It occurred to me that one way to potentially eliminate all filament swap waste from purge towers or Bambu-style filament “poops” is to instead do something similar to a “purge object” or a “wipe object” but where the object is… filament.

The idea was somewhat inspired by Stefan’s video from a few months ago, which first introduced me to the idea of printing filament: https://m.youtube.com/watch?v=CQ-N1fr4N0w

If the purge object is new filament, then you don’t need to figure out what to use it for immediately; you can store the mixed-color filament for later use until you have a genuine need for a color-agnostic or structural object. No more piles of fidget spinners you didn’t really want.

A few other thoughts:

  • The filament probably needs a minimum bend radius so it can feed into an extruder without breaking. One option would be to print a circle or a rounded square / squircle type of shape toward the perimeter of the build volume (potentially multiple concentric ones, or spiralized shapes). Another option would be to print disconnected straight segments.
  • The filament doesn’t need to be printed flat. It could potentially spiral upwards (likely with supports). That could help avoid print head collisions for taller models (in which case you might not want the head to have to get back down to <1.75mm from the print bed). In the most general case, the filament could be printed at arbitrary angles, even vertically, and could start at some height above the build plate (potentially supported by the object itself, for example in an object where color changes don’t start at the very bottom). Maybe the angle could even be optimized to provide the best match between purge volume per layer and volume of new filament printed on each layer.
  • A good solution probably requires a good way to connect multiple segments of printed filament together
  • Mixing materials rather than just colors seems like a bad idea to me (e.g. soluble supports) but I’m going to mention it as a possibility anyway; you could conceive of something inspired by Stefan’s composite material and intelligently organize the different materials within your new filament.
  • You could potentially control the mixes / transition colors that go into your purge filament, or even choose multiple different new filaments for different color transitions. For example, you might have separate new filaments for red-green, green-blue, and blue-red transitions. Maybe you want it create particularly pretty new filament, or avoid particularly ugly combinations.

I don’t actually have a multi-color / multi-filament printer, and I don’t have time to experiment with this (even though the first prototype could be as trivial using a filament shape as the wipe object in PrusaSlicer). I’m mostly sharing this to establish prior art in case someone nefarious seeks to patent something similar in the future (which is also why I added some half-baked thoughts that might make other things become obvious to someone skilled in the art), though it’d be great if the idea is actually good and someone could implement it well. Or for all I know this does already exist.

Thanks for taking the time to read this! Feel free to repost/share/steal the idea if you like it.

  • NuXCOM_90PercentEnglish
    1·
    5 hours ago

    Unless you are changing filament VERY frequently, the energy costs will almost guaranteed erase any environmental benefits from filament waste.

    Because you basically have the exact same problem that most of the filament reuse methods have: If you want a good connection, you want to have both the filament coming out of the nozzle AND the end of the filament you are printing onto to be hot. That is a LOT of engineering effort as you would likely need to keep the current tail of the output filament hot for the majority of the print so as to not add significant stalls when you change filament. This is why most of the tools to fuse to strands have that sleeve that you heat up

    Because the moment you start adding supports for your output filament? Holy crap.

    I dunno. I still think the answer is more cost effective recycling facilities. I’ve enjoyed Stefan’s various attempts to reuse filament but outside of the splicing methods for near empty spools, they are all a giant mess requiring multiple tools for an often subpar result. Just standardize a cheap and effective way to throw our poop, failed builds, and near empty spools into a box and send it to a filament company. Then give us a discount for doing so. And let said company use their industrial machines to reuse that plastic.

    One other complexity: Again, unless you are changing materials constantly AND doing a super long print, the amount of filament you print during any given print is going to be minimal. So you need to maintain state on the build plate/apparatus in between potentially months of prints.