The blueprints for a 3D-printed gun first appeared on the internet in 2013. It was fairly low tech, with the end product definitely not something that could be used in a fire fight. As the years progressed, however, technology improved. DIY’ers went from printing handguns that could only fire one round to components for AR-15s.
Still, the tipping point didn’t come until March 2020 when a design was released for a 100 percent printable semi-automatic rifle that is durable enough to shoot thousands of rounds of 9 mm bullets. For $350—which includes the printer and a handful of hardware supplies—an individual with nefarious intentions can print a non-traceable semi-automatic weapon.
According to the U.S. courts thus far, it is not illegal to share the design for or print a semi-automatic weapon; so, researcher Zhanpeng Jin approached the problem from a different angle.
Jin, an associate professor of computer science at engineering at the University of Buffalo, has found a way to track the origin of 3D-printed items that is as unique as the human fingerprint.
Three-dimensional printing, also called additive manufacturing, works by adding successive layers of printing material according to a digital design. Every 3D printer has an "extruder," which pushes the building material along. The extruder's hot end then melts the material, and places it on the print bed to build the model.
In a study published in IEEE Transactions on Information Forensics and Security, Jin and his team discovered that each extruder's hot end has its own heating properties that are so unique, they can be used for identification purposes.
A printer’s “ThermoTag,” as Jin and his team call it, is as unique as a human fingerprint. And like AFIS for fingerprints, an associated database for 3D printers would be necessary for comparison purposes. Theoretically, once that database exists, experts can use it to try to match a 3D printed object to various extruders. Once the model printer is identified, classic detective work could help find or at least narrow down the purchaser—and printer of the assault rifle used to cause harm.
For proof-of-concept, Jim and his team examined and compared the ThermoTag features of 45 different extruders of the same model. The researchers correctly identified the source printer with an accuracy rate of 92%.
"This ThermoTag will behave like the fingerprint of the 3D printer. When you print out a new product, you can use watermarking," Jin said, noting that watermarking can be used to invisibly embed such information as the printer's manufacturer, label and serial number in the product. "That would make this watermark of this particular product unique."
Of course, it is possible to replace the extruder on a 3D printer between jobs, opening up a possible route to avoid detection, or at least muddy a printer/person’s association to a specific gun.
“That’s why it’s so important to create a database of these parts for comparison,” Jin concluded.
Photo: University at Buffalo tech can identify machines by their unique “hot end,” which could aid intellectual property and security. Credit: University at Buffalo