Otimização de parâmetros do processo de manufatura aditiva por fusão em leito de pó a laser da liga AISI 316l com foco em produtividade e propriedades mecânicas

Abstract

As Additive Manufacturing of metals developed, attention was drawn to its capabilities to build complex and tenacious components. In this matter, the Laser Powder Bed Fusion (L-PBF) technology has been largely viewed as promising, due to its high precision and ability to produce high quality parts. However, to fulfill its potential, LPBF has yet to enhance its productivity and lower its costs. Research has shown the positive impact of increasing layer thickness on productivity, although only a few works have optimized this parameter in a systematic approach, considering its interactions with other parameters and weighting productivity and mechanical properties simultaneously. In this path, this research did a series of experiments to identify a processing window for stainless steel AISI 316L and to develop a parameters’ set with improved productivity. The methodology used began with single tracks followed by cubes, which its relative density was used to create a regression model. The model indicated some optimized parameters, which were validated trough fabrication. One set of optimized parameters was then chosen for mechanical evaluation and compared to guide samples and standard requirements. The optimized set has shown a less refined microstructure than the guide samples, higher occurrence of defects and more fragile characteristics. Nevertheless, tensile results of optimized samples surpassed standard definitions in four out of five categories, as did the guide samples. For the analysis of machine cost per fabrication, the same build job was used to manufacture with the standard and the optimized set of parameters, and the results showed 56% cost reduction when utilizing the optimized parameters. This result is mainly attributed to the significantly processing time reduction of 59%, entailed by a higher productivity rate and a reduced coating time, that are directly affected by the 5x thicker layer of the optimized parameters set.