PET Bottle Recycling: Waste to 3D Printing Filament 

Is Pultrusion a big waste of time of a feasible way to make good filament? Oh, and check out our CNC Kitchen products at cnckitchen.store/ or at our resellers www.cnckitchen.com/reseller

Whether it is a waist of time or not, the best part about doing something like this is the excitement it creates to go out and clean up trash. Plastic bottles are everywhere.

Little tip if you struggle to remove the POM wheels from the bearings : leave them a few minutes on a 90°C bed, they will easily fall off

Note for those of us with SAE drill bits. A 1/16th bit will produce a 1.57mm hole on a drill press but chucked into a handheld drill gets much closer to 1.7mm. The test hole I drilled in a scrap piece of steel fits a cutoff of 1.75mm filament snugly.

Nice video as always Stefan. Just few important notes for fellow makers: 1. Dry your bottles (or bottle strips) in dehydrator for 8hrs on 70C before pultrusion 2. Pump the pressure of the bottle before heat treating surface with schrader tire valve. 2. Check the thickness of the bottle and readjust bearing cutter by using Petamentor2" calculator or manually (thinner the bottle wider the strip) 3. Each bottle brand might need specific pultrusion and printing temperature (eg. some of my bottles even couldn't be purged below 280C)

But the Pfand tho 😭😭😭 jk it looks so cool

im glad you are bringing this to a bigger platform, a lot more people are going to try jt now

Thanks! Insane that you made one. We are keeping support and sponsor Joshua / ReCreator, but we finally decided not to make a kit for it, since it’s much easy to get parts from scraps, and save cost. Great video, thanks for spread recycling sprit! See you soon.

I would very much love to see some more testing of this material. But there may be too much variables to control to do it justice. I went down the rabbit hole and built myself a rig for pet recycling during COVID. I gathered all the info I could and used it to make my own experiments and I consider it to be a great overall material, similar to perg, but seems even better in some aspects. However, I ended up using entirely different set if parameters than yours, with still great results (though I have no hard data to prove it, other than my opinions and observations). To me, it is only a viable option if it can replace my need for commercial filament. This means it must be easy to make in enough quantity when I need it and produce comparable results. It also needs to be cheaper, which comes down to how much time it takes to make as in my country (Serbia), the bottles are everywhere. First, the quality - it's definitely comparable, perhaps only except color. You are a bit more limited there. You can try and color the strips and it somewhat works, but the colors aren't as saturated. Second, the time it takes to make - I used different parameters for pullstrusion and I can get a 2l coke bottle done in 25mins and it yields 15-20g of filament if prepared right. An order of magnitude faster than what you mentioned. I don't know if there are downsides, I didn't measure, but the quality is perceptibly the same as the longer methods. I used 235°C and a fast enough feedrate to land on 25mins for lets say 18g of filament. Even if this is slow for some, the process can be parallelized by having multiple machines running at the same time. But the device has to be cheaper than an old ender 3. Luckily, there are people who managed to bring the cost down to very cheap. Check out Petamentor for example, but there are more. Now we come to the biggest problem - quantity. 20g of filament is simply not enough for a lot of the prints and if it is, you are left with a few grams of unusable filament in the end either way. A solution is to join more filaments together. Joining then reliably is a problem. The biggest one for me, which in the past has made me give up on using recycled material. First - forget about joining strips, they simply can't survive the stress of pulling through the nozzle. Plus, like you mentioned, the process doesn't actually melt the strip, so it won't melt the strips together either. I tried joining strips in multiple ways and never had a successful pull through. I tried 100s of times - melting, mechanically interlocking, chemical solvent based bonding, and combining it all, it doesn't work in a diy environment. Otoh, joining finished filament is deceivingly simple. You take a piece of PTFE tube, melt two ends by flame or heater (I used a soldering iron) and while still molten, you slide on the tube. When it cools, the joint is done. However, the joint is very stiff and brittle and often breaks even if you just try to wind it on a spool. If it kinks before it gets in the extruder - forget it. Relieving the pressure on the joint helps somewhat, by making the bends before and after it to increase mobility. I tried faster and slower cooling, various things, but IMO, this is still the biggest issue and stops you from making a spool of say 200g of filament and just hitting print. I recently stumbled upon a guy that solves it by making a huge diameter spool, which is a genius idea (function3d on yt), he has many different good ideas too. This prevents the filament from bending too much and breaking the joint. Also, fresh off the machine, the filament tends to uncoil and spring back to a big diameter, making spooling it impossible. But not if you have a big enough spool. I will have to try that, as it looks like it's the missing puzzle piece for making the process of making filament viable. Next - printing. I found that printing at high temperatures enabled faster printing. I had little zero issues of crystalization. I print on 280°c and have great results. I think layer adhesion is great, but have no numbers. I think overall strength is good, but have no numbers. Other notes: it's prone to moisture absorbtion like petg. Drys easily. Appears to be somewhat self-regulating in terms of density consistency, I had good results on same settings with vastly different strip widths and thicknesses. Once the joint makes it into extruder, it doesn't cause problems. It frequently snaps while entering. Clogs weren't common. Cleaning wasn't too important, except for label adhesive. White spirits/turpentine works good, less toxic than acetone. I blow up bottles with a bicycle pump to 2 bars before treating with hot air gun, that way the bottle doesn't shrink. More than 2 bars risks explosion. I'd love to see annealing at least, but other tests too!

