PSI - Issue 54

Nikolai Kashaev et al. / Procedia Structural Integrity 54 (2024) 361–368 Kashaev et al. / Structural Integrity Procedia 00 (2023) 000 – 000

364

4

2.2. Fatigue testing and characterization A Testronic 100-kN resonant testing machine of Russenberger Prüfmaschinen AG was used for the load-controlled uniaxial fatigue tests. The tests were performed in accordance with ASTM E466-07 at a resonant frequency of approximately 80 Hz and a load ratio R F = 0.1 and at room temperature. Specimens with a constant test section of 7.5 mm × 2.0 mm within a length of 20 mm were used (Fig. 3). The same testing machine equipped with the crack length measuring system FRACTOMAT was used for fatigue crack growth (FCG) tests. The crack length measuring system FRACTOMAT is based on the indirect potential drop method and continuously indicates the measuring values. It generates an accurate result of the crack length within the single-digit micrometer range and can also be used to control the propagation of the crack. The tests were performed using the standard compact tension C(T)50 specimens with a thickness of 3.0 mm and a width W of 50 mm. The sketch for the extraction of specimens from the WAAM fabricated structure is shown in Fig. 2(c).

Fig. 3. Geometry of the specimens used for the uniaxial fatigue test. All dimensions are in mm.

Load-shedding in accordance with ASTM E 647 was employed to determine the fatigue crack growth threshold. Basically, fatigue cracks were initiated at intermediate values of the stress intensity factor range  K and subsequently, the load was continuously reduced in dependence of the crack extension. The normalized gradient of the stress intensity factor K -gradient was set according to the recommendations of ASTM E 647 in order to avoid an influence of the plastic zone size on the crack growth rate and the fatigue crack growth threshold. For the metallographic analysis, cross-sections from the WAAM-fabricated walls were extracted, ground, polished, and etched with Kroll´s solution. The microstructure was analyzed with an optical microscope (OM) Leica DMI5000 M (Leica Microsystems GmbH). The microstructure of the WAAM-fabricated structure in the "as built" condition consists of metastable martensite with acicular  ' as a result of diffusionless  →  ' transformation (Fig. 4(a)). The present microstructure indicates that high cooling rates must have been present. For the diffusionless martensitic transformation of Ti-6Al-4V, cooling rates of at least 410 K/s are required (Ahmed and Rack, 1998). The subsequent heat treatment at 920 °C transformed the martensitic  ' structure into a fine lamellar structure with  lamellae (Fig. 4(b)). Similar microstructures have been reported by Wang et al., 2013 and Syed et al, 2021 for WAAM fabricated Ti-6Al-4V. Fracture surfaces of the fatigue specimens were examined using a KEYENCE digital microscope VHX-7000 of KEYENCE Deutschland GmbH. In this analysis, the size of the crack-initiating defect and the distance from the center of the defect to the nearest specimen surface in the case of the internal defect were measured.

Fig. 4. (a) Obtained as-built microstructure and (b) the microstructure after the applied post-building heat treatment.

3. Fracture mechanics framework Fatigue life prediction can be performed based on a crack growth equation, which is used to calculate the length of a fatigue crack a , that grows due to cyclic loading. For this purpose, the simplified NASGRO equation (NASA, 2001) was used in the present work, in which only the short-crack and the stable crack growth were considered (Eq. 1):