Print Your Own Filament

Ask anyone with a 3D printer what they make the most. They’ll probably say “trash.” There are extra pieces, stuff that oozes out of the extruder, support material, parts that didn’t stick to the bed, or just parts that needed a little tweaking to get right. No matter what you do, you are going to wind up with a lot of scraps. It would be great if you could recycle all this, and [3D Printing Nerd] looks at the FelFil Evo Filament extruder that promises it can do just that. You can see the video below.

As you’d expect, the device is a motorized auger that extrudes filament through a hot end not dissimilar to your printer’s hot end. You have to run a bag of special material through it first to clean out the plastic path. After that, you can create filament from standard pellets or pieces of old plastic.

The device didn’t work right away. Turns out a wire was loose inside. After that, how did it work? Watch the video and find out.

We will say, it is very slow and you can’t run the machine for very long periods of time to prevent overheating. It also wasn’t clear to us how you’d keep the filament neat as it comes out. The video never shows using shredded up plastic parts, either, but you’d think that wouldn’t work any better than the pellets.

As a gadget, the idea of a filament extruder is pretty cool. For recycling, it is a great idea. However, with filament relatively cheap and readily available, any desktop filament extruder is going to have to work pretty well to win a lot of users.

We’ve seen homebrew extruders, of course. Some of these are pretty inexpensive and use common parts. So if you wanted to experiment yourself, there isn’t much to stop you.

25 thoughts on “Print Your Own Filament

    1. Tons of times, every time, people discover that extruding filament isn’t hard. But creating a reliable extrusion and spooling it properly is. So it never gets beyond a toy level.

    2. Indeed, both the mill and winder kit are about $400
      The speed is certainly an issue, but this option uses a 1.75mm tube as a draw-plate… Several users also describe the operational quirks, as professional industrial units use a 3mm nozzle, water cooling baths, and simply adjust spooling rates for smaller diameters.

      Full disclosure, I do not personally own a filastruder as I run ABS, and recycle my support material in acetone solvent for surface adhesive (thin layer smeared onto glass at 95’C sticks extremely well, and must be under 30’C to remove).

      1. @Joel
        Why use water cooling baths? And how well would that work for hygroscopic filament anyway? (They don’t put a desicant bag in ever spool you get for no reason. Moist filament is a nightmare!) If you look at Joels Youtube channel for videso from ProtoPasta you can pretty well see how they just run the filament along a few meters in a V shaped aluminium channel before it goes to the spooling station. There is no water involved to cool it, just some fans blowing air around.

        1. I had a chance to see with my own eyes a professional filament extrusion line. There were like 3 different water baths involved – each cooling the filament further down from the extrusion temperature.

          1. You are both correct, as I saw some lines also use water cooled rollers that progressively tension the filament as it cools (that was in a China Factory, as in the US it seems water baths are more popular… Sorry Magpie, I have no idea which is better or why.)

            I suspect those little desiccant bags don’t last indefinitely, and have a finite shelf life. At one point, I just used large reusable safe-desiccant bags in spool box setups… given the humidity can get ridiculous around here… we just microwave the bags and leave the safe-desiccant to steam-off in the open a few times.
            Note, the ABS I use doesn’t seem to get steam pock marks even if left unopened for several months, but PLA starts to sound like popcorn after 3 weeks. We now mostly print flat stuff for mechanical mounting brackets, and like the fact transition-cement will weld ABS to cheap PVC conduit.

            Maybe someone can try a mini forced-air pellet/spool dryer/dehydrator based on a 60w incandescent light-bulb and muffin fan….

        2. I can’t reply to your deep reply. I think the bags do last if they are sealed. There is only so much moisture in the sealed bag and once the silca gel grabs it, there is no more. Until you open the bag. Here’s a tip, though. The cheapest way I know to buy reasonable amounts of silica gel is to buy “crystal” cat litter. Same stuff. Huge bag at Walmart for $10. Just be careful going to Walmart and asking for Crystal. In some cases, she’s one of the cashiers. In other cases, you’ll get a big surprise and possibly jail time.

          When I worked for GE they would extrude plastic through a water bath prior to chopping it into pellets. I knew this one was going to have trouble when you see the blob of plastic weighing down the extrudate as it exited. That probably thinned out that part of the filament. That was Valox, I think. Or maybe Ultem. They made both in that same plant.

