Particle velocity measurement in powder gas jets of coaxial powder nozzles for laser material deposition

Laser material deposition (LMD) is a process in which a laser beam creates a melt pool on the surface of a substrate while a powder nozzle delivers a powdered additive into the melt pool, where the powder material then melts. A dense, metallurgically bonded layer forms once the melt solidifies. The velocity of powder particles in LMD is one of the key factors that influence the amount of particles' energy absorption (along their individual trajectories) through the laser beam on the way between the powder nozzle and the substrate. The amount of energy that is absorbed by the powder particles above the melt pool determines the temperature on the substrate surface and, hence, influences the formation of the melt pool and the deposition of the powder material. Data about the particle velocity are, therefore, essential for modeling the LMD process-not only for physical simulation where a particle's energy input can be integrated along its trajectory, but also in an experimental environment of process parameter studies, where understanding the interdependencies between the process parameters is crucial. In this study, a setup using a high-speed camera and an illumination laser is used to measure the velocity of powder particles in the powder gas jet of a coaxial powder nozzle. Several parameters that are known to influence the particle velocity are varied: Feed gas rate, shielding gas rate, nozzle geometry (width of annular gap), and powder mass flow rate. In this study, four different powder types are used. The influence of these process parameters on the particle velocity is measured. Four different methods for tracking individual particles and calculating the velocity distribution within the powder gas jet are used and compared: Manual frame-by-frame particle tracking and manual evaluation from multiple exposures in single frames, as well as particle tracking velocimetry and particle image velocimetry (PIV), which incorporate region-of-interest boxes into the algorithm. Sufficient accordance as to the measurement results is found in comparing the four methods. Further, using the highly automatable PIV method, the influence of main process parameters on particle velocity is measured. Out of the examined parameters, the feed gas rate is found to have the most immediate impact with a linear correlation to particle velocity. A correlation between different powder particle size distributions and measured particle velocities is shown as well.