PSI - Issue 54

Jenny Köckritz et al. / Procedia Structural Integrity 54 (2024) 423–430 J. Köckritz/ Structural Integrity Procedia 00 (2019) 000 – 000

425

3

(Bolchoun et al., 2014). Fatigue assessments for multiaxial loaded welds with several methods and comparisons to experiments are performed in Mei et al. (2021), Bolchoun et al. (2014) and Brahami et al. (2019). For tube structures, Hobbacher (2016) refers to Zhao et al. (2001), which is only applicable for steel and can therefore not be applied for a multiaxial loaded aluminum tube structure. Nomenclature 1 fatigue exponent r bending ratio SR1 fatigue coefficient N load cycles R load ratio stress amplitude force amplitude log. standard deviation 2. Materials and Methods 2.1. Loading of cargo bicycle welds The cargo bicycle frame was tungsten inert gas welded from EN AW 7020 aluminum tubes with a T6 heat treatment. The welding was performed manually with an EN ISO 18273 S Al 5356 weld wire. No heat treatment was applied to the finished frame. The cargo bicycle, shown in Fig. 1 (a), was assessed according to the requirements for cargo bike frames in the DIN 79010 (2020). The load cases involve a cyclic pedaling at each pedal with 1200 N and the load ratio R = 0 , which leads to a torsion of the frame along its moving direction. Braking or impacts with obstacles lead to horizontal forces with R = -1 . The horizontal force amplitude was calculated as 1225 N. The cyclist and cargo are accelerated while cycling over uneven ground. This leads to cyclic vertical forces, which are tested separately according to DIN 709010 (2020) but were considered as a combined load case here. The 1200 N driver load and the maximum loading of 500 N for the cargo were applied to both load platforms simultaneously for increased safety. Lastly, the rear brake moment might cause fatigue cracking at the fallout connection to the frame and thus was also included in the fatigue assessment, although not required. Considering maximum brake deceleration and a safety factor, a brake moment of 1650 Nm was applied. Additional to this constant amplitude load cases, reaction forces at the contact areas of the cyclist and accelerations at the axles were measured on extensive test drives to enable the fatigue assessment under more realistic loads, as presented in Köckritz et al. (2022). A simplified load collective was derived from these measurements for the experimental and numerical investigations of specimen. 2.2. Experimental Setup Experimental investigations were performed on simplified specimen, see Fig. 1 (b) and (c). The fatigue tests under multiaxial load were conducted on welded tube-plate specimen. Additional tests were performed under pure bending on plate-to-plate T-joints, see Fig. 1 (c). The specimens are manufactured in the same way as the frame, therefore no heat treatment or other postprocessing is applied. The specimens have a weld thickness in the tube-plate specimen of about 5 mm and in the T-joint specimen of about 6 mm with small difference between specimen and along the weld.

(a)

(b)

(c)

Fig. 1. Cargo bicycle CityPed® (a), simplified tube-plate specimen (b) and T-joint specimen (c)