Abstract
Metal Additive Manufacturing is a promising technology for the manufacturing of tomorrow. Allowing for complex, customized geometry and less wastage than traditional machining, it is already seeing applications in some industries, such as aerospace, marine and offshore, medical, automotive, etc. Although it is becoming increasingly popular, metal AM still has some major drawbacks, which have slowed widespread adoption. Current technologies have a relatively low production speed, a limited size of parts that can be produced, and are expensive to own and operate. Wire Arc Additive Manufacturing (WAAM) is an emerging technology that has the potential to address these drawbacks. It can build large parts quickly and at a lower cost than powder metal AM technologies. However, the WAAM process has many variabilities and is difficult to control due to the many process parameters involved. Current solutions to address this involve coupling WAAM with a subtractive process to remove defects. However, the two processes need to be better integrated, as each process needs to have information about the actual geometry that resulted from the previous process. This is especially so for the subtractive process to determine which areas in the printed part to remove. This dissertation proposes a scanning system that can report in-situ geometrical data of a part being printed. This would allow WAAM and Hybrid WAAM systems to run in a closed loop, allowing print errors and deviations to be detected, and compensated for in the next step. This would improve the WAAM process's viability and make it attractive for adoption by the industry. The proposed system uses a laser line triangulation scanner to generate a point cloud of the printed part. The laser scanner is mounted on a robot arm, which allows it to be moved through scan toolpaths with high accuracy. Additionally, the six degrees of freedom of the robot arm allows the scanner to be placed in different orientations, enabling the scanning of complex parts. The proposed scanning system is able to produce complete three-dimensional scans of parts with an overall accuracy of 0.5 mm when compared to a GOM ATOS 3D scanner. It is also capable of scanning the top profile of a print to determine the height profile of the current layer, with an accuracy of 0.05 mm. Lastly, an analysis was performed on the business viability of WAAM technology, focusing on Hybrid WAAM systems equipped with a scanner for in-situ geometry reporting, such as the one proposed in this dissertation. Business viability was explored through the possible business models that can be employed, the current market and its external factors, competitors to the technology and strategies for growth.