"Modeling and structural simulation of thin-walled additively manufactured components considering material anisotropy and inhomogeneity"
Modeling and structural simulation of thin-walled additively manufactured components considering material anisotropy and inhomogeneity
Sprache des Titels:
The unrivaled design freedom of additive manufacturing makes it a valuable technology for industries. The premise to exploit this benefit is the accurate structural response prediction prior to part fabrication. However, the unconventional mechanical material behavior of additively manufactured structures that arises in various processing technologies, still constitutes a major challenge for this task. Recently, for instance, it was shown that the mechanical properties of laser sintered polyamide 12 not only exhibit considerable degree of anisotropy but furthermore depend drastically on the local wall thickness of the generated part. At present, the interaction between anisotropy and thickness dependency remains unclear. It is particularly critical for thin-walled structures as it may lead to components that either deform excessively and fail prematurely or to overly conservative dimensioned parts for the intended load cases. Hence, it was the aim of this thesis to improve the validity of structural finite element simulations for thin-walled additively manufactured parts by considering the local effects of orientation and wall-thickness. Therefore, the peculiar mechanical material behavior was characterized in extensive tensile testing campaigns, in which both, pure and short-fiber-reinforced polyamide 12 material was considered. To gain a more complete understanding of the observed behavior, porosity characteristics including spatial distributions and pore morphology, were analyzed by means of X-ray micro-computed tomography for the pure polyamide. Subsequently, of the obtained experimental data material models were derived that account for the influence of thickness and orientation. Thereby, different approaches were pursued for the pure and reinforced material. For the latter, the model was extended to additionally capture the scatter inherent in the mechanical material properties. Next, a script for automatic mapping of the inhomogeneous material properties in shell meshes based on local finite element orientation and thickness was developed. In a subsequent step, a property clustering algorithm was implemented in order to improve efficiency of the approach. Consequently, finite element simulations were performed to predict the load displacement behavior of different structures on sub-component level. For the reinforced material, additionally, a thickness dependent orthotropic failure criterion was proposed. For validation purposes, physical thin-walled parts were fabricated via laser sintering and loaded in experimental setups. The tensile tests disclosed distinct losses in Young?s modulus with decreasing wall thicknesses for all coupon build orientations. Also, strength values exhibited a strong dependency on thickness in both material variants. The microstructural porosity analysis delivered valuable insights into the origins of the peculiar macroscopic structural response and yielded a better understanding of the material as well as potential for improvement of the laser sintering process. The comparisons of finite element analyses with mapped inhomogeneous and conventional constant properties disclosed considerably improved prediction of stiffness and failure. For the latter, however, substantial deviations between simulation and physical experiment remained, indicating that further research is necessary to effectively asses the load bearing capacity of thin-walled additively manufactured structures. The efficiency of the mapping strategy was significantly improved by use of the clustering algorithm. Furthermore, the variability observed in the structural response of loaded parts could be reproduced by the stochastic simulations. Ultimately, it is worth noting that the method is not only applicable to laser sintered polymers but relevant for all structures, where anisotropy and thickness must be considered.