PSI - Issue 34

Federico Uriati et al. / Procedia Structural Integrity 34 (2021) 184–190 Author name / Structural Integrity Procedia 00 (2021) 000–000

188

5

state before fatigue testing. The surfaces of the calibration beams are also left in the as built state and underwent the progressive support cutting procedure to generate the inherent strains coefficient used by the process simulation software. More details of the calibration and residual stress analysis are given elsewhere. 4. Component testing The structural function of the lower suspension arm has driven the selection and development of the present additively manufactured component. Therefore, the fatigue performance is a crucial aspect to be taken in consideration for part design and manufacturing process qualification. AM technology is influenced by many factors and is important to achieve a comprehensive knowledge of all the aspects that could contribute to improve or worsen the overall mechanical performance of the part of interest. Since the behavior evaluated from numerical simulation of the model and the experimental results completed on the real part must be proven consistent, the authors planned to perform laboratory tests and correlate collected results. The test configuration was developed to fit within the dynamic testing machine MTS 810 (MTS System, MN, USA) while replicating the loading condition experienced by the lower suspension arm in service, Fig. 5a. A critical issue is the reproduction of the actual boundary conditions. Two different solutions for the cylindrical joints were tested: i) nylon bushing and steel pin as shown in the model of Fig. 5b and ii) close fitting steel bushing in reamed aluminum hole. FE simulation showed that the selected joint type influences the stress distribution in the vicinity of the boundary.

a) b) Figure 5 a) experimental setup; b) model reproducing the experimental configuration

Dynamic loading has been then applied to optimized lower arms to determine their actual fatigue strength. Although yet limited experimental evidence is presented here, additional tests are on-going, and the main conclusions are expected to hold true. Constant amplitude loading at R = 0.1 with reference maximum force F max = 3.75 kN was applied instead of a realistic variable amplitude loading to simplify the assessment phase. Fig. 6b shows the fatigue crack that developed in about 156000 cycles. Contour plots of the loaded component obtained by FEA (Altair Hypermesh, Altair Engineering USA) and shown in Fig. 6a confirmed a matching of critical location with crack localization. The local maximum elastic stress was estimated in σ max,comp = 285 MPa while local residual stresses from part fabrication simulation were negligible.

a)

b)