A new computational method is proposed to simulate the balling phenomenon during powder bed fusion (PBF) metal additive manufacturing processing. The method combines the discrete element method (DEM), which simulates powder behavior, and smoothed particle hydrodynamics (SPH) method, which simulates fluid and elastoplastic behavior. Balling is a phenomenon by which spherical defects form on a product surface because of surface tension and poor wettability between the molten metal and the surrounding powder or solidified material. This phenomenon is particularly evident in materials with high surface tension, which can cause discontinuities in the melting path and which can affect manufacturing quality. Simulating such complex behavior necessitates consideration not only of the molten metal behavior (including surface tension), the metal powder and solidified metal behaviors, and the melting and solidification phenomena: it also includes their mutual interactions. The proposed method can simulate these processes. Moreover, performing simulations under conditions of various combinations of laser power and scanning speed confirm qualitatively that the proposed method captures the balling phenomenon. The proposed method can capture trends of experimentally obtained results obtained from earlier studies, i.e., under the same laser energy per unit area, the balling phenomenon becomes more likely to occur as the scanning speed and laser power increase. This approach can be a valuable tool for optimizing manufacturing conditions and for improving the quality of metal additive manufacturing processes.
Tokunaga et al. (2026) studied this question.