I love these recycling-themed videos you do. Praying a cheap and reliable pultrusion machine comes to market soon!

The hollow result makes me want to see if you can stuff other materials into the center before pulling making composites, would that be something you could try?

Does this filament release toxins or off gas while printing?

I've been gathering so many different versions of these. But no easy to follow video like this. This is exactly what I was looking for as I've been starting one of these PEt recycle projects! Thanks!

I had the same thoughts on using PET Bottles here in Germany so I looked out for a different material that is used everywhere and always thrown away. Joghurt and curd cups are abundant and made from polypropylene (PP). PP is a pain to prepare and cut with the bearing cutter but there is a razor cutter somewhere on thingyverse or cults3d that works nicely. My recreator version is not ready to test yet but I made some PP filament in the past and found that PP has some really interesting material properties like environmental and chemical resistance and being dishwasher compatible. Might be interesting to have a look at! Thank you for sparking new interest for turning waste into useable stuff!

I'm not sure I have time to do this. But, I love that it exists. Everything good starts like this and then, who knows?

I’ve built a machine since your previous video about this topic, its a quite nice hobby, but the best thing in my opinion would be desgning a lowcost filament extruder in a sinilar fairly easy to build manner. Then we could use the whole bottle, use the caps of the bottles or even something like a milk jug or containers that pudding comes in. In my country (Hungary) we have a bearly functioning collection strategy with no deposit so I’d love to build an extruder, but the ones alaready in the wild are either really expensive or built from literal scrap

I would be interested to see a full test round on crystallized PET. Amorphous PET will momentarily become extremely sticky while melting, and that basically stops up any industrial machine melting plastic. For that reason PET pellets are crystallized in special machines that keep them moving while heating them, like a washing machine but with infrared light instead of water. From experience crystallized PET is really similar to PBT but possibly more brittle, usually in manufacturing we try to avoid crystalization afrter remelting by cooling the parts quickly. Slower cooling after remelting results in crystalization - opacity and brittleness

I'm really happy you were able to get around to making one of these! I have a MK2 that I recently updated to include the geared system of the MK6 and it's a blast to make the filament. We don't have deposits for bottles in my state (sadly) but this does mean I have great incentive to make filament myself! the 2L bottles give me 20+ grams of PET and the prints are great. Like you, I've used it for several practical prints - mainly repairs. And DEFINITELY test annealed PET against PET! Something I've noticed is that the PET can even "anneal" itself on certain prints, particularly those with layers with smaller cross-sections. Those areas will come out opaque while the first few layers are clear.

