Deceptively Simple Process Turns Bottles Into Filament

If you know that most soda bottles are made from PET plastic, you’ve probably thought about how you could make filament from them and have an endless supply of cheap printing material. [Mr3DPrint] says he has a method and shares a few videos that make it look easy. We wonder if the quality of the filament is up to par with commercial products, but assuming the videos are accurate, it appears that the resulting filament gets the job done.

The details are a little sketchy, but it looks simple enough. THe first step is to get any indentations out of the bottle. He has several demonstrations of this some using pressurized air in the bottle and some without. In each case, though, a drill holds the bottle through the cap and spins it over a flame until the surface is smooth.

Once mostly smoothed out, he cuts the bottom off the bottle and uses a mounted razor blade to cut the bottle into a thin strip. The next jig is a standard hot end with a 1.75mm nozzle mounted horizontally. We assume he drilled out a standard nozzle with a 1.75mm hole or, accounting for die swell, maybe a bit smaller hole.

Presumably, the razor produced strips narrow enough to fit the top of the nozzle. He pushes the plastic through and uses a printer heat break to tie it to a string and a little machine pulls the rest through while winding it up on a takeup spool.

Our guess is the diameter of the filament is all over the place and we wondered if placing the filament vertically or horizontally would give the best results. However, this seems like it would be an easy thing to cobble together if you wanted to try it. In the comments, [Mr3DPrint] mentions some temperature ranges to get you started, but we didn’t find any reference for the designs of the machines. On the other hand, they seemed simple enough to build, especially if you have a 3D printer.

The extruders we’ve seen are much more complicated. Most of those, though, use pellets.

68 thoughts on “Deceptively Simple Process Turns Bottles Into Filament

      1. @Alice. 95% of all plastic ending up in the ocean comes from just a few rivers in Asia and Africa. Look that up to convince yourself. If you aren’t living in China / India / Nigeria, then your plastic waste isn’t going into the ocean.

        1. Most of the developed world ships a very high percentage of their plastic waste to china and other countries for processing. So don’t think just because you don’t live in one of those places that it isn’t your trash ending up in the ocean

        2. A lot of that plastic originates in other countries and is shipped over to them. Reusing it here does have potentially save some amount of pollution elsewhere. And the time involved shown in the video is negligible.

          1. Except, most of plastic from these bottles, if reused in printers, will end up in land fills anyway. Typically, 3d printed “trinkets” are not family heirlooms that are kept around for long..

        3. When you look at the content of that leaking plastic its nearly all western world ‘recycling’ that didn’t get recycled, so taking the plastic away from your local waste managers that clearly can’t be trusted to sell/subcontract to those that actually will recycle it is a good thing.

    1. Why is everything about money? The plastic is getting reused and it doesn’t look that labour intensive. It could be a option for anyone that doesn’t have easy access to filament.

      1. Because it may cost a lot of money. In countries that implement a bottle return system, you pay something like 25 cents for a large soda bottle and get it back when you recycle it.

        If one 2 liter soda bottle contains 70 grams of plastic, of which you may recover maybe 30 grams (most of it is in the bottom and the neck), that’s $83.33 per kilogram of plastic. That’s about twice the cost of buying a spool of new filament. Add the labor cost and you’re well off into not making any sense to recycle it.

        You might still think “It’s only money”, but when you spend more money to do something, you make the economy run and consume more resources anyways.

          1. Right you are. However, adding in $14 an hour minimum wage, processing those 34 bottles like this would take several hours and bring us back to the same point. If you take 5 min per bottle, that’s about 40 bucks plus the cost of materials, which is still more than the typical cost of filament per kg.

        1. I pay 10 cents deposit on a 1L, and like 99% of people I get zero for recycling it at the curb. If I were to collect a pound of containers I could get around $1.40 for them at the current market rate. If I don’t crush them I get 10 cents if I find a store with a reverse vending machine. I probably lose $50-$100 in CRV a year by handing it over to my garbage company.

