Amsterdam-based MX3D, a specialist in robotic metal 3D printing (many will know of its famous canal bridge project), teamed up with software company Altair to 3D print an optimized industrial robot arm for a robot provided by ABB Robotics. The project, which leveraged Altair’s design expertise with MX3D’s large-scale metal 3D printing capability, demonstrates how generative design can optimize and customize robotic arms for increased productivity. The project also showcases how MX3D’s metal AM technology can efficiently produce large-scale metal parts with complex geometries.
From design to production
The 3D printed robotic arm was based on an existing part from ABB and was redesigned by Altair engineers using Generative Design Customization. The design process aimed to create a more efficient shape for the robot arm to not only increase productivity but also to improve handling (through lower kinetic energy), precision and build rates. These three optimization goals were carried out using the Digital Twin Design process, which consisted of performance requirement identification through kinematic motion simulation of assemblies.
Generative design was also employed in the redesign process in order to integrate optimized topologies with improved geometric efficiencies. Various design options were generated and the best was chosen after a validation process that considered performance and manufacturing feasibility.
Eventually, the optimized robot arm was 3D printed using an advanced version of MX3D’s Wire Arc Additive Manufacturing (WAAM) process using stainless steel material. The advanced WAAM process is equipped with superior geometry processing which enables complex, organic-inspired geometries to be printed upright. The WAAM process is also controlled by intelligent algorithms which found the most efficient build strategy for each feature of the robot arm.
>50% material weight reduction
In the end, the 3D printed robot arm was superior to the original in many ways. While the original arm weighed 150 kg, the optimized 3D printed arm weighs just 73 kg. The manufacturing process also proved to be more efficient: 3D printing the arm only took four days with 24/7 production. To finish the arm, the partners used a 3-axis milling machine to fine tune the tolerances at connecting points.
Ultimately, MX3D and Altair aimed to showcase how their respective technologies could enable the production of custom metal replacement parts, and especially large-scale parts. Whereas it would traditionally require extensive tooling and long lead times to order a large-scale metal replacement part, MX3D’s metal 3D printing platform has the potential to be used in-house for dramatically faster turnaround times and—in the case of the robotic arm—improved part performance.
“We’re getting much interest from corporates to print large-scale heavy equipment parts,” said Thomas Van Glabeke, Project Lead and Technical Business Developer at MX3D. “This is either for R&D purposes, low volume production or in case a critical part breaks down unanticipated and need to be replaced fast. MX3D is able to print these parts reliably and fast with WAAM technology. Moreover, the business case improves even further if an optimized design of the part is printed.”
“We are thrilled to collaborate with MX3D and are proud to contribute to its vision of developing tailored robotics,” added James Dagg, CTO of Modeling and Visualization at Altair Engineering. “The challenge in this showcase was particularly exciting as our Altair team supported the robot customization redesign from the earliest stages, helping to identify requirements for optimal operational conditions with a digital twin representing the entire system. We are looking forward to seeing the project progress as our efforts help to increase MX3D’s productivity.”
Last week at Formnext, MX3D announced the release of its basic MetalXL software, which makes it easier for users to 3D print large-scale metal parts in house. Metal XL Beta\01 is now in the process of being tested by select users and an advanced beta version will become available early next year.