Stefan, I thoroughly enjoyed this video, and was inspired to see you involve your daughters in this effort. While it will always be easier to have perfect spools of PETG shipped in for printing, being able to make filament this way seems like the fullest evolution of this technology. I will be making one of these. Thank you from VA, USA.

If I watched this video two days ago, I would have asked for a third option: Living in a world with PET bottles, but not using one for the last two weeks. But yesterday, I was actually using one.

Just the process of assembling the entire project already looks like a lot of fun. I doubt if the effort is worth it to 3D print with PET for me, but the machine itself is really awesome! Straight forward + good explanation video! 🙂

been making pet filament for 1 year now, and for the price of less than 2 euros per kg (electrical bill) , it has been great, better thermal resistance, better uv resistance, easy to print. made my own version of a pet pull, even made a tutorial how to build one. the downsides: short pieces of filament, not easy to splice together, not the best looking prints(transparent filament, standard blueish colors, more stringing )

Anything that helps recycle OR gives inspiration to create something better is always a good idea. Danke!

I have built this with the exception of grinding the bearings for the bottle cutter. Works well

I wonder if you could also re use the heated bed in order to make a dehydrator that connects to the rest of the filament maker. That would be super cool! Maybe you could look into it?

Smoothing out the bottle was my fav part, such a genius idea. Worked amazing.

This is a brilliant device. I admire how they were able to reuse so many parts from the Ender 3. Amazing design. It must have been a long design process, but I bet it was fun!

Love this sort of build! good job describing it and including the "bumps" along the way. Have you though about pelletizing? I think it would be harder with more specialized equipment (ie, not recycling an Ender 3) but could have a much higher efficiency since you could pelletize the tops and bottoms. And probably wouldn't risk your fingers with the knife! Would probably need 3 or 4 different machines. It probably could be used to make solid filament, though.

I'm surprised that the pull process is so slow. Presumably, this is due to a thermal limit, so could the ribbon be pre-heated? (Using the spare bed heater channel)

I like your optimism, not a lot of people have that lately.

Absolutely flawless video. Well demonstrated and critically important information presented. Thank you for making the 3D printing community more scientific and accessible.

What do you think, is it possible to pull textiles through the machine? There are 100% PES clothes which could work. Textiles are often just recycled "thermically"--> burned, even in countries where bottle recycling is quite good

I think it’s a great idea. I’d really like to make my own pulltrusuon setup, but I don’t want to take apart a printer, or use so much filament/time printing printing the base. A budget version, which doesn’t need massive investment of expensive materials would be great.

Thank you for making this video, I've been planning on building the mk5 kit for months but never finding the time but your video has given me the motivation to finally embark on that journey lol

The fact this just works. And so simply.... So cool.

Definitely a cool project, but the amount of manual labor for 30g of filaments seems pretty impractical. Albeit expensive, shredding and using a proper extruder is probably still the best option, as that not only cuts down on manual labor, but also increases yields, since the whole bottle can be shredded, and allows for much larger batches

Ive been really interested in this process. Have you found any reliable way to weld the filament pulls togather for a longer strand?

Great video! Bonus: when you do the recycling yourself you know the material is TRULY recycled, not just stuffed in a landfill somewhere.

Heck yeah, thanks Stephan! I've been thinking about trying to do something like this.

So much informative. As always. Thank you very much. Do you think this way one can convert 3mm filament to 1.75 mm? So not melting but just softening.

I wonder, since the filement is hollow, could you feed a fiber in for fiber reinforcement? You could feed it though on the middle of the bottle strip before it is folded over. You'd need to print continuously on the printer though since you wouldn't be able to stop extruding.

I appreciate this pitch, first video I've hit like for in at least a year, good job

Great video and analysis at the end. Yes, not an energy efficient way to recycle but shows what may be done by the ingenious minds of us makers. Bravo!

Perhaps you could use an extruder and second, slightly hotter nozzle after the first to add pressure to the system going through the second nozzle to eliminate the hole in the center? If you synchronized the extruder to the filament roll there shouldn't be a speed discrepancy, and you'd still have a pulling mechanism through the first nozzle. I'd try it myself but my printer is dead atm...

