I've got a cardboard box next to my workbench that I used to call the "failure pile." Warped first layers off the Flashforge Adventurer 5X, a half-finished AMS test print from the Bambu Lab A1 that picked up a layer shift at 2am, rafts and supports I trimmed off and tossed without a second thought. For years that box just... filled up, and eventually went in the trash. Somewhere between $40 and $60 a month in plastic, by my own rough math, just thrown away.
That box is basically empty now. Not because I stopped failing prints — I still do, regularly, that's 3D printing — but because I finally built a recycling loop with Felfil's gear, and it's changed how I think about every spool I buy. This isn't a theoretical "you could recycle your prints" post. This is what actually happens on my bench, what worked, what didn't, and whether the math makes sense for you.
Why I Bothered in the First Place
I run two very different machines side by side — the Adventurer 5X for fast, disposable prototypes and the Bambu A1 with the AMS for multicolor stuff — and between the two of them I generate a steady stream of scrap: failed calibration prints, support structures, purge blocks from AMS color changes, and the occasional full print that just didn't work out. I'd been buying filament in bulk to keep costs down, but I kept circling back to the same annoyance: I was paying for plastic twice — once to buy it, and again (in time and material) when a print failed and the scrap went straight to landfill.
Digging around r/3Dprinting and a few of the filament-recycling threads on the maker forums, I saw the same debate over and over: is DIY filament recycling actually worth it, or is it a novelty that burns more time than it saves? I decided to just build the loop myself and find out, rather than take anyone's word for it.
The Setup: Shred, Extrude, Spool
There are three steps to turning a pile of failed prints back into usable filament, and I'm running Felfil gear for all three because it's the only desktop-scale system I found that doesn't require an injection-molding background to operate.
Step 1 — Shredding
I started by running everything through the Felfil Shredder+. This is the step people skip when they try to DIY their own recycling with a blender or a hand-crank shredder, and it's the step that actually determines whether the rest of the process works. Feed size matters a lot — if your shreds are too big or too inconsistent, the extruder downstream will choke or feed unevenly. The Shredder+ chews through rafts, supports, and failed prints into a fairly uniform flake, and I run the blades hot right now because I'm pushing through a lot of volume. Worth knowing up front: the blades and combs do wear, especially if you're shredding anything with infill patterns that create tougher chunks, so I've already got the replacement blades & combs kit on the shelf for when mine dull out.
Step 2 — Extruding
From there, the shredded flake goes into the extruder. I'm running the EVO Extruder & Spooler Bundle, which pairs the EVO extruder with Felfil's automatic winder so I'm not standing there hand-winding filament for an hour (I tried that early on with a 3D-printed spool jig — don't bother, just get the Spooler). Temperature control on the EVO is the whole game here. PLA extrudes cleanly in the 170–190°C range for me; it's forgiving and predictable. The moment I started trying to blend in PETG scrap, I had to dial in temps more carefully and slow the feed rate, because PETG is far more sensitive to moisture and will string and bubble if you're not patient with it.
Step 3 — Spooling
The automatic spooler is genuinely the difference between this being a hobby and this being a chore. It winds at a consistent tension and calibrates as it goes, which matters more than people expect — uneven winding tension is one of the most common complaints I see in recycling threads online, and it's almost always a spooling problem, not an extrusion problem.
What Actually Works (and What Doesn't)
After a few months of running this loop, here's where I've landed, material by material:
- PLA: The easy win. Consistent results, minimal drying needed, and it's forgiving if your shred size varies a bit. This is where I'd tell anyone to start.
- PETG: Doable, but it humbled me. PETG is hygroscopic and recycled PETG is even more sensitive — if you don't dry your shred thoroughly before extruding, you'll get bubbling and inconsistent diameter. I now dry PETG scrap overnight before it ever touches the extruder.
- ABS: Workable but smelly, and I only do this with ventilation. Recycled ABS also loses a bit of its impact strength each time you reprocess it, which lines up with what I've read from people running print farms who recycle at scale.
- ASA, Nylon, anything fiber-filled: I don't bother. The temperatures and drying requirements aren't worth it on a desktop setup, and fiber-filled material chews up blades fast.
Does the Math Actually Work?
This is the question everyone actually wants answered, and the honest answer is: it depends on your volume. The Felfil gear isn't cheap, and if you're an occasional hobbyist printing a handful of parts a month, you will not "break even" on the hardware by recycling your own scrap alone — the value there is more about waste reduction and having a reliable source of support-material-grade filament for test prints than pure cost savings.
Where it gets genuinely worth it is if you're running anything close to print-farm volume, or if you're doing a lot of multicolor AMS work where purge waste adds up fast, or if you just hate the idea of plastic going to a landfill when it's sitting right there on your desk. I'm not replacing my primary filament purchases with recycled material for anything customer-facing — color consistency and mechanical properties aren't quite there for parts that need to look or perform a certain way. But for test prints, jigs, internal tooling, and prototyping, recycled filament has become my default, and it's taken a real bite out of my monthly filament spend.
My Actual Routine
If you want to copy what I'm doing: collect scrap by material type (do not mix PLA and PETG scrap — I learned this one the hard way), shred each batch separately with the Shredder+, dry anything that isn't PLA before extruding, run the EVO at a conservative feed rate rather than pushing for speed, and let the Spooler+ do its thing without babysitting it. Label your spools — recycled filament from mixed batches of failed prints often comes out with a marbled, slightly mottled color, which honestly looks great on anything mechanical or industrial-styled.
If you're curious about building this same loop, I'd start with the Shredder+ and the Extruder & Spooler Bundle together rather than buying a standalone extruder first — the automatic winding is what actually makes this sustainable on a regular basis instead of a one-time experiment. The full lineup, including replacement parts, is in our Filament Making & Recycling collection if you want to see everything side by side.
I still fail prints. I just don't throw them away anymore.