          1. Over here we get 25 cents for soda bottles. In every shop.
            This has led to poor people rummaging through garbage bins in public places or walking through trains from front to end in search for bottles and signs being put up that ask for bottles to be placed next to garbage bins instead of in them. The same is true for aluminium cans, but you rarely see them nowadays.

            And while refilled bottles are much more valuable for the industry, you get much less when returning them (8 cents for a beer bottle). People still collect and return them for the money.

        2. In countries where plastic bottles are not involved in any return system, they’re garbage and free once used. In these places, it’s MUCH cheaper to recycle them as filament than to throw them in the trash. So it really just depends on the situation. Sometimes you’re right, sometimes you’re wrong.

        3. Your logic is economical, but unsustainable.
          Remember people might be trying to “save the planet” (even though that’s a bit of a misnomer, since the planet’s recovered from worse things than humans in the past). And for that equation you need to disregard financial costs and look at the environmental costs.
          Energy right now is produced from various sources, some of which are admittedly not very environmentally friendly.
          There isn’t a real environmental impact associated with you working on something like this, since the activity isn’t physically taxing you’ll use the same amount of resources doing this as you would sitting down doing nothing.

          So it’s down to pollution from landfills (plus the resource budget to get the old plastics there + buying a new spool of filament and the resource costs associated with that process) vs recycling in your kitchen (plus the resources associated with that).

          So anyone wanting more than just save money would have to calculate those factors for themselves and decide what’s really better. In the meantime it’s a neat hack.

        4. im not sure i understnad your thinking.. labor for what? Why would i charge myself $14 an hour to process bottles for a home diy project on my day off? thats like saying i should charge myself to pee, thats alot of work sometimes, or charge labor for my wife to make dinner. Somethings are just part of life and making a kick ass filament machine that uses plastic i have sitting around the house is one of those things of life.

  1. Good one, nice post thanks for sharing :-)
    Recycling options abound, I do a bit – border line hoarding though, eek !
    In Western Australia we get 10c for each water drinks bottle, cans, glass bottles only minor exceptions, so its a toss up to collect bottles for dosh vs converting to filament in any sort of commercial review of Time, so will get to video in due course. Good for educating kids though and the retired people I meet in their 60+ years of age who only just started to discover science, let alone engineering principles, one wonders how they lived most of their lives one in particular without late primary school arithmetic who imagines the only reason we are not squashed flat by Earth’s gravity is the universe’s gravity in some way “balances” that, yikes…

      1. Ah, well I have no idea what your over there is, as to where its your “here” ? Though I hear you get more money – is it taxable as income but, in any case who over there subsidizes the program Or does it make a profit – unlike over here where drinks in those containers have gone up more than 10c even 50c or more but, unclear of the commercial relationship between manufacturers and government ie. If some of the bottle manufacturer’s revenue increase supports the recycling scheme.

        1. Finland;
          0.5l PET bottle – 0.20€
          1.5l PET bottle – 0.40€
          Aluminium can – 0.15€
          Glass bottle – 0.10€

          Working system – hardly see any of them thrown around for any longer period of time.

    1. on the other side of the island we also get 10c per bottle so cost wise buying a reel is probably a cheaper option – but

      Im bemused by how easily it appears to pull through the hot end – I would have thought the tension would have just pulled the molten part away from the hot end and pulled more plastic ribbon in – unless it is not hot enough to melt the plastic fully but just soften it sufficiently to squeeze it into a new shape – Ill have to go back and have a closer look at the description on the video

      1. Exactly right: he’s deforming the ribbon into a tube (potentially completely solid) at a temperature above the plastic deformation temp, but below a full melt. He mentions having a very consistent ribbon width as important. Using a lower temp like this should have positive effects on the quality of the polymer as well: the more often and hotter you melt most plastics, the more degradation you see.

