PSI - Issue 75

Arthur THIBAULT et al. / Procedia Structural Integrity 75 (2025) 509–518 Arthur THIBAULT/ Structural Integrity Procedia (2025)

512

4

Furthermore, Vickers microhardness tests (HV0.1) revealed significant differences between the weld bead and the base metal (BM), with average hardness values over 10 measurements of 110.2 HV for the base material with a standard deviation of 4.08 and 96.1 HV for the weld with a standard deviation of 3.96. The material present in the weld bead is therefore a mixture of aluminium 6061 and 4043, and although these two grades have similar chemical compositions, their mechanical properties are not exactly the same. Microhardness traverses show an abrupt transition between the weld bead and the base metal, suggesting that the HAZ (Heat-Affected Zone) is very narrow (Fig. 2c). Indeed, it [the HAZ] is not discernible from the hardness traverses, even though the indents are only 250μm apart. These two findings support the need for using different constitutive laws for the base metal and the welded zone. Residual stress measurements were conducted in collaboration with the Laboratory of Mechanical and Material Engineering LASMIS (UTT) in order to quantify the influence of welding. The samples tested had already undergone the T6 heat treatment. The objective of these measurements was to investigate whether residual stresses persisted in the weld bead even after heat treatment. These measurements were performed on two butt-welded tubes. Measurements were taken at a series of points around the weld bead to check the homogeneity of residual stresses within it. Additionally, measurements were taken along a line starting from the centre of the bead and moving away from it, in order to quantify the zone of influence of the weld on the residual stress field. The results show low-magnitude residual stresses concentrated in the weld bead, as the measured stresses drop to practically 0 MPa once outside the weld. This highlights a difference between the welded zones and the rest of the structure. It is also noted that the measured stresses are predominantly compressive, and that normal stresses are significantly dominant over shear stresses. The values presented in Table 1 correspond to the measurements taken around the weld bead, have been normalised, and are expressed as a ratio of the maximum value. Table 1. Measurement of residual normal and shear stress around the weld bead Measurement 11 (MPa) 22 (MPa) 12 (MPa) 13 (MPa) 23 (MPa) 1 (0°) -0,5 -0,42 -0,01 -0,07 0,01 2 (90°) -1 -0,53 0,03 -0,30 -0,03 3 (180°) -0.36 -0,72 0,01 -0,04 -0,01 4 (270°) -0.73 -0,65 0,08 -0,10 0,02 5 (270° 2) -0,78 -0,73 0,08 -0,10 0,03 Average -0,67 -0,61 0,04 -0,12 0,004 These various preliminary measurements have highlighted metallurgical and mechanical differences between the welds and the other parts of the structure. They therefore highlight the importance of considering a specific local constitutive law for the welded zones, even though these zones are composed of an alloy similar to the rest of the structure and undergo the same post-assembly heat treatment. Micro-tensile tests Once the preliminary measurements were completed, the objective was therefore to determine a local elasto-plastic constitutive law in the welded zones. To this end, micro-tensile tests instrumented with Stereo Digital Image Correlation (Stereo DIC) were conducted. For these tests, samples were extracted from scooter decks. This choice stems from the fact that the scooters used are also made of aluminium 6061, assembled using the same welding method, and undergo the same heat treatment as the bicycle frames. To ensure sample representativeness, the scooter decks used were taken directly from production at a Decathlon supplier. After the assembly and heat treatment of the decks, test specimens were extracted from the weld region using Electrical Discharge Machining (EDM) in order to minimise alteration of their mechanical properties (Fig. 3). 1.2.