Finally, turning plastic pellets into 3D printer filament

Here’s the situation: a kilogram of 3D printer filament costs about $50. A kilogram of plastic pellets costs less than a tenth of that. Does anyone have a solution to this problem?

For years now, the general consensus was making your own 3D printer filament at home was nigh impossible, dealing with temperatures, pressures, and tolerances that home-built machines simply can’t handle. [Bradley] sent in a filament extruder he made because he was disturbed at this current mindset that desktop filament factories have huge technical issues that have yet to be overcome.

[Bradley]‘s extruder is based on the Lyman Filament Extruder, a machine that has successfully demonstrated taking plastic pellets, forming them into a filament, and having this filament used in the production of 3D printed parts. [Bradley]‘s improvements include a variable-speed motor, a larger hot end, and an automatic timing system to produce set quantities of printer filament.

Of course, since Inventables threw $40,000 at the problem of creating filament at home there were bound to be more than a few successful designs making their way out into the public. When we last covered the developments of home filament manufacturing, the Filabot seemed to be in the lead. Now with [Bradley] (and  [Lyman])’s machines turning out usable filament, it’s only a matter of time before the 40 grand prize is snatched.


  1. Ren says:

    So, is it just price gouging by the mfgr’s at this point?

    • pelrun says:

      Not really – it’s more to do with economies of scale. There are large startup costs with doing a run on an industrial extruder, so you really have to make a lot of filament in a single go before the cost-per-length starts coming down.

      Since the 3d printer market is still pretty small, few people were going to take the risk of a huge, expensive run just to be left with a large quantity of filament that they can’t sell.

  2. Hack Man says:

    Not loving that unguarded roller chain and sprocket. That’s a problem.

  3. Rich says:

    Stupid question — weedwhacker string? Would that stuff work?

  4. John U says:

    Always good to see someone doing what everyone else thought was impossible.

  5. raster says:

    If weedwhacker string were PLA or ABS it would be a ‘maybe’ but even then it probably doesn’t have the tolerances needed. I think most weedwhacker string is nylon. This questions gets asked a lot. ;)

    • rasz says:

      why would you need precise tolerances? its melted in the extruder anyway

      • c0derage says:

        Volume in = volume out. If the feed stock’s dimensions aren’t precise then the volume of extrudate leaving the hot end varies per mm of feed stock fed into the system.

      • bennitkin says:

        The printer feeds filament in linearly, assuming that the filament has a constant volume per unit length. If the actual diameter varies, then the print will be inconsistent.
        (It might have blobs hanging off where too much plastic was extruded, or voids where the filament was thinner than expected.)
        Of course, if a printer could measure input volumetrically, uneven filament would be fine.

  6. rue_mohr says:

    weedeater string dosn’t work(not the good stuff anyhow) . The filament used for 3d printers was origionally designed for plastic welders, which iirc, also take a cheaper, triangular filament.

  7. Ragman says:

    It’s probably a stupid question, but why can’t 3d printers just use pellets instead of filament? I’m assuming the plastic they use never gets very flowable and the pellets would result in air gaps…or something.

    • Justin says:

      The setup that turns the pellets into a steady stream for the 3D printer’s “head” is rather bulky, as you can see. So you don’t really want to mount it on a head that moves around the place (which is the leading mechanism). So now you have a pellet melter that feeds a stream of plastic to a mobile head a bit further. The larger that distance (determined by how far the head goes) the more difficult it gets to maintain a flow of heated plastic, it’ll cool. So you have the choice:
      1. keep the plastic molten over a long-ish distance without a re-heater at the head. This is very cumbersome.
      2. let the plastic cool and reheat it at the head.

      To sum up: Having the plastic fed from pellets limits the distance the head can move and makes for a more complex setup.

      • Rupin says:

        You are correct, about the head moving, but can the head be stationary and the platform moves? I think it then solves the concern you raised.

      • Justin says:

        @Rupin’s reply.

        Sure it can have a mobile platform. But the problem is then translated in other issues such as how big will the total footprint be. And you’ll need to scale up your steppers.

        As another person replied, it has been done. But in the end you end up juggling trade-offs and at the moment the pre-made wire sits in a sweet spot.

      • Ragman says:

        I should have thought of that. I guess I was underestimating how fast the plastic cools and how far from the head you could move the molten plastic.

    • attrezzo says:

      They can. And they did.