    2. Gee, was WA that backward?

      I’m in NSW and almost 70. I learnt about DNA, Physics, Chemistry, Geology, Geography, Maths and English in High School. In maths we learnt about i^2 = -1 which became j^2 = -1 in electronics the following year.

      1. Ah well a Peter Sellers fan, great to have been there ;-)
        I understand in NSW as well as other states besides WA at the end of the 3rd year high school (mandated, unless for medical reasons) the kids (with likely parental influence) have a choice of 2 more years schooling for so called 4th and 5th year just just before uni – that was a national/federal guideline. Nothing to do with WA being in any way backwards – it was a matter of choice. So if you wanted to do 4th 5th year with those units you mention then same as I but, many dropped out after 3rd year which I have met they went into law, accounting, sales, naturopathy, farming, counselling etc – so far only one seems to have missed last year primary – I could find out if home schooled (likely).

        My best friend of the time didn’t continue after 3rd yr, instead worked for various places whilst I continued studies, I recall him lauding over the fact he had more spending money than me, at the time :P
        I really didn’t care, I was and still am curious of so many disparate areas of science – especially physics and engineering technology – and towards medicine which I recall will become or already is ahead of time a branch of engineering ;-)

        Where those key units you mentioned are taught with some elective options which could be switched half year or after 4th year – in my case I didn’t do biology, instead everything for subsequent electronics engineering at the then W.A.I.T. mid 1970’s now Curtin University (same campus). My interest in so called imaginary numbers (ie. a number which is simultaneously an operator) wasn’t foremost in my mind – instead I was mostly into chess and poker and thus stats. Engineering of course in my first year broached the issue of filters and phase where i (as operator mostly) became essential as without it filters cant be designed. So if any pragmatist out there who doubts the value of so called imaginary numbers then let them try some filter design and RF even AC circuits and especially 3-phase eg so called tri-plen harmonics and passive filter networks – though that was mid 1970’s. These days of course active filters and high end “alien like” MOSFETs scarily handle huge currents at great efficiencies (have a few for a physics experiment switching 1500Amps) and with recent SiC for high voltages in tiny packages. ie Finding a MOSFET with 8 micro Ohms resistance in On state versus many Giga Ohms in Off state pretty amazing. Just saw a 7KV SiC device that does an Amp switches from 10v gate, early lab proof of concept with time to market shortening each few months – benefit of competition.

        Then I went back to human biology facets through food chemistry post grad 2010 – Everything has changed !!

        Best thing I ever did, regret a little not doing a biochem unit at WAIT 1970’s and therefore regret not doing biology at high school. Ive since caught up rather well, bacteriology, distinction in food chemistry, considering PhD – unless my physics experiment works even 1%, visitors not allowed ;-)

  2. There are handicapped people in need for some kind of job. They usualy work in some kind of subsidized supervised workshop, making ceramic trinkets to be sold as part of fundraising campaing for their cause. I think this would be perfect match for that kind of operation. Also people in some rural parts of asia often manualy recycle lots plastics and electronics. I even saw documentary about people in china making recycled plastic pellets from plastic bags in their backyard, since they live near the place where local government dumps plastic waste.

  3. I can’t help noticing that PET isn’t one of the choices for 3D printer filament. I assume that’s because it doesn’t work well.

    You can’t dispute that used PET bottles are readily available, but an inexhaustible supply of the wrong material isn’t useful.

        1. I see both, along with various grades of nylons and about a million other thermoplastics… They are all available, and often any of them can be used, for many prints looks/shape not material properties are what matter…

          Though there are occasions you need to print in the right material too – generally despite its many downsides I stick with ABS for the easy sovlent welding to modify most things – as ABS is hugely common in the world, and its generally good material properties. Still not really given a proper go to vapour smoothing, but at some point I’ll have to.

          1. i guess there are big regional difference because i have never seen PET filament here ( Switzerland)
            i never use ABS because of environmental concerns and because of the smell/toxic fumes it emits event thought as you point out there are many advantage to using it.