  1. This looks like a pretty terrible product. Not even spooling the filament coming out of it makes it pretty worthless, but then seeing how much effort he had to go into, trying to dial the settings in, and he was just using one type of PLA-filament — different manufacturers’ filaments may require a new round of dialing settings in, different colours may require it too (e.g. the white PLA from Prusa that I have is very stringy and oozy compared to the silvery one and requires different settings to print with, if I want similar results), and how does it react to variances in ambient temperatures?

    1. yeah… that dialing in of a temperature and speed values must have been really hard work.
      But seriously now, what wondered me the most was the complete lack of anything to wind the filament on. Oooohhh look how the filament is all tangled up, well that’s no surprise? Or is it? The filament has no way to go but to twist and curl while cooling. What’s the use of a pile of filament, can you feed it like that into your machine, nope, it must be on spools not on piles, otherwise (if not already) it tangles up and you print stops. Why does the “review” (because that’s what it is, a shameless review video) does not cover the lack of a filament winding method?!?!? And why does it look like that there is no proper cooling of the filament BEFORE it leaves the extruder. If there was then the diameter of the produced filament would be much less affected by the speed and the outside temperature. Now it takes while to tune in the device, get everything right and then you must shut down the device because the manual says it might overheat, now that not a very (this device is completely safe) ensuring thing, or is it?

      The product concept is very interesting, but the way it was made how it works and how it performs do not justify the cost. Though it was very interesting to see such a device aimed at the 3D printer/consumer market.

      1. So many people in 3d printing have this attitude and it’s really weird. Practically nothing about 3d printing is consumer-grade, fire-and-forget, layman-friendly tech. All this stuff requires calibration, knowledge, modification, and that’s the fun part.

        If you have a 3d printer and need something to spool this filament up, print a spool holder and a reduction gear to make the spool spin in proportion to the auger extruding filament from this device. Looking at the product, it’s clearly not meant to be a polished, complete consumer gizmo. It’s a pile of exposed heating elements and industrial machinery. I think the target market is people who can take that and make it work. You know, hackers and such.

        1. Not sure how this one would fare, but just raising the nozzle up off the ground a good distance and pointing the nozzle down lets you use gravity to create your tension. The tension on the filament is proportional to how far your raise it, it straightens out on it’s own, has time to cool on the way down, and then tends to form a coil all by itself. If you can’t tilt the extruder, you can use a roller right after the nozzle too, but that will leave a flat spot on the filament.

  2. This article is misleading. I expected a method to print filament with a title like “print your own filament” but instead it’s a link to a product review and said product doesn’t print filament. It prints a stringy, un-useable mess. Let us know when something that actually works is publicly available.

    1. We can do that math. 30 euros a kilo of filament (if you have expensive filament). Raw materials are about 4 euros per kilo. Which would put the break-even point at about 11.5 kg.

      (If this machine would reliably produce quality filament. Which is most likely not the case)

      1. i would be interested how you came to that conclusion because nice try but no the basic kit which you based your calculations on is hardly a fully functional machine. the full kit costs 600 and your raw material estimates are very optimistic.

        1. Not to mention running costs and man hours involved pushing up any estimated costs, also there will be failed batches which push those costs up further, finally how many times can you recycle filament?

          Also how much filament do you use in a year really? Is it even worth the time and investment to recycle a few kilos of filament a year? I mean if your the owner of a make a space or company, I can see benefit judging on how much filament you would go though, to recover say 30% (guess at the waste involved in printing lol) would be a good amount of cash saved if your using a few hundred kilos a year.

  3. While this product looks a little half-baked, it does show promise. My feeling is that the lack of a take-up spool is not the big issue this article makes it to be. Any self-respecting hacker can use an empty spool and print his own gearing to make one. I’d even suggest that the manufacturer add a bit of value by including a free plan for such a mechanism in with the device. The slow speed and the tinkering required to get consistent output are the real problems. Better power regulation and increasing the motor size and heater output might help there.

  4. Some youtuber reviewed this a few months ago.. It “Works” but barely. It was inconsistant in it’s thickness no matter what settings he used, and even after he contacted the company who made it. He had several problems and ultimately decided the time spent with it, it just wasn’t worth it. It is a great idea in concept, but wrong execution of design. Your mileage may vary.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.