      The filabot people did it. I’m assuming the intermediate step is more economic in the end. You don’t want your print head heaving around a hopper full of plastic pellets and all the machinery to melt/compact them in a smaller printer.

      On the filabot website, and more interesting to me, is the concept of using recyclable plastics instead of pellets. Just chop up Jr’s old toys and make new ones!

  8. ginsublade says:

    I used to work in a plastic pipe factory and they had a vacuum pump degassing the plastic where the plastic pellets first started to melt in the extruder….this was for PVC so I don’t know if it applies to this or not…thought it might be useful infomation …

  9. nhitze says:

    I just leave this videos here and you folks decide what to make of them, ok? :)

  10. Very cool, I enjoy the competition, but what do you mean “finally”? A month ago when you guys covered my extruder (over at or we were making HDPE filament and weve been refining our tolerences (and adding more torque!) since then! I wouldnt call it “finally” I would call it “as well”. That said this is a sweet design, damn I love the 3D printing world

  11. Isotope says:

    That price point is still tough to hit. I’ve said it before and I’ll say it again – it takes a lot of energy to extrude plastic and the more energy a machine consumes, the more expensive it’s going to be. The parts have to endure higher forces and temperatures and have some degree of precision.

    The fact that there’s a $40k bounty tells me these guys already did the market research and if they can produce these machines in quantity for $250 each, they stand to make considerably more than $40,000. My advice to anyone who hits this price point – keep it to yourself and go into manufacturing. You’ll earn more than $40k.

  12. Malcolm says:

    This is an important step in the right direction.

    I can see a time when you have a bunch of hoppers of different colored pellets, all ready to be fed just-in-time into the production.

  13. Cazna52 says:

    What I am really wanting is one that you can just drop pet bottles into that will shred them then turn them into filament.

    I have done a little molding using pet before and the temperature has to be pretty specific and cooling has to be slowed down a bit but if you could build a machine to make filament out of it you pretty much have free filament.

  14. NewCommentor1283 says:

    noone believed me when i said it may one day be possible,… to make 3D printer filament at home

    looks like mr crazy is sane again :)

    next step? recycle already-printed parts back into filament, possible? crazy? my last idea of the “future” was crazy(making filament), so maybe this one is possible too!(recycling) :)

  15. bothersaidpooh says:

    What about using a dual core wire
    i.e. polycaprolactone/ABS in order
    to mix it at the printhead?
    Best of both worlds, essentially
    a plastic “epoxy” that once
    hardened becomes stronger than
    both on their own.

  16. Malek says:

    So, uh… I want one? I want it? I need it? Where can I get it? GIVE IT TO ME!!!

  17. Makesthings says:

    As far as nylon weed trimmer filament being hygroscopic the treatment is the same as if you were using nylon pellet feed stock, you dry the feedstock before molding it. Conventional dryers used for dehydrating food would work for weed trimmer filament, in the injection molding industry hot air is pumped through hoses attached to the feed hopper so you dry the pellets on the fly. You need quite a bit of moisture to mess up an injection molded part from water vapor, so a simple trial and error approach with recording the drying time for each particular test batch of filament would render the correct time for the material. start at 100 to 120 degree F or check for a generic nylon resin and use the reccomended dry times and temperature ranges to start with.

  18. MPerez says:

    A meat grinder / sausage stuffer is all you really need to make your own filament. The main challenge is to get the heat away from the machine, but this can be easily done if you use a meat grinder attachment instead of an actual machine. The attachment will need to be mounted on a fixed base with a pulley or chain drive and a motor as picture above. A funnel shaped sausage stuffer attachment can be used to step down the flow from the meat grinders large opening. The end of the funnel will need to be fitted with the tube shaped heating element from an old coffee maker (be sure to descale it first), and the orifice will be somehow mounted to the heating element (by means of brass compression coupler with a cap on on other end. The cap gets drilled to the diameter of the filament (1.75 or 3.0mm). Heat can be applied to the meat grinder using stainless steal wire as an element and an electric control from an old electric stove to regulate the flow of power. A low voltage transformer can be used to maintain a low voltage and eliminate any risk of electrocution. Insulation of all the hot parts will require angel hair porcelain matting. And of course the filament gets dumped into a basin full of water made from an old 55 gallon drum. A motorized spool collects the filiament at the same rate it is fed by the auger.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Get every new post delivered to your Inbox.

Join 96,369 other followers