Fail of the Week: How not to Build a Filament Extruder

It would be great if you could create your own filament. On the face of it, it seems easy to do, but as [Thomas Sanladerer] found out when he was a student, there are a lot of details that can bedevil your design. His extruder sort of works, but he wouldn’t suggest duplicating his effort. In fact, he hopes you can learn what not to do if you try to do it yourself.

In all fairness, [Thomas] was a low-budget student and was trying to economize. For example, he tried using a drill to drive the auger. Why not? It looks like a drill bit. But he found out that wasn’t satisfactory and moved to a pair of wiper motors with their built-in gear train.

The wiper motors allowed him to get some ABS filament, but the machine had more troubles. Other lessons learned were to keep the water cooling tank closed so water doesn’t splash out onto electronics, and that it is hard to look at filament with a CCD sensor.

The controller is a simple Arduino. There are three heat zones before the plastic reaches the die. As you might expect, there’s a PID controller to regulate the machine.

[Thomas] says the flow rate was too high, so slowing down the production might have helped. A smaller auger is also on his list of things he’d do differently in hindsight. The melting area needed a heatbreak much like a 3D printer’s hotend to keep hot plastic from creeping towards the cooler part of the tube and clogging.

With his current experience and a bigger budget, we have no doubt he could end up with a workable extruder. As it is, we always enjoy learning from other people’s suboptimal builds. It is a little humbling to show you failed projects on the Internet, but it is a valuable service, indeed.

We want an extruder that can recycle our junk parts. We’ve seen some really cheap builds, but we don’t really know how well they work.