Very cool video! I always thought that this process needed a feedback loop, but seeing how easy this is, i might have to try this myself! what happens if you increase/decrease the pull speed?

A curious child is a wonderful thing and I couldn't think of someone better to be the teacher. Your daughter is going to be one smart kiddo

The squeaky joint noises when showing the pneumatic robot arm... your ad placement is probably the best of any RU-vidr. I hated dealing with the bottles and got a filtered water pitcher instead. So I really don't have PET bottles. But if I did, I'd probably make one of these because they're fantastic.

Do you think the lower print speeds for bottle PET could be related to the increased part strength?

How much does the selfmade Filament cost per Kilo and how much would the same have been as Pfand?

I just downloaded the files about a day before you posted this video, I'm excited to reuse some of my parts lying around. Since I moved, I've been drinking bottled water, so I'm looking forward to experimenting with recycling those as well as pop bottles. I don't think I'll be going around collecting garbage just yet, that comes once I used up all my own waste 😆 I always like the testing mindset and the surprise bonus strength of annealing is definitely something I'll keep in mind! Great work!

Hearing your voice always makes me happy. I hope you're well!

1.75mm drill bits are hard to get in the US, but you can use 1/16th or a 1.5mm and then use metal polishing compound on a string to sand the hole out to size. As a bonus this removes any burs that can damage your filament and makes pulling the filament much easier. Just be careful to widen the hole evenly and test the size often because it gets bigger faster than you'd think.

Does this process produce VOCs? If so do we need to create a filtration process?

I work in a diesel shop and we get pallets of water bottles in to drink. All of these are tossed in the garbage but the advantage is that they're all the same brand they're all clean and the shop even gave me a recycling bin so that I can use the filament out of it! I pretty much have the perfect setup and although I have doubts about how much time it would take I'm sitting on a potential Gold mine of filament

Nice, I've always wanted to print a waterbottle from a waterbottle! And when it breaks, I can just do it again! Infinite waterbottle!

The expression "Penny wise, pound foolish" comes to mind.

Next project for Stefan: a filament stitcher. The fibre optic ones use laser, I beleive, but perhaps a heat block with 1.75mm ID split in half that clamps down and softens two ends of filaments may work to join them.

Legit have been waiting on this lol, I saved all the components of my old E3 for this exact reason. Thanks Stefan!

"Recycling bottles into 3D printing filament is so smart! Can't wait to try it out! ♻️"

IMHO, this transformation should be done by the recycler companies; as an individual, the cleaning, label and wrinkle removal is a lot of work with possibly polluted filament and low yield. As for solving the environmental pollution issue, I think it is better to restrain oneself in the usage of these kind of recipients. Avoiding waste is better than recycling it, isn't it ?

It's worth mentioning that you should always use a release layer like gluestick when printing PET filament - it *will* rip chunks out of smooth PEI and glass sheets!

😎 I have been using this method for about two years, but the PETPull-2 version. Since then I have not bought filament again.

Very cool and fantastic design from the creator. Ty for the video.

One of the reasons you’re getting terrible flow out of the PET filament is because you aren’t printing it hot enough. If you turn up the temperature to about 300°c, you’ll get noticeably better flow than PETG.

I'm currently making one of these and stumbled upon this video. Awesome to see you're joining in on this too!

Ur doing the nature's work

Definitely something i need to try

I’ve been using the recreator 3D funsize for quite some time, it turns out great filament

i've had their homepage open in one of my 1000 tabs for well over a year now. finally gonna build one, it really doesn't seem terribly complicated and well worth it. but i'll use a CR10 and wood for the most part, over a spool and days of printing doesn't seem necessary. i think i might also not wash out my cola bottles too well and perhaps end up with some nice caramel filament. thank you for the push, its a great project.

A great proof-of-concept, especially basing it on a widely-available standard base (Ender 3). I think that easy shredding of plastic is the next problem we need to be tackling.

I'm absolutely going to be making one of these sometime soon!

big help thank you

I'm excited, I finally have a new printer to replace my ender 3 on the way so I could turn the old thing into this~

Great video! I love videos on recycling plastic! Please do a full video on crystallized PET! Keep up the great work!

