect population of the samples was studied through micro X-ray Computed Tomography (µCT). The authors expect with this work to encourage further studies on the effect of the process parameters on the fatigue behaviour, envisioning the adjustment and improvement of the Processing-Structure-Properties-Performance relationships of AlSi10Mg produced via Additive Manufacturing. AlSi10Mg fatigue specimens were produced in the horizontal and vertical orientation, see Fig. 1a, on an EOS M400 machine. Table 1 gives an overview of the 12 different groups manufactured and of the respective conditions used. 7 specimens were manufactured for each group using a scan strategy combining an inner hatch / filling parameter and a contour parameter which is illustrated in Fig. 1b. An exception is condition ID 7 where the contour parameter was not performed. Two different parameter sets were applied for the inner hatch / filling: a set with 30 µm layer thickness with an energy input ( E ) of approximately 50 J.mm 3 , used to produce high quality parts, and a set with 90 µm layer thickness at about 22 J.mm -3 for a drastically increased build up rate. The energy input ( E ) is described in Eq.1 where P is the laser power (in J.s -1 ), v is the scan velocity (in mm.s -1 ), d h is the hatch distance (in mm) and h th is the layer thickness (in mm) (Thijs et al. (2010)). Both inner hatch parameter sets used a scan regime which rotates the hatch orientation about 67° from layer to layer to reduce the texture in the grain structure. Furthermore the majority of the samples were produced on a pre heated build plate at 165 °C except condition ID 1 where the base plate heating was set to 35 °C. = /( × ℎ × ℎ ℎ ) (1) 2. Materials and Methods 2.1. Specimen manufacture