Printing metal as easily as it is to printed with thermoplastics has been a dream for a very long time, with options for hobbyists being very scarce. This is something which [Rotoforge] seeks to change, using little more than an old Ender 3 FDM printer and some ingenuity. Best of all is that the approach on which they have been working for the past year does not require high temperature, molten metals and no fussing about with powdered metal.

Additive manufacturing using friction welding. (Credit: Ruishan Xie, et al., j.mtcomm, 2021)

Rather than an extruder that melts a thermoplastic filament, their setup uses metal wire that is fed into a friction welding tool head, the details of which are covered in the video as well as on the GitHub project page. Unlike their previous setup which we reported on last year, this new setup is both safer and much riskier. While there’s no more molten metal, instead a very loud and very fast spinning disk is used to provide the friction required for friction welding, specifically friction and rolling-based additive manufacturing (FRAM) as in the cited 2021 paper by [Ruishan Xie] et al. in Materials Today Communications. By the same lead author there’s also a 2025 paper that explores more complex implementations of FRAM.

With this method the feed wire (e.g. aluminium) will experience plastic shear, causing it to become somewhat fluid and capable of adhering to other surfaces. This same method can be used for other materials, including plastics and glass. As can be seen in the video, FRAM certainly seems viable, though it is a veritable rabbit hole of complications since you’re dealing with a high-velocity engineering challenge.

Here’s where [Rotoforge] found that slitting saws are a good, off-the-shelf option as they have basically the same high-speed-but-safely requirements. This left the motor part, which has to keep the friction wheel going. After a lot of trial and mostly error, it was found that the motor in a Dremel tool provided the solution in the form of a universal AC motor. Unlike brush-less DC motors, these AC motors are far more simpler, cheaper and can keep up a constant speed much better, which is probably why they’re still used in power tools everywhere. Ergo some cheap Vevor flex-shaft grinders were bought and adapted for some FRAM purposes.

The Rotoforge FRAM printing in action. (Credit: Rotoforge, YouTube)

Initial experiments with Al1100 aluminium alloy showed very good layer adhesion, to the point that they were very similar to a solid bar of aluminium. Due to the layered nature of the prints, they perform better than solid Al1100 parts in some tests. Thus the next challenge was to try more advanced printing techniques than a single straight line, which posed the bigger challenge and is where the basic Ender 3-based prototype met its match.

Next a fourth axis will be added to hopefully resolve some of the issues encountered with the current prototype, along with a host of other improvements to make printing more reliable and versatile. Although it’s clearly still early days for FRAM, it is rather exciting that even in a hobby setting without massive monetary investment it’s already possible to do this much.

Of course, it should definitely be said that eye- and hearing protection are absolute requirements if you intend to do some FRAM printing yourself. The video gives some idea of how loud the process is, and high-speed discs and wheels together with metalworking always introduce the exciting possibility of high-velocity shrapnel.