Automatic Error Correction Delivers Reliable 3D Printing of Truck Tire-Sized Parts
When you're printing things the size of a truck tire — or a truck — you need a little help to ensure the best possible quality.
Researchers at the US Department of Energy's (DOE) Oak Ridge National Laboratory (ORNL) are looking to make 3D printing more reliable for large-scale manufacturing — by developing a computer vision system that can detect and correct errors during the printing process.
"It is novel that our controller can sense what is happening and react in real time," claims lead researcher Kris Villez, who partnered with University of Tennessee graduate student Chris O'Brien on the project to improve how 3D printing works for large-scale parts. "It controls the process almost like a human would: by observing and nudging the setting until it reaches the desired outcome."
Fused filament fabrication (FFF) printing, in which a molten material is extruded out into precisely-controlled layers to build up a 3D object, has revolutionized prototyping and small-scale manufacturing. When you scale objects up, though, the reliability drops and errors in the printing process build up — which is the problem Villez and colleagues set out to solve, targeting items bigger than a truck tire to prove the process.
The system works by using an array of six thermal cameras positioned around the extruder, monitoring the material both as it's extruded and as it cools. If the material's temperature is too high or too low, either of which could ruin the print, the controller can automatically compensate by adjusting the hot-end temperature or the print speed in real-time and without human intervention.
"There is a vast opportunity space to make these machines more intelligent and more responsive," Villez claims of the potential impact of the system. "In the end, we'd love this to work like baking bread: You set the oven temperature, put in your dough, and return when the timer goes off to see if it’s done. You don’t have to monitor the oven temperature in real time throughout the baking."
At the time of writing, the ORNL team had not released any further details on the system — though a paper detailing an earlier version of the same error correction approach, written in partnership with researchers at the University of Tennessee-Knoxville, is available under closed-access terms in the proceedings of the Sampe Indianapolis 2025 conference.