Selective laser melting (SLM) is a promising additive manufacturing (AM) process for high-strength or high-manufacturing-cost metals such as Ti-6Al-4V widely applied in aeronautical industry components with high material waste or complex geometry. However, one of the main challenges of AM parts is the variability in fatigue properties. In this study, standard cyclic fatigue and monotonic tensile testing specimens were fabricated by SLM and subsequently heat treated using the standard heat treatment (HT) or hot isostatic pressing (HIP) methods. All the specimens were post-treated to relieve the residual stress and subsequently machined to the same surface finishing. These specimens were tested in the low-cycle fatigue (LCF) regime. The effects of post-process methods on the failure mechanisms were observed using scanning electron microscopy (SEM) and optical microscopy (OM) characterization methods. While the tensile test results showed that specimens with different post-process treatment methods have similar tensile strength, the LCF test revealed that no significant difference exists between HT and HIP specimens. Based on the results, critical factors influencing the LCF properties are discussed. Furthermore, a microstructure-based multistage fatigue model was employed to predict the LCF life. The results show good agreement with the experiment.
1 January 2018 in International Journal of Mechanical Sciences
An accurate defect assessment procedure is needed to ensure integrity of girth welded steel pipelines while avoiding unnecessary repairs. An integral part of such a procedure is an estimation scheme for the crack driving force. In this work, a modified reference stress solution was developed for the assessment of elastic-plastic pipes with surface breaking defects. For the particular case of interest (reeled offshore pipelines), the loading is phrased in terms of applied strain. The stress-based approach has therefore been extended to the case of strain/displacement control. The results obtained using the modified reference stress solution have been compared with a three-dimensional finite-element analysis. Very good agreement has been obtained. The predictions are also significantly better than those obtained from existing approaches. A comparison between the proposed method and full scale testing will be presented in a separate paper.