Chris Wolfe

Feeling nerdy this Friday so we’ll tell materials stories. One lesson I’ve learned over the years: most engineering problems aren’t caused by bad data. Rather, they’re caused by incomplete data or viewpoints. For a long time, 2D metallography has been the standard for evaluating materials like 6xxx series aluminum alloys. It’s familiar, fast, and useful. But as performance expectations rise in automotive and aerospace manufacturing, the limits of 2D analysis are becoming harder to ignore. Intermetallics such as AlFeSi are a great example. Their morphology directly affects extrudability, surface quality, and mechanical performance. Plate-like β-AlFeSi phases can trigger undesirable cracking and defects, while more equiaxed α-AlFeSi phases are far less harmful. We were approached a while ago by a manufacturer with an issue that could potentially be traced to one of two suppliers. The challenge was finding evidence. Conventional metallography analyses, looking at single 2D slice was ineffective. This is where automated 3D serial sectioning changed the game. By reconstructing more than 100 precisely controlled slices into a true 3D volume, technologies like Robo-Met.3D® move us from estimation to evidence. Instead of inferring feature size and distribution from one image plane, we measured volume fraction, morphology, and spatial distribution directly, as the video here shows. The difference between the suppliers in terms of how much undesirable β-AlFeSi phase was in their products - was very stark. We had convincing evidence for our manufacturing client! What stands out to me from this study isn’t just the imaging technology. It’s the mindset shift. 3D analysis isn’t just about nicer visuals. It’s about reducing uncertainty, improving predictive models, and making better process decisions the first time. I’ve seen again and again that when engineers can see the whole picture (no pun intended), they design better processes and avoid costly downstream surprises. As materials systems grow more complex, relying solely on 2D data is increasingly a risk. Those who adopt volumetric, data-rich characterization won’t just understand their materials better. They will move faster and make smarter decisions. PS – Another lesson I learned with this – make sure you keep appropriate confidentiality in mind when presenting technical results. Having stripped this data of identifiers, I was presenting the results of this study at a materials conference about five years after the study. Among the highlights was an extensive discussion with a technologist from one of the Aluminum suppliers in this study. He agreed that this would be a very useful insight for their material as they were active in the field. I was able to keep my composure and agree wholeheartedly! #MaterialsEngineering #ManufacturingInnovation #3DMetallography #AdvancedManufacturing #EngineeringLeadership https://lnkd.in/gpD5UM4j

18

Tooling is still a great and underutilized AM application. "Subaru of America, Inc., the U.S. subsidiary of Japanese automaker Subaru Corporation, has reduced tooling development time by more than 50% and cut fabrication costs by approximately 70% by adopting large-format 3D printing for accessory installation tooling." https://lnkd.in/gAYKSr4F

57

6 Comments