Automatic geometry alteration when designing for metal additive manufacturing
Paper in proceedings, 2020
During the last decades, the aerospace industry has been interested in converting some of its production from traditional manufacturing to metal additive manufacturing. This technology is attractive to the aerospace industry due to the low production volumes and the need to minimize product mass that characterize the industry. The transition from traditional to additive manufacturing technologies requires designers to take heed of a new set of design and manufacturing constraints. A well-known limitation for the implementation of additive manufacturing is the need for support structures which increase material consumption and post processing time. The research presented in this paper suggests a novel algorithm for automatically altering the geometry of the product to reduce the need for support structures. The algorithm operates directly on the vertices of the 3D mesh by relocating them to mitigate overhang issues. A proof of concept in the form of a software prototype was developed using the Python programming language to demonstrate the algorithm, and how it can be applied to 3D models. The prototype software utilizes a 3D STL-file as input, and it outputs an altered file of the same format. In the process of developing the software prototype some issues surfaced, the most prominent of which was the inability of the algorithm to handle overhangs parallel to the substrate of the powder bed printer. Such overhangs are also referred to as 0°-overhangs due to their normal vector being parallel to the z-Axis, thus resulting in an overhang of 0°. Aside from that the algorithm also struggled with models of high complexity, which in some cases resulted in invalid or malformed STL geometries. The research conducted for this paper indicates that further research and development is required. Future research concerning automatic geometry alteration should focus on how to deal with the 0°-overhangs. It should also involve investigating the possibility to alter the geometry of CAD-models instead of STL-files, as CAD-models can contain more information, such as constraints. Information of that nature could enable the algorithm to take customer requirements into consideration as it alters the geometry, thus improving its ability to maintain product functionality.