Toward In-situ Characterization of Additively Manufactured Parts Using Nonlinear Resonance Ultrasonic Spectroscopy

ABSTRACT

In this study, we investigate the feasibility of using nonlinear resonance ultrasound spectroscopy (NRUS) for in-situ monitoring of additively manufactured (AM) parts i.e., while they are still on the build plate. In NRUS, the test specimen is excited around one or more of its resonance frequencies with increasing driving amplitude. The linear shift in resonance frequency with the increasing driving amplitude is a measure of the constituent material’s hysteretic nonlinearity (α), which is itself related to some degree of micro-damage in the test specimens. We conduct NRUS on test specimens glued to a thick plate. The specimens are excited using a piezoelectric transducer (PZT) adhered to the bottom of the plate. We use this setup to measure the hysteretic nonlinearity parameters (${\alpha }_f$ and ${\alpha }_Q$) of several cylindrical AM specimens fabricated by laser powder bed fusion technique as well as a few non-AM metallic specimens. The measured nonlinearity parameters for the specimens on the build plate (process monitoring mode) are compared to those measured without the build plate (quality control mode). We observe a systematic decrease in the measured nonlinearity when the specimens are tested on the build plate. An analytical study demonstrates that we measure the weighted average nonlinearity of the specimen and build plate, which itself has a lower nonlinearity. Despite the observed difference, the measured nonlinearity parameters of the specimens with and without the build plate are highly correlated. With further investigations, the proposed test setup can potentially be used for characterization of AM parts in situ.

KEYWORDS:

• Nonlinear resonance ultrasound spectroscopy (NRUS)
• additive manufacturing
• hysteresis nonlinearity
• in-situ monitoring
• process control

Acknowledgments

This work was partially funded by the ASNT 2019 fellowship award to all authors. We thank LANL and CIMP 3D for providing the AM specimens used in this study. PM thanks Chun-Yu Ke and Linying Gao for proof reading/reviewing the paper. Submission of this work to RNDE is a requirement of the ASNT fellowship.

Disclosure Statement

No potential conflict of interest was reported by the author(s).