From Failed Builds to Innovative and Economical Designs

One of the challenging aspects of additive manufacturing (AM) technology adoption is how to conceptualize and create part designs that are optimized for function and manufacturability. As additive manufacturing processes can create parts with increased complexity, functionality and integration, our engineering design thought process and software tools have not fully evolved to take advantage of these potentials.

These challenges became apparent in a collaborative project where a team tried to bring to life an innovative optic circuit. The team included quantum physics researchers from the Quantum Physics Laboratory (QPL), Institute for Quantum Computing (IQC), University of Waterloo; and AM researchers from the Multi-Scale Additive Manufacturing Laboratory (MSAM), Mechanical and Mechatronics Engineering, University of Waterloo. The problem was complex, with optical mounts requiring a custom 3D-mechanical arrangement, utilization of a low coefficient of thermal expansion material, mass reduction, adjustability and tight alignment tolerances. We decided that metal additive manufacturing was the only viable solution and embarked on this path, not because it was easy, but because it challenged the status quo of what can be done in the field of laser interferometry and additive manufacturing alike.

To add to the challenge, the entire program was orchestrated from concept to print in less than six months, without our team members ever being in the same room. Orchestrating designs, coordinating manufacturing plans, troubleshooting failed builds and iterating through to success is a true story of why digital manufacturing is so powerful in terms of pooling talent virtually to create parts physically.

Presented here are reflections on our effort, with an optimistic anticipation of our final product being printed successfully. I hope the reflections on design for additive manufacturing will inspire others to learn from our failed but exciting builds. Three aspects come to mind that hindered innovative design thinking and our team:

1. Starting with a design problem space that is rooted in a design concept applicable to conventional manufacturing. The challenge with this approach is that it can lead to a constrained solution, without exploring a diverse family of design concepts. When following this path, our team found that instead of leveraging the additive manufacturing potentials, we were tempted to spend a large amount of time making the design manufacturable via additive manufacturing instead of making it better altogether. Converging to design solutions from experience happens organically, based on what designers have been predominantly exposed to up to that point. The solution to this problem is to bravely gain experience and be willing to ask for mentorship and assistance to critique the design path. In our situation, the design proved to be challenging in terms of manufacturability. This is similar to the issue of modifying for additive manufacturing vs. design for additive manufacturing highlighted previously.1
2. Solving all problems, except for the obvious problem. It is extremely challenging to take a step back and make the decision to go back to the basic question: what is the actual function of the design? Once the design team has gained experience with what works and what does not work for a specific design family, often the sane solution is to not give up, but to be brave and try something new. A complete overhaul of the entire design could lead to new opportunities in terms of more optimized productivity, manufacturability and overall better leveraging of the additive manufacturing potentials. In our situation, the overhaul resulted in shorter build times, a higher success rate and an overall sound design. ... Performance remains to be tested.
3. Integrating material, process and design into a manufacturability success score is challenging. One of the more challenging aspects was validating our designs via large-scale simulations. Finding the correct process parameters that are stable for complex, bulky designs, all the way to designs with refined features, as well as figuring out scanning strategies that produce a sound build file even if your design is “perfect,” is still challenging. Printing small pieces of the design to test out hypotheses was great, but when we put it all together, the build could still fail. We found the best approach was to use our intuition and go for it. This resulted in a few costly, failed builds and hours of forensic autopsies.

   

In deploying additive manufacturing, the knowledge of the complex relationships among material, process and performance are crucial in producing manufacturable parts with good mechanical properties and shape fidelity.

We look forward to exploring what the next generation of machines and software tools have to offer as well as to see it all under one roof at RAPID + TCT later this year.

Acknowledgments

This project was supported by the Department for National Defence Canada DND 4596-E — Innovation for Defence Excellence and Security program.

1-“Designing for Additive Manufacturing = DfAM + MfAM.” 3Dprint.com, 2020