PSI - Issue 57

Martin Matušů et al. / Procedia Structural Integrity 57 (2024) 327 – 334 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

329

3

2. Fatigue strength analysis

2.1. Description of specimens and experimental setup

This study investigates the fatigue behavior of AlSi10Mg aluminum alloy manufactured additively using the Laser Power Bed Fusion (L-PBF) technology. Hourglass-shaped fatigue specimens with a critical cross-section diameterof 9 mm and a transition fillet radius of 60 mm were used for the fatigue experiments (see Fig. 1a). The fatigue specimens had a machined head with M18x1 thread for fixing it to the Amsler HFP 422 resonant pulsatorwith the 100 kN load cell. The loading was imposed by the pulsator only in tension ( R = 0.1). Specimens with a rectangular critical cross section of 6x3 mm and a length of the reduced section of 36 mm were used for static tensile tests (see Fig. 1b). The gripping length was set to 30 mm.

a)

b)

Fig. 1. (a) Fatigue test specimen; (b) Static tensile specimen with thickness of 3mm.

All specimens were printed in the vertical direction within one build cycle by Concept Laser M2 printer (with printing parameters accordingto [6] experimentalcampaign regarding printing parameteris well presented here [9]). Four groups of specimens (differing by different heat treatment conditions) were tested, with each group consisting of 14 specimens designated for fatigue experiments and 3 for static tests. Eleven specimens from each fatigue related group were used to establish the S-N curve, while the remaining 3 were used to analyze the heat dissipation via the S H tests monitoring the surface temperature throughout the range of tested amplitudes. All specimens were left with surfaces in the as-built state, no machining (except of threaded heads) was applied. Throughout the text, the series are described using shortened names and color coding according to Table 1. Table 1. Series description using color coding to distinguish heat treatment of the specimens. ID-code Heat Treatment Color coding NoHT Specimens were left as built without any additional heat treatment. Red

Specimens were subjected to a conventional heat treatment recommended by the supplier of the printing powder, General Electric (GE), which also manufactures the used 3D printer. Specimens were heated to 240 °C and held for 6 hours, followed by cooling on air. Blue Specimens were heated to 200 °C for 2 hours, followed by air-cooling to ambient temperature. Green

T240

T200 T300

Specimens were heated to 300 °C for 2 hours, followed by water-cooling.

Orange

The surface temperature of the specimens during the experiments was measured using an infrared thermal camera. The Flir A315 was used at CTU in Prague, while the Fluke RSE600 was used at OTH Amberg-Weiden. Both cameras had a NETD (thermal sensitivity) of 50 mK and 40 mK, respectively. To ensure accurate temperature measurements on the specimen's surface, a high emissivity paint from LabIR was applied. The emissivity of the paint was measured to be 0.963 at ambient temperature. To assess the surface quality of the specimens, surface roughness measurements were conducted using the Perthometer M1 from Grant for tensile specimens ( R a = 2.11 μm , R z = 14.19 μm ) and MarSurf LD 120 drive unit for fatigue specimens ( R a = 3.09 μm , R z = 24.28 μm ).