          2. PET on its own is not at all common, but having a quick look through some of my usual suppliers they do offer it alongside some other less common filaments.

            And yes while I don’t see ABS as a major environmental concern at all – better turning that oil into long life plastic parts than burning it, I have often looked at other materials and wondered if its worth trying to switch to reduce the need for that hassle dealing with the fumes – my current setup doesn’t really deal with them at all, just close the cupboard door to keep it mostly in the cupboard, and close the room door while its printing to keep it out the rest of the house – then vent room throughly when done – far from ideal but a more high quality 3d printing experience (or printer for that matter) hasn’t made the top of my list yet…

    1. Not every part needs to have the best material properties. Hell, in my workplace it only just became widely acceptable over the last couple years to use 3D printed parts of any kind for anything, before then you’d have to write up a risk analysis for every single 3D printed part, and they would always be a point of contention with customers, no matter how little load they had to support. It was often easier (and cheaper from a time cost perspective) to just buy some milled aluminum for a part that would take 30min on our filament printer.

      Sometimes you just need something with the right shape and strength doesn’t matter.

      1. but that’s kinda the point Carl was trying to make PET is not the same as PETG !!! notice the extra G? that not there just for fun it stands for glycol so the plastic is different at the molecular level. similar yes but not identical. you cant just magically make it have the same properties/printability as PETG.

    2. If PET actually prints then I have an interesting idea for a large filament source.

      Industrial strapping tape that you find wrapped around large boxes.

      My idea was to design a hot end that would accommodate the tape directly accounting for its size and thickness.

  4. Okay. How does one get I touch with mr3dprint? I would like to have a bit more information. I’m retired and have time on my hands would enjoy trying it some time.

  5. I think the process works, but I also think we are being shown ‘cream of the crop’ video.
    The wall thickness of a blown PET bottle is all over the place; thick at the ends, thin in the middle section.
    A ribbon of constant width doesn’t automatically equate to constant volume per length!
    In other words, parts of that ribbon should be cut wider (where it is thin) and narrower (where it is thick).
    Great idea, but I can’t help but feel that lots of details have been left out.

    1. Worse: many soda bottles are double layer with PET/PE for the mechanical properties, and something like nylon for a gas barrier. Single plastic bottles may also use a double layer construction with some chemical coating between the layers for the same purpose. Monolayer bottles can also be chemically coated on the inside, either modifying the plastic, or infusing the top few nanometers with something like silicon dioxide to fill in the gaps in the polymer.

      This is also the reason why PET bottles are hardly recycled. They come with such a variety of additives that you can’t mix a whole lot of reclaimed plastic into the new bottles or the quality would suffer. Instead, they downcycle most of the bottles into cheap fiber and fabric, which are basically throwaway products since the mechanical quality of mixed recycled plastic is bad.

      The resulting filament will become a mixture with sub-par properties. It may also de-laminate and break apart either in the making or in the printer.

    2. That heating under internal pressure should even it out rather well done right – when it balloons looking very similar everywhere it will be more equal, though I agree a simple feedback mechanism to slow/speed the pull really would be sensible and improve the results.

      1. It tends to balloon out more where the material is thinner, making it even thinner. I don’t see a reason why it would magically equalize the film thickness unless you bring it all the way to liquid and make like a soap bubble.

        1. It balloons only where you soften it with the heat – so softening it, focusing on the thicker areas till it look like an even a balloon means the thickness will approach equal – as that is why its looking like an even balloon. Its not heating the whole thing all at once.

          1. Dude if you watch he isn’t blowing the whole thing up bigger, he focuses the heat on the thicker bits till they balloon about as much as the middle – its thinning the thick bits by softening them and letting the air pressure work against it, which draws them out expanding into a thinner more even sheet with the rest of the bottle.

            Though I’m not saying its anywhere near 100% effective and evening the wall thickness you can see just looking at it how much the thicker parts swell to match the thinner parts.

