Date | Time
07/06/2016 | 14 h 00 min - 17 h 00 min
Ross Cunningham, Samikshya Subedi, Sneha Narra, Tugce Ozturk, Harshvardhan Jain, Elizabeth Holm, Brian DeCost, Robert Suter, Jack Beuth, Anthony Rollett
Carnegie Mellon University
5000 Forbes Ave., Pittsburgh PA 15213, USA
Additive manufacturing (i.e. 3D printing) is increasingly being implemented for making structural metallic components, yet the nature of microstructural defects and their influence on the mechanical properties are still work in progress. It is important to understand the microstructure and, in particular, porosity in additively manufactured metallic parts as well as the powders used as feedstock in many of the machines. Apart from surface flaws, pores are the primary origin of fatigue failures under cyclic loading. The morphology and location of these pores can help indicate their cause; lack of fusion or keyholing pores with irregular shapes can usually be linked to incorrect processing parameters, while spherical pores suggest trapped gas. Synchrotron-based 3D X-ray microtomography was performed at the APS on additively manufactured samples of Ti-6Al-4V using electron beam powder bed and Al-10Si-1Mg using laser powder bed. The spatial and size distributions of the porosity over a range of processing conditions were determined. Five Ti-6Al-4V samples were fabricated with parameters varied to produce a range of melt pool areas. Imaging samples were sectioned from the bulk and the contour-bulk interface. Similarly, three samples of Al-10Si-1Mg were made with varying process conditions. Marked variations in the type and amount of porosity were observed as a function of the melt pool area.
Beyond measurements of porosity, 3-D printed parts are known to have residual stress as a consequence of the shrinkage that occurs on solidification as well thermal contraction. Thanks to recent advances in high-energy (synchrotron) x-ray methods, a combination of near-field and far-field high-energy diffraction microscopy (HEDM) enables the mapping of both 3-D grain structure and the lattice strains. Preliminary measurement results are presented for printed Ti-6Al-4V. There are many ways in which data analytics could be applied to additive manufacturing. An example is given of the application of machine vision to the classification of different types of metal powders.
Keywords: Additive Manufacturing, 3D Printing, Crystal Plasticity Simulation, High Energy Diffraction Microscopy (HEDM), Computed Tomography (CT), Machine Vision
Acknowledgments: Jon Almer, Edward Cao, Ross Cunningham, Peter Kenesei, David Menasche, Tugce Ozturk, Suraj Rao, Hemant Sha, Samikshya Subedi, Robert Suter, and Xianghui Xiao are thanked for their contributions. NSF, DOE, APS(ANL), the Commonwealth of Pennsylvania, and America Makes supported various aspects of the work.