ABSTRACT
Insufficient data are available to fully understand the effects of metal additive manufacturing (AM) defects for widespread adoption of the emerging technology. Characterization of failure processes of complex internal geometries and defects in metal AM can significantly enhance this understanding. We aim to demonstrate a complete experimental measurement process and failure analysis method to study the effects of AM defects. We utilized simultaneous implementation of tensile tests with high-resolution X-ray computed tomography (XCT) measurements on 17-4 stainless steel dog-bone samples with an intentional octahedron-shaped internal cavity included in the gauge length and also containing much smaller lack-of-fusion (LOF) defects, all generated by a Laser Powder Bed Fusion (LPBF) additive manufacturing process. The LOF defects were introduced by intentionally changing the LPBF default processing parameters. XCT image-based linear elastic finite element (FE) simulations were used to interpret the data. The in-situ tensile tests combined with simultaneous XCT measurements revealed the details of the failure process initiated by additively manufactured rough internal surfaces and porous defect structures, which experienced high stress concentrations. Progressive collapse of ligaments leading to larger pores was clearly observed, and the resulting porosity evolution until failure was quantitatively analyzed. The high stress concentrations were also directly confirmed by the FE simulations. The experimental methods described in this paper enable the quantitative study of the complex failure mechanisms of additively manufactured metal parts, and the image-based FE simulation method is effective for identifying and/or confirming possible failure locations and features.
ABSTRACT
Pore structures of additively manufactured metal parts were investigated with X-ray Computed Tomography (XCT). Disks made of a cobalt-chrome alloy were produced using laser-based powder bed fusion (PBF) processes. The additive manufacturing processing parameters (scan speed and hatch spacing) were varied in order to have porosities varying from 0.1% to 70% so as to see the effects of processing parameters on the formation of pores and cracks. The XCT images directly show three-dimensional (3D) pore structure, along with cracks. Qualitative visualization is useful; however, quantitative results depend on accurately segmenting the XCT images. Methods of segmentation and image analysis were carefully developed based, as much as possible, on aspects of the images themselves. These enabled quantitative measures of porosity, including how porosity varies in and across the build direction, pore size distribution, how pore structure varies between parts with similar porosity levels but different processing parameters, pore shape, and particle size distribution of un-melted powder trapped in pores. These methods could possibly serve as the basis for standard segmentation and image analysis methods for metallic additively manufactured parts, enabling accurate and reliable defect detection and quantitative measures of pore structure, which are critical aspects of qualification and certification.
ABSTRACT
Near real-time visualization of complex two-phase flow in a porous medium was demonstrated with dynamic 4-dimensional (4D) (3D + time) imaging at the 2-BM beam line of the Advanced Photon Source (APS) at Argonne National Laboratory. Advancing fluid fronts through tortuous flow paths and their interactions with sand grains were clearly captured, and formations of air bubbles and capillary bridges were visualized. The intense X-ray photon flux of the synchrotron facility made 4D imaging possible, capturing the dynamic evolution of both solid and fluid phases. Computed Tomography (CT) scans were collected every 12 s with a pixel size of 3.25 µm. The experiment was carried out to improve understanding of the physics associated with two-phase flow. The results provide a source of validation data for numerical simulation codes such as Lattice-Boltzmann, which are used to model multi-phase flow through porous media.