Novel Method of Predicting Deposition Efficiency in Cold Spray by Incorporating Sphericity into a 1D Analytical Model

Sensitivity of the cold spray (CS) additive manufacturing process to the variabilities in its process parameters is still being understood. Among the process parameters, particle morphology can have significant implications on drag forces and therefore the particle impact velocity. This in turn affects the deposition efficiency (DE) and the quality of the final products. In this work, a new approach is introduced for computing DE by incorporating particle size and size dependent sphericity distributions into a 1D analytical multiphase gas dynamic model that uses the Haider and Levenspiel drag coefficient that compensates for particle sphericity. To this end, size distributions and size-based sphericity distributions of Al alloy 6061 (Al6061), Ti, and Cu particles were measured using static optical microscope images. The analytical model was then used to estimate size and sphericity dependent velocity curves for helium and nitrogen spray cases for aluminum alloy 6061, and nitrogen spray cases for Ti and Cu powders, which showed that sphericity dependent velocity variations can be on the order of hundreds of meters per second. Following velocity predictions, a DE calculation method was developed by calculating the ratio of particles that exceed a semi-empirical critical velocity. Size informed velocity predictions were compared with predictions for using a spherical particle assumption and particle velocity measurements conducted by particle velocimetry. Ultimately, perfectly spherical particle assumption showed an underestimation of particle velocity while the sphericity data informed velocity predictions were successful in capturing the size versus velocity mean and the velocity range observed in measurements with acceptable accuracy. Furthermore, for nitrogen spray cases, DE was measured experimentally and compared to the newly introduced DE estimation methods, which showed that critical velocity is underestimated for non-spherical particles. Thus, critical velocity equation was calibrated for a more accurate DE estimation for given cases.