How do you prepare for an AM series use case from a technology perspective? In today's blog post, I would like to explore this topic in more detail.
First, let's differentiate between prototyping and a potential high-volume order. In prototyping, time is of essence and the goal is to get the parts delivered to the customer as quickly as possible. Costs and production preparation are adapted to the schedule, so the process details are not considered in the first step. Software tools (orientation optimizer, automatic support) are often used in process preparation to simplify the work preparation process. After all, the customer wants to have his component in his hands quickly to install it in a prototype, for the next management presentation, or to validate his simulation etc..
When it comes to preparing for series production, the mindset changes. Much more time is spent on process preparation, since the entire AM process chain is impacted at this stage. Parts are closely analyzed and aligned in different orientations on the build plate. Virtual part slices are made in x, y, z directions to understand every single detail of the part and include it in the part orientation process in advance. This is followed by optimization of the support concept, if supports are required. The goal is to design the support structures in such a way that they can be removed later by machine or hand as easy as possible. Experience has shown that this requires a number of iterations, as the process preparation and post-processing department has to align step by step.
Once the part is supported, we get to my favorite area, the process parameters. In laser powder bed fusion, there is a wide range of optimization that can be done to satisfy the customer. These include dimensional accuracy, mechanical properties, and density of the part.
Understanding the process parameters can only be an advantage, as a machine parameter set supplied by an OEM is usually only the first basis to start from. Additional fine-tuning is done based on the customer's geometry requirements, which is crucial in terms of cost and part quality.
After process preparation, we take a deeper look on the shop floor. The setup of the machines and the selected concept are essential for the uptime of the system in a future series production scenario. High OEE (Overall Equipment Effectiveness) must be achieved to optimize the utilization of each machine. For this purpose, the physical and digital interfaces at the machine as well as the build boxes to be transported in the future production line have to be considered. A zero-point clamping system for the subtrate plates, standardized digital interfaces, and an adaptive modular design of the machine are key factors. Their importance for manufacturing can be read in one of my articles on the IDAM research project (link).
However, a scalable machine concept plays a decisive role for me. Reproducible part quality, OEE above 80%, accessible physical and digital interfaces, digital integration into the system architecture, possibilities for process monitoring, user-friendly interface to the machine, short and reliable machine Idle times for hardware preparation are my main factors for such a concept.
A good understanding of the machine is needed before the series use case can be processed. Anyway, I am sure that most of us are checking frequently features of the new machine generations, so that should be no problem.
Finally, data handling should be considered. Initially, I would have rated it much lower, but with the digital tools available today in the field of AI, it needs a good database to further optimize manufacturing processes in the future. In my opinion, the trend in the AM market is increasingly moving in this direction, but some users still lack a properly filled data lake as a starting point.
Data is important for serial production of AM components, because in case of a customer claim, the root cause of defects can be quickly found and eliminated. Examples include in-situ process monitoring data, order status data, machine equipment vitality for preventive intervention, etc. If we ensure good handling and processing of this data afterwards, we will quickly learn to modify our process in terms of part quality and can guarantte a high machine vitality.
I am interested in your opinion. What are your three most important factors to prepare for a Metal AM series production?
Look forward to your answers.
Comments