[Update October 13th] Working very hard right now to organize the shipping of the Pellet Extruder and other tasks related to Greenboy3D like organizing a supply of pellets for everyone and a pellet extruder wiki, which is why I sometimes have trouble answering everyone immediately. Most parts for the extruder have already arrived and have been processed & packed for shipping, however the pellet extruder screws are still in manufacturing since this is the most complicated and costly part of the pellet extruder. I believe that the first units will be shipped out by the end of October. Due to the additional organizational tasks related to the supply of pellets, there is simply more work that needs to be done. This includes staying in contact with many companies selling different types of plastics & watching out for good offers, requesting samples, testing them, negotiating prices, organizing the transportation of tons of plastic pellets, and so on. These tasks also take up some of my time, which would otherwise have been spent on the Wiki/Knowledge Base platform and the guide videos. Just a couple days ago, for example, I ordered and paid for 8,600 kg of PLA pellets, which will likely be transported to my location from Slovenia next week. 2500 kg PETG pellets are already in my possession, but more of other types are needed like: ABS/ASA/ABS-GF/ABS-GB PET TPU/TPE (various Shore levels) PA GF/CF Additionally soon there will be a new update about the ongoing progress (including the one about pellets) 🙂 If you want to get some additional pellets along with your order then that is possible however you need to consider that, while the pellet extruder will be shipped for free there will be additional shipping costs for pellets. However I will keep you updated regarding the pellets. The only thing I am not sure about is whether I should also get these extremely elastic pellets I showed in one of my videos, after doing more research it turned out these were not Shore 5A, but Shore 30 (with out the "A") which is significantly below Shore 0A meaning even more elastic and flexible than Shore 0A The only problem is that these pellets are quite expensive compared to other plastics and could only be offered for prices like 12,99 USD/EUR per kg (way higher than the other plastic types like PLA, which will probably go for 3,5 USD/EUR) Should you have more questions or concerns please feel free to contact me anytime - I am here to help.
Hi My idea for coloring is using different color inkjet bottles they feeding directly from side your extruder simply using stepper motor pumps. So you can run the color(s) slow or higher speed, or in a mix of different colors into the extruder.
@@madison66 a similar thought occured to me also. Realistically, for any innovative 3d printing system to try and take on the giants like prusa, creality, or bambu labs, then it needs to have an answer to multi material or multi colour systems with a user experience at least half way as refined. Automated side feeding of colourant into the feeder could be part of that.
@@madison66 theoretically you would only need red blue yellow black and white to make any color combination you could use two or more print heads to achieve full color 3d prints
I loved your research and the video - but I’ll be blatantly honest about the print quality: it’s not there yet. The Benchy has VERY inconsistent layer lines. On a commercial printer I would have returned it. The question is: are the inconsistencies generated by your pallet extruder? Or is it the mechanics of the printer that are suffering from a heavy print head? I would love to see it on a stiffer printer / maybe it gets good on a printer with a more rigid gantry, linear rails or wider gantry extrusion… more tension on the belts. Please, don’t dismiss the importance of print quality - people are willing to pay more for better quality filament exactly because they want smoother prints. Keep going the awesome research!
The issue is caused by the inconsistency that plastic pellets in general have. It's like having a Filament spool with fluctuation a diameter between 5% but because of this it costs also 5 to 10 times less. This issue can be solved in the future for example with a pressure sensor in the extruder barrel or a small AI camera monitoring the extrusion rate and real-time-micro adjusting it, which is why pellet extruders will have in the future basically the same print quality as filament printers. But this this kind of technology still needs to be developed... If I make enough sales, I will try to develop this technology with extrusion measurement myself, since it would be my dream to bring a huge innovation to the world of 3D Printing... Wish me LUCK🙂
@@greenboy3dyou could try an esp32-cam or 2 with a macro lens focused on the hotend, as a bonus you can detect blobs before they cause issues. Imo another option might be double extrusion, with a pellet extruder extruding into a bowden tube and the filament being picked up by a regular hotend, that allows you to use light to measure the diameter, and to use a way lighter moving assembly. (Also depending on print speed you could have a shell that's solid but the core still above 60C, which means more flow rate.) Yet another option (imo the easiest to do currently) would be to use an idex system and do the outer wall on a regular .4mm nozzle, then fill in with the pellet extruder on a 1.2 or 1.6mm nozzle. The great thing with that solution imo, is that you don't need good quality pellets, and can use whatever color recycled plastic in it and still have an uniform print. And also the large diameter nozzle means that you don't need the heavy pellet extruder to move fast.
