Abstract

Three groups of Ti–6Al–4V specimens were printed on an industrial laser powder bed fusion system using three volumetric energy densities (47, 38 and 30 J/mm3), called High, Optimal and Low (HED, OED and LED), and subjected to stress relief and annealing heat treatments. The reference OED group had the smallest amount of porosity <0.005%, while the two other groups (HED and LED) had similar levels of porosity (0.03–0.05%), but formed by flaws of different nature: gas pores (HED) and lack-of-fusion defects (LED). First, chemical composition, microstructure and porosity features of each group of specimens were analyzed. Then, uniaxial tensile testing, axial fatigue testing, fractography analysis and fatigue crack growth testing were carried out. Similar chemical compositions and bi-modal microstructures were observed in all the specimens. Micro-computed tomography (μCT) confirmed the targeted porosities and types of pores for each group of specimens, with only the LED specimens having a significant number of pores larger than 0.07 mm. Tensile properties were similar for all the specimens with the exception of elongation to failure, which was significantly lower for the LED specimens as compared to the HED and OED specimens (∼11 vs. ∼14%). Fatigue testing of the HED, OED and LED specimens showed runout (107) stress ranges (Δσ) of, respectively, 450, 495 and 360 MPa, while fatigue crack growth testing resulted in a threshold stress intensity value of ΔK = 3.9 MPa m½ for all the specimens. A Kitagawa-Takahashi diagram was built as a function of the μCT-measured defect size and showed to be an appropriate while conservative tool for the defect tolerant design of Ti–6Al–4V LPBF components. A strong influence of the defect size and proximity to the surface on the fatigue behavior was observed, while the defect shape was shown to have little incidence on this property. Overall, this study showed that the influence of pore size on the fatigue resistance becomes particularly significant once a certain pore size threshold (√area ≈0.2 mm) is exceeded, and that this eventuality coincides with a ∼20% decrease in the volumetric laser fusion energy density as compared to its optimum value.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call