Finally a good use of Ender 3

Fantastic! we definitely need to be recycling these plastic scraps 👍👍👍

Useful, this will make a worldly difference that needs to be promoted (as you are :), thank you!

I would like to see more about heat treatments for PET including differences made by changing the cooling rate. For example, does the crystalline strength change if you heat and then immerse in a liquid, put it in a refrigerator, etc.

I am actually looking forward to this. I have an ender 3 that makes excellent impromptu birds nests and no amount of tinkering has solved its issues. So, I'm going to tackle your project and make it into my filament maker for my other two 3d printers.

I would love to see a deep dive into the strength, especially for use in combat robotics

I feel like something is never a waste of time if it solves a problem or produces something of value, and there's no arguing these bottles are a problem begging for a solution. Sure, a discussion can be had around efficiency and it won't be for everyone, but I think that reusing an old tool in order to reuse a material was quite an elegant and interesting solution.

ngl bro this idea inspires me to build this first whenever i can finally afford to buy a 3d printer

What youre teaching your daughter about the environment is amazing 7:40

Strip protrusion could be the ideal method for diy continuous fiber filament because 1st it should produce a decent fiber core concentricity without needing a specialized nozzle because it simple wraps the strip around the fiber, and 2nd it provides the necessary tension on the fiber.

I bought a bambulab printer and figured it's probably a good idea to look into recycling filament... the timing on this video... considering I'm one of the people that bought a CE3Pro and didn't really care to hook it back up. Love this channel

Absolutely love it.

What about overall energy required to recycle a single pet bottle?

Thank you so much for this vid, awesome stuff

One of the best creative content creator on youtube

So bekommt mein Sidewinder eine neue Aufgabe! Eine sehr tolle Upcycling Umsetzung

Really cool dood!! Yes i tempted

I would love to see the comparison of annealed filaments across multiple materials. Low temperature crystral structure reflow could be an interesting investigation for not just pet but also pla, asa, etc.

Turning the Ender 3 parts into a different tool has a similar vibe to the alternate builds of a Lego kit which they used to show on the back of the box 😁

I've been wanting to build one of these for some years

I am interested to make this machine. I will watch this video again and again until this knowledge sinks in my mind.

I hope the community comes up with a working recycling machine that works with shredded bottles and produces a homogenous mixture of the entire batch.

Excelente vídeo, ya te habias tardado, pero sabia que tarde o temprano ibamos a tener un video tuyo de este tema, profesionalismo y honestidad con un toque de mas humano Saludos desde mexico

Ideally I believe melting the plastic down into a thin block, then using a series of heated rollers to press it into a thin sheet which can be pultruded may yield better results. Under such a method, any plastic could be pultruded given it would be first formed into an appropriate block.

I've been watching this PET bottles to filament projects from at least 2 years. With all the respect to your hard work Stefan, in my opinion this has nothing to do with ecology. Yeah, great words about saving the planet, clean up the environment, recycling materials etc. Dirty bottles are unusable as long as you don't want to spend clean water we suppose to save (how many litres per single bottle?). Detergents, acetone to remove labels and then dirty paper towels (harmful to environment). Reshaping empty bottle using 600-1200W heatgun (around 50% of electricity in Germany is produced from fossil fuels). 2 hours of "printing" to get 31g of material out of 80g (low efficiency, low productivity, electricity). Pure PET is hard to print, require higher than PETG temperature and slower print time (again wasted electricity compared to other materials). Now the most important part for me. What should I do with this short pieces of filament? I've never seen a single "bottles to filament" project making even 100g of filament in one piece. If we keep using electricity, acetone, cleaning materials and our precious time to keep printing 3D benchys, were not helping our environment. We are working against it. I'm sorry for being honest.

Love the video, a good one to print instead of this(which requires an unused ender 3) is one by Tylman design, it is a kit that you can buy all the hardware and print the casing out yourself. Really fun project and he ships the kits out super fast