          2. > Its not heating the whole thing all at once.

            Even worse – you’re modifying one part and leaving boundaries and transitions. Remember, the bottle expands in all directions, so thin and thick bands can have the same outer diameter because the bottle is elongated.

          3. Dude he is focusing the heat at the bits that need it, which is not at all the same as having no heat anywhere else its all spreading so there are not hard transitions!

            The whole point I’m trying to make is if the bottle still thicker and stiffer anywhere under that internal pressure it won’t balloon at all evenly – it will really show as it did when he first pressurised the bottle cold… Apply some heat to those areas so it softens and watch as the pressure pushes it out to about the right thickness!

  6. I find those discussions about who lets more (or less) plastic slip to the oceans rather funny, if not, quite sad.
    Plastic going into the oceans is a really worrysome problem, but countries who are big plastic producers and consumers, like the USA, aren’t doing any more good in “reducing their share of plastic in the oceans” when they fail at reducing plastic production dramatically. If the plastic output is not reduced, nothing else makes sense: recycling rates are usually very low or insignificant and recirculating plastic is only a new chance for it to shed more unrecoverable microplastic or for it to eventually miss recycling, and other alternatives for it to not go to the oceans are incinerating it, sending it to a landfill or delegating his destiny to another country, who will not have a better solution too.
    All the talk about “plastic in the oceans” seems just like smokescreen for hiding the big users and producers, and pointing fingers at those who receive all the foreign scrap, after all, I don’t think Dow Chemical, Basf, Braskem, Exxonmobil, or any similar industry, has Nigeria or the Philippines as their main consumers.

  7. Interesting comments. So where does most plastic end up when it is recycled? I mean from a regular consumers pov of “recycling”, like the containers we fill for the waste management company to take, or the local recycling center for that matter. If a project just interrupts the cycle of recycling, I’m not sure it is a good thing, when thinking about recycling that is, as one is just adding at the least energy cost to that lump of plastic before it actually gets recycled, assuming it gets recycled at some point. Maybe this is a difference between recycle and repurpose, perhaps reuse as well. All good things I suppose, but it may depend on what the item is, how much of a resource is in said item, how much resource was brought to bear to bring it about. Initially, I’m thinking there might be a difference between reuse/repurpose of a piece of wood as opposed to a plastic bottle. Good article, more stuff to think about.

    1. Problem with most ‘recycling’ for plastics we see in the western world is that it is a flat out lie, maybe a tiny percentage really is sorted and recycled properly, but most of it isn’t, just get dumped in the ‘third world’ somewhere…

      So taking it out of the ‘official’ cycle of recycling to make useful objects, or even just use as the early prototype filament is definitely a win. Rather than turning into waste, probably after just one use, maybe if its lucky after its recycled once its got a long useful life as that object, or at least saved precious virgin plastic from being used for the rough prototyping phase.

      I’d agree though there is a big difference between reuse/repurpose and full recycle/remanufacture that varies by material value and easy of recycling – Metals and Glass are easy to recycle and have a great deal of energy invested in their initial purification/creation from raw materials, where Wood really can’t be recycled, just repurposed or turned into not at all woodlike paper products (which is still a good thing). Plastics are somewhere in the middle they can be recycled, but often for a loss of material properties.

      No matter what though reuse is always best, at least from an ecological standpoint – even with glass bottles say, where using them to make new glass bottles is entirely possible and doesn’t have major quality issues just giving them a proper clean and reusing is vastly less energy.

    1. Jon, he’s pulling the filament from the hot-end with the winding spooler mechanism, not pushing it into the hot-end at all. So no drive/feed mechanism is needed, just a pull/spool mechanism.

  8. It’s too bad that less than 2% of what you put out for recycling ACTUALLY gets recycled.

    Virgin plastic is cheaper than recycled plastic. Everything in the chain is for-profit.

    You should have connected those dots before now…

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.