Selective Laser Melting is an additive manufacturing (AM) process which uses high-powered lasers to rapidly melt and solidify metal powder into three-dimensional structures. The SLM process is highly complex and can have undesirable results such as stochastic defect formation and non-equilibrium microstructures. To fully understand the AM process and ensure a high-quality manufacturing process, novel high speed sensing methods are needed to capture key physical phenomena associated with the AM parts at high resolution. Through our seed funding from Hopkins Extreme Materials Institute (HEMI) our team has developed novel spectroscopy techniques capable of measuring the build process along the laser path. This technology has contributed to a greater understanding of laser-materials interactions, and has provided insight into solidification and defect formation as a function of processing conditions. Our sensor captures the thermal spectra of the molten powder, which is closely matched to the laser focal spot size. The key to our hardware technology is our home-built, high-speed spectrometer which allows us to measure and estimate the true temperature of the melt pool regardless of changes in material geometry, shape or emissivity.
Complementary to our novel hardware are the Machine Learning (ML) tools developed, to interpret these signals and reveal trends or anomalies that are not visible to the human eye or via post-mortem inspection of the printed components.