| Summary: | Additive manufacturing (AM) is a family of manufacturing techniques that are increasingly being used for the production of high performance or highly customised parts. Greater adoption of AM for end-use manufacturing is hampered by unpredictable process outcomes and limitations on both geometry and build orientation. Powder bed fusion is currently used to manufacture end use parts in both metal and polymer, with greater industrial adoption expected. However, it is widely acknowledged in literature that process control and, therefore, process monitoring, are needed to make powder bed fusion more suitable for widespread use in manufacturing. If a defect can be detected during the process, it can either be repaired in-situ or process parameters can be changed to reduce the occurrence of further defects. A record of the detected defects in a part could also contribute to a quality certification procedure for applications with strict certification requirements, such as aerospace part production.
In this work, point cloud data was collected to demonstrate the feasibility of in-process fringe projection measurements of the powder bed surface of laser sintering. Additionally, a fringe projection measurement system was designed to obtain measurements from the powder bed of a high speed sintering machine. Both sets of data were analysed in a similar manner to differentiate between different types of defects and establish indicators of normal processing conditions. However, due to the high dynamic range of the high speed sintering powder bed, only partial comparisons of data could be made with laser sintering.
This thesis demonstrates a novel in-process monitoring system using fringe projection to provide near real-time feedback on the state of the build, potentially enabling process control decisions to be made. This thesis does not demonstrate the calculation or implementation of process control decisions, but does show how three dimensional measurements of the in-process polymer powder bed can be used to provide high level information about the progress of the build and, therefore, used as feedback for a process control system.
|
|---|