I have access to literal tons of both engineering and commodity resins for free that's thrown away from my workplace. I think you 100% need to pick a nozzle that's optimal for pellet extrusion more so than maintaining broad compatibility.
1kg PETG cost here 6-9 usd per kg. That is so cheap, if it would be free, I would not print more. In the company we also put hugh amounts of PA6 and POM in the junk
Special nozzle all the way. If a standard platform for pellet extruders eventually comes out as a result too, it just means this will have set the ground for that standard! If there is demand, standardization will come, and people will prefer the performance gain and the eventual affordability.
My 2 cents on your question about "standard nozzles": not critical, especially if they are suboptimal for this extrusion technique. Most important is a pellet extruder with optimal heating & retraction results. So if that means a different nozzle type, so be it. Fantastic work by the way ! 👍 Been saving a ton of PLA parts in the hope of recycling them into new prints. But nothing out there seems affordable in terms of time , money or reliability. Most require recycling into filament first, which feels like a dead end for consumers: too many issues, too time consuming, somewhat expensive. Your approach that just requires careful grinding seems the most promising to me so far. And seems good enough for my prototyping workflow for functional parts: many iterations, no need for perfect surface quality. Thanks for your 2 years of effort on this front !
I agree on this. CNC Kitchen even tried to build a high flow nozzle out of a regular brass nozzle and a copper insert when HF nozzles were a novelty. So already manufactured nozzles can be adapted "quick and cheap" with a round copper insert by just press-fitting I guess, and new manufactured nozzles just with one hole less! I am sure that with the visibility you are earning through your videos, you can achieve something like that pretty easily. However, I would let the nozzle compatibility window open.
Agreed - for PLA the time and effort isn't worth it - aesthetic prints like cosplay items, busts, toys etc. wouldn't have the look or shape consistency needed, and the functional parts may not have the tolerances needed - it could be great for printing support or infill perhaps but then that introduces a whole other series of challenges in slicing or IDEX integration. I would recommend this extruder for printing thermoplastic elastomers that are too soft to turn into filament, or perhaps make printed parts from materials other than standard and widely available filaments.
A question that can't get out of my mind is what people would do if they had a nozzle clog, which can happen on any 3D printer due to some particle blocking the nozzle hole. If the "special nozzle" is not easily available globally because only I am selling it, then this might be a problem. Any ideas?
@@greenboy3d depends on your goals, I guess. If it is to make money on nozzles then indeed it could become an impediment for buyers. If not, publishing the design specs will get a long way toward copycat nozzles appearing on aliexpress or ebay. With many biz models in between (partnering w 3D printing accessory distributors on each continent, selling through specialized online stores that have figured out worlwide buy & ship like CNC Kitchen's, etc). BUT the nozzle is only a tiny concerning piece in my opinion. How about other proprietary or uncommon parts of the extruder and providing spare parts for repair ? Like the extruder screw, or its housing ? Screw extruders do NOT have a great reputation on the hobby scene. So, as a buyer I'd be less worried about the nozzle (I'll likely find a way to unclog it) and more about whether I'd be able to get spare parts if anything breaks down ? Bambu Lab might be an example to ponder. The highly specialised parts in their printers initially turned many people off for fear of ending up with an expensive unrepairable dead paperweight. But their policy of selling most parts online at a reasonable price went a long way to assuage these fears. Maybe something can also be learned / copied from that ?
@@greenboy3d I'm quite interested in more details on shredding the plastic to make it workable, or is it really as simple as just throwing parts in a kitchen blender and sifting the result? I would also like to know more about your 2-year journey (what major hurdles did you encounter and how did you tackle them) to get at this point, and on how to get my hands on the result to be able to play with it.
Idea: try adding some new pellets to your shredded waste, somewhere between 10-40% of new pellets. This is normally done in recycling as it increases quality of the end product by a lot. And 90% of recycled material will still safe a lot of waste
I mean, 100% recycled material or 90% recycled material should safe the same amount of waste, since you are propably using more waste than you produce (if over 90% of your prints are waste, you are propably doing Something wrong)
Put me down as "don't care" about nozzle compatibility. Nozzles are cheap, they last a long time, and if a special extruder needs a special nozzle, I can understand that. As long as I can GET them, or I can modify standard ones to work. Are you thinking of tapering the inner profile? What would make a standard work better on a pellet extruder?
A question that can't get out of my mind is what people would do if they had a nozzle clog, which can happen on any 3D printer due to some particle blocking the nozzle hole. If the "special nozzle" is not easily available globally because only I am selling it, then this might be a problem. Any ideas?
@greenboy3d the idea to modify a stndard nozzle from @BrightBlueJim would be the easiest way in terms of accesability. Even when you break multiple ones during the process of modifying or if they clogged faster you can get a inexpensive replacement. The tapping tool could be a custom cadfile ready to cnc manufacture via different big manufactures, like JLC for example, same for special nozzles.
@@greenboy3di mean if you’re the only one selling it it could be a problem, but if you can solve the distrubution option it wouldnt be an issue imo. You could look into if modifying a nozzle thats readily avaliable is an option because then people could just get that nozzle and modify it themselves if buying one from you isnt an option.
@@greenboy3d I work in blow molding. We use stainless steel screens and place them after the extrusion screw. This catches contaminants and solves your clog. Your only option after that is solvents when removing your nozzle.
I am in absolute awe. You are ABSOLUTELY correct. There is a huge market that falls under the 'Good enough' category. It's not even bad print quality. I'd MUCH rather have a bunch of good enough prints made from all of my plastic waste. I will be ordering soon.
Thank you for your appreciation 🙂 If you have any sort of questions, feel free to contact me anytime. For example, through email or Instagram. I also suggest you to join the Greenboy Discord Community Server where you will encounter more than 1300+ like-minded people that discuss daily all sorts of things about pellet 3d printing > discord.gg/UuFDbFj6FC
Honestly the majority of new printers have proprietary nozzles, so if an optimal nozzle can be close in price, i dont see it being an issue. Really interested in this for an option for my sunlu s8 with a bigger nozzle to print my failed prontd
A question that can't get out of my mind is what people would do if they had a nozzle clog, which can happen on any 3D printer due to some particle blocking the nozzle hole. If the "special nozzle" is not easily available globally because only I am selling it, then this might be a problem. Any ideas?
@@greenboy3d Good point indeed. Humans on all the sides of the planet. may be you go as some success company do - make a cheap option and a pricy one. So you will also not let your interest go away for some ideas.
Cool to see progress on this! Reg.nozzle: why not just make custom nozzles that use m6 threads? You maintain compatibility, and you can also get the characteristics you want
A question that can't get out of my mind is what people would do if they had a nozzle clog, which can happen on any 3D printer due to some particle blocking the nozzle hole. If the "special nozzle" is not easily available globally because only I am selling it, then this might be a problem. Any ideas?
@greenboy3d well, it's not like you can buy normal nozzles at the local super market either, i think as long as you maintain compatibility with standard nozzles, it should not be an issue, otherwise you can sell them in packs of 10, that gives people alot of time to order a new set when they need it. I suspect that the custom nozzle is going to be all about having as little fillament volume as possible to avoid oozing, right? In that case, it might be even easier to clear the nozzle, and it would make it even more a non issue :)
@@satibel I think so is for metals and well, many other materials that melt at high temperatures, so i suppose it should be a similar story with plastic since it's so similar to those just in a much weaker and lighter (and cheaper) form
@@ERIK-457 yeah for metals the main issue is that most of them are alloys and you can't precisely control the quantity of each metal, so you have higher variance on properties with recycled metal compared to metal that is made from known sources.
I think a possibly better option is to look at making filament yourself. If you can make filament then you can use any off the shelf printers or parts. Splitting it into separate steps is likely better. That also removes the print quality concern. Another option may be to put the pellet extruder on top of the printer and have it create filament that goes straight into a Bowden tube down to the hot end, although this would likely not be as good as making the filament separately.
Supporting standard nozzles will bring more people in, so in the short term that's probably the better plan, but the next step will be to develop custom nozzles and setups that work better.
In this case, in terms of speed, I think a bed-slinger might actually perform better than a CoreXY. On a CoreXY the heavy toolhead would affect the resonances of both axis, which would limit accelerations. On a bed-slinger the heavy Y axis performs worse than the X, bottlenecking the X axis speeds. But in your case, both axis are heavy, so they don't limit each other, and because they're separate, it's probably going to be easier to deal with their respective resonances. But that's just my theory.
For recycling it's typically done by mixing the old plastic into a new batch. You could try how well the system works at different ratios of old and new. Ideally ofc you would want to shred it to similar granule size, but the hopper and screw system should be able to handle varying sizes. Would like to see some tests on this. Recycling old plastic inhouse can be a huge saving overall. For the nozzle question, I think you should aim to have optimal nozzle for the printing and not be chained by universality. If the system becomes widely used then it becomes the new standard :)
I used to work at Tektronix, and there was a shade of color the used a lot on plastic molded parts that we called "Tek blue". I learned one day that they made this by mixing medium blue, white, and black RIT fabric dye (the dry powdered version) to get this color. Of course, those were injection molded parts, so I don't know how that would work here.
A question that can't get out of my mind is what people would if they had a nozzle clog, which can happen on any 3D printer due to some particle blocking the nozzle hole. If the "special nozzle" is not easily available globally because only I am selling, then this might be a problem. Any ideas?
Nozzle clog can be solved by buying another one. Profit. Sell a 3D design of the nozzle for a substantial amount, or just publish it for free for the street cred
I calculated 112g/h = 38.3 mm^3/s, am I correct? Calculations below: Assuming PLA density is 1.24g/cm^3: 112g * 1.24 = 138cm^3/h 138 / 60 = 2.3cm^3/min 2.3 / 60 = 0.0383cm^3/s = 38.3mm^3/s flow It would be cool to see you perform the orcaslicer max flowrate test on different nozzle sizes. It makes it easier to see the actual limits of flowrate.
A bit of an overestimate on flow rate unfortunately - if the PLA is more than a gram per cm^3 then 112 g should represent less than 112 cm^3. I think the calculation should be as follows: 112 g/1.24 g/cm^3 = 90.3 cm^3 90.3 / 60 = 1.50 cm^3/min 1.5 / 60 = 0.025 cm^3/s = 25 mm^3/s Still better than many standard nozzles, but not matching top of the line high flow. I think beyond this, the cooling rates will be a bigger limits but with klipper input shaping the weight of the extruder may be much less of an issue than with previous firmware. Still a very impressive project in general!
I'll look into the "orcaslicer max flowrate test", might test it this way next time. Thank you for your very interesting input 🙂 What would you like to see next?
For the nozzle, maybe go with a CHT nozzle, since it has three channels in which the molten plastic flows through, meaning it might have more consistency in the heating and flow. As for the small discrepencies in the melting process, maybe add some small notches on the side of the extrusion screw so that any trapped air could be potentially allowed to bubble out or maybe as a way to control a bit of pressure. As for dying your batch, could try going with alcohol dyes and do it in the same format of mixing it in an enclosed jar/container.
I like the idea, however I think print quality is a vital thing to consider, especially from those of us who make a living off of making prints. Especially on dimensional accuracy, print quality is something needed.
The retraction explanation makes perfect sense when you consider that reversing the conveyer screw in a pellet extruder is functionally doing the same thing as a "retract" move on a filament extruder; namely, they both are reducing the pressure inside the melt chamber. Which is the real goal of a "retract" move anyways.
Some of the problems you're running into are also issues in small extruders. I.e. if the extruder screw is running at 60 rpm, expect an uneven "squirt" of resin every 1 second out the end of the nozzle/die.This is why metering pumps are often used at the end of an extruder (to tightly monitor flow). The 3d filament printer brilliantly solves this issue inexpensively by using filament. The filament feeding system is acting as the gear pump. Another issue will be safety. If you build up pressure behind molten plastic that has frozen off at the nozzle, watch out. Randcastle has made a great miniature extruder for decades to learn some of the inherent issues.
Very good video and so happy to see your progress. I have 40 years of high production high precision injection mold making and molding experience. You are dead on in the process you outlined and are mastering the rheology of thermoplastic flow . The only difference in the outcome is that you are not able to get the density of an infection molded part because you do not have a packing phase in 3d printing. As a result you may have aligned molecules and a grain like wood in printing that has some advantages and disadvantages. But with carbon fiber we may be able to produce a similar strength . Not sure yet all the doors you are going to unlock but accolades to you this this an amazing project. But I can tell you that the injection molding press industry has already figured out how to do a lot of what your doing they just have not figured out how to 3d an extruder for 3d printing
About the nozzles: could one not just cut away most of a standard nozzle with a hacksaw, leaving just enough threading to screw it in? This would eliminate the entire part where plastic is melted and leave just the final conical shrinky bit needed for printing. Given that this extruder isn't likely to be acquired by anyone non-technical, I think having "SAW ALL THE THINGS" as part of the build process isn't too demanding a step.
you are right on your conclusion but yet it would still make things more complicated for everyone... because if you would do it this way and optimize the pellet extruder for a shorter threat than people without this modification couldn't properly use the pellet extruder which is a problem :(
@@greenboy3d I was thinking more like - have a standard unmodified plan for a normal nozzle that everybody can follow, but mention that this is a possibility so that anyone willing can saw the needless bit off and enjoy a little extra effectiveness.
Honestly, if the nozzle is open source, it wont matter if you make it specific or universal, people can come out with their own to help improve etc. This is an awesome project and you're doing the community AND the environment a huge favor. Keep up the good work 👍💪
I’m all for recycling and reducing waste, also I like saving money, so I love it! Most prints just don't need good quality. Stuff like hooks, holders, storage containers and so on work just the same even with a +-1mm tolerance, so the reduced surface finish is no problem.
I once read an article about an Indian university student who couldn't afford a 3D printer and so he designed and built his own, the only things he bought were the arduino chip to build his control board and the steppers, he was also having trouble affording filament and so he also designed and built a pellet extruder as he could get the pellets for pennies. I have a few old bedslingers that I would love to convert into pellet printers!
Let's not forget that pellet 3D printing is also better for the environment. You're cutting out the entire step of shipping to the filament producer, extruding filament, spooling it, and packing it. I'm not a type of "oh my god the environment is everything and we have to drop everything to protect it"-person, but it's still nice when something is more efficient. I do wonder, what's causing the visible layer lines? Is it the relatively inconsistent extrusion rate?
I have collected around 8-9 kg of PLA waste in the course of 14 months and building an extruder to use them to print functional parts will be a much better use for them instead of giving them for recycling. I am very excited for the design files to come out and know where I can source the screw from so that I can recycle my waste into useful parts!
For recycling 3D prints, CNC Kitchen got the best results by repelletizing them. He was making filament but the basics were: 1.) Shred 2.) Sift 3.) Repeat till consistent 4.) Form filament 5.) Pelletize this first batch 6.) Form final filament Clearly you can skip 6 and probably combine 4 and 5 to be a single operation.
Nice. For recycled plastic the bulk density will be different. Building a small pelletizer extruder would be more consistent. I worked as a tool maker in an extrusion plant for 10 years. There are ways to make it more consistent but the complexity goes up.
Please note that if you have quality pellets (clean) you can produce filament on professional liine for 4-5 eur/kg. For instance TPU 65 shore D costs 3,8-4,8 euro kg now (Resinex, Ravathane 130 D65 NATURAL), regranulate 2 eur. Total is 6 euro kg while in store you pay 40 euro.
I just filled the form 😌 In short, for me it would be all about reducing waste from prototype prints: I would shred them and re-use the plastic. Most of the time, quality doesn't matter that much when prototyping iterations especially if you add some more tolerances if needed.
@@greenboy3d I'd like to see how the shredding process could be optimized, maybe with finer particles, filters and multiple iterations to reduce variance in extrusion. I'm guessing that pellets are a convenient form factor but not made with tight tolerances. And maybe it's possible to increase the print quality by shredding everything, including pellets themselves into finer particles with tighter tolerances.
You should check out the quantum delta 3d printer by Kilian Gosewisch, instead of moving the printhead it moves the entire bed on all axis and the extruder stays stationary, that way you can eliminate the size and weight constraints of your extruder.
Nozzle compatibility across filament printers makes sense, but it does not have to be that way for a pellet extruder. It makes more sense for the pellet extruder that uses a different extrusion method to have a different nozzle. Then from there, hopefully all future pellet extruders would have the same nozzle compatibility. Maybe if you just keep the same threading, it can be up to the user to decide if they want to use an easily available filament nozzle, or a pellet nozzle.
I feel that its worth mentioning that spraypaint isnt a great thing to be melting down, depending on the paint it could be very toxic. Additionally, when mixing pigments into plastics the formulation is tricky as the pigments can easily ruin your material properties...
PLA pellets can only reliably sourced at $10-$15/kg. The cheapest PLA filament is $15/kg… last I checked. 0-30% cost savings was not 300% (missinformation factor 10x to infinity)
My personal opinnion on the nozzle type is to stick with the current, widely available nozzles and once its widely adopted, offer specially designed nozzles for improved print quality. I would preffer that because of 2 reasons: - It would get a lot more people interested if they can retrofit their printers - They would have as may interchangeable parts as possible Moving on to the specialized nozzle: - It would offer to the people who initially are focused on print quality and not function a reason to recyle and be more sustainable - The hobyists would also benefit from having something to upgrade to
This is an awesome idea. It would be really nice to be able to actually make our own spools from pellets. Im sure you've already posted a video on why you didn't do that here.
Hey, this is great! I was looking for a convenient pellet extruder. I currently recycle old plastic and create my own filament, but that is much more work than directly printing the shredded material.
Nozzle compatibility isn't important if it comes at a cost of performance. What would ultimately be nice is to design it to function with a breakout board, and be compatible with toolswap plates such as the Hermit Crab or WhamBam Mutant for Ender 3, or be able to easily mount it to a toolchanging printer. That's a much nicer quality of life, to be able to quickly swap between a pellet extruder and filament extruder.
I dont like too much all the "green" theme, but i like the idea of self-sufficiency, i might use recycled plastic for low quality prototypes and only use top quality expensive filament for things that i want to be more permanent and of higher quality, i am a fanatic of quality by the way
Neat video, I love the amount of work folks put into FDM printers just to get a new way to print or not waste materials. I am not sure I would go through the trouble of doing all this, just to save a few bucks in materials, but I feel the pain of wasted materials as well. Resin printers create so much waste with support materials only being used and tossed. I love that you took and Ender 3V2 and did this, those things are like Honda Civics, cheap and easy to mod.
Love to see that project continuing... any estimation on when this will be available for testing? Would love to adapt one of those to my voron-toolchanger 😁
I love what you got here. The problem is layer lines. I would like to see this mod on something really good with that. If it's your system then open-source it to see how you can compress it. Current superchargers come to mind with compression.
Personally, I do not enjoy tinkering... that's why I went with a Bambu. However, this seems promising for prototyping. Keep pushing the envelope. That's the only way progress is made.
I think significantly better print quality is more important than full nozzle compatibility. However it would be great if you could maintain some level of compatibility, for example by keeping the M6 thread. Then the pellet extruder would work best with the custom nozzles but still function ok-ish with a common nozzle. Some compatibility would prevent complete downtime and allow the early adopters more ways to experiment - an easy one would be shortening common nozzles as from what you're saying a stubbier nozzle is closer to ideal.
Could you technically print pellets with the shredded prints so the size is consistent? As for the standard nozzle question, I'd love to see the difference in print quality with a non standard nozzle
Depends on how consistent the shredded prints are. If you have for example cylindrical shaped shredded plastic, and you plan to mix it with round shaped pellets, then you might get even more inconsistency... But yes you can...
I really like what you are trying to do. Even if the up front cost is most likely going to be more, the saving on the filament is huge. Making custom colors is interesting too. Are you using regular spray paint?
Yes, regular acrylic based spray paint. But like I mention there are multiple ways and spray paint types... By the way, what would you like to see next?
SAW VIDEO 1 AND NOW THIS!! Thank You! Can a pressure lift switch be added between the nozzle head assembly and the actual carrier mount guide rail? maybe this would be fore a different design. say 0.6 nozzle layer eight travel is 1.2. the stepper feed goes till the head is lifted ill the trigger head lift switch is tripped or lifted to that height, then the travel steppers are engaged to conduct gcode shape. maybe be do a reverse stepper feed code for retraction. Head should relax and lift switch switch off for next travel movement and no feed. Maybe CNC kitchen or Dr. Dflow. Dflow did a very large scale pellