Ultrasonic Evaluation of Laser Scanning Speed Effect on the Spectral Properties of Three-Dimensional-Printed Metal Phononic Crystal Artifacts

Abstract

Additive manufacturing/three-dimensional printing (AM/3DP) processes promise a flexible production modality to fabricate a complex build directly from its digital design file with minimal postprocessing. However, some critical shortcomings of AM/3DP processes related to the build quality and process repeatability are frequently experienced and reported in the literature. In this study, an in situ real-time nondestructive monitoring framework based on the dispersive properties of phononic crystal artifacts (PCAs) to address such quality challenges is described. Similar to a witness coupon, a PCA is printed alongside a build while it is interrogated and monitored with ultrasound. A PCA is substantially smaller than the actual build. Due to its periodic internal structures, a PCA creates pass and stop bands in its spectral response, which are sensitive to the variations in its process and material parameters. These periodic structures, representing the geometric complexities of an actual build, are designed for a specific monitoring objective(s) in AM/3DP. As a model application, in this demonstration study, the effect of the laser scanning speed of a slective laser melting (SLM) printer on the spectral properties of metal PCAs (mPCAs) is ultrasonically evaluated offline. The dependency of the pressure and shear wave speeds, the apparent Young's and shear moduli, and Poisson's ratio on the scanning speed are quantified, and it is found that they are highly sensitive to the laser scanning speed of an SLM printer. The sensitivity of the peaks of the pressure and shear spectral waveforms acquired for the identical mPCA designs printed on the same build plate with the same process parameters is also quantified. For powder-based AM/3DP technologies, where scanning speed is among the crucial process parameters such as laser power and bed temperature, the reported correlations between scanning speeds and the mechanical and spectral features of the mPCAs are expected to be instrumental in developing in situ real-time monitoring systems.

Keywords: in situ real-time process monitoring; laser scanning speed; phononic crystal artifacts; selective laser melting (SLM); stainless steel 316L; ultrasonic process monitoring.

FIG. 1.

(a) Schematic diagram of the experimental setup and ray tracing diagrams in pulse-echo mode with a BP and an mPCA sample, (b) schematic diagram of the cross section of an mPCA sample with its periodic internal structure, and (c) an image of the mPCA sample sets bonded to the bottom side of the BP, microscopic view of the samples with a scale bar of length 800 μm at scanning speeds p_s (d) 800 mm/s, (e) 1000 mm/s, and (f) 1200 mm/s, (g) selected windows in the pressure waveforms based on the reflections from the BP-mPCA interface $W\begin{array}{c}P\\ BP\\ \end{array}$ (solid lines) on the BP waveforms and mPCA layer–layer interfaces and the bottom surface–air interface $W\begin{array}{c}P\\ S-BP\\ \end{array}$ (dashed lines) on a subtracted sample waveform, and (h) selected windows in the shear waveforms based on the reflections from the BP-mPCA interface $W\begin{array}{c}S\\ BP\\ \end{array}$ (solid lines) on the BP waveform and mPCA layer–layer interfaces and the bottom surface–air interface $W\begin{array}{c}S\\ S-BP\\ \end{array}$ (dashed lines) on a subtracted sample waveform. BP, build plate; mPCA, metal phononic crystal artifact.

FIG. 2.

Normalized pressure (a) and shear (b) waveforms of the mPCA sample sets with the BP response waveform (dotted line) at three levels of scanning speeds (p_s = 800, 1000, and 1200 mm/s).

FIG. 3.

Normalized subtracted pressure (a) and shear (b) waveforms of the mPCA sample sets with the BP waveform (dotted line) at three levels of scanning speeds (p_s = 800, 1000, and 1200 mm/s). The sample responses are highlighted with gray-colored boxes.

FIG. 4.

Scaled normalized subtracted pressure (a) and shear (b) waveforms of the mPCA sample sets with the BP waveforms (dotted) at three levels of scanning speeds (p_s = 800, 1000, and 1200 mm/s). The asterisks (*) represent the responses from the bottom surface of the samples.

FIG. 5.

The pressure (a) and shear (b) spectral responses (solid lines) of the mPCA sample sets with the BP spectral responses (dotted) at three levels of scanning speeds (p_s = 800, 1000, and 1200 mm/s). The shifted peaks (⚪) are marked as P₁, P₂, and P₃ in the pressure spectral responses, and S₁, S₂, and S₃ in the shear spectral responses.

FIG. 6.

Average frequency shifts of the spectral response peaks (P₁ to P₃) of the mPCA sample sets at three scanning speeds (p_s = 800 [●], 1000 [■], and 1200 [ × ] mm/s) from the spectral responses of subtracted pressure waveforms (a) and S₁ to S₃ from the spectral responses of the subtracted shear waveforms (b).