Acoustic monitoring of additive manufacturing for damage and process condition determination
Is Version Of
Additive manufacturing (AM) is the process of forming materials in a layer-by-layer fashion, as opposed to more traditional, subtractive manufacturing methods. The method enables the inclusion of features not possible with classical methods such as complex internal and external geometry, and gradations of material composition for some processes. While these geometry and composition variations are an enabling capability for design freedom and customization, uncertainty remains regarding the "quality" and in the resulting material properties and defect distributions for AM parts. There remains a need to identify appropriate means and methodologies to inspect them. AM is well suited for low production volume, complex, high value components and thus, real-time, in-situ characterization on a part-by-part basis of these materials has become of interest to academia and industry. Acoustic methods have been utilized for monitoring cutting, milling, welding, and laser processing of metals and polymers in the past. In this work, an acoustic monitoring array was utilized to monitor directed energy deposition of Ti-6AL-4V onto a steel substrate. Temporal waveforms were recorded intermittently and passively, and later analyzed using temporal and spectral methods. Metallographic analysis and comparison of crack densities with acoustic metrics are shown to correlate well as a material damage indicator. Low amplitude process noise is also shown to correlate with the process state, for which 3 variations around nominal deposition parameters were tested.
This proceeding may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing. This proceeding appeared in Koester, Lucas W., Hossein Taheri, Leonard J. Bond, and Eric J. Faierson. "Acoustic monitoring of additive manufacturing for damage and process condition determination." AIP Conference Proceedings 2102, no. 1 (2019): 020005, and may be found at DOI: 10.1063/1.5099709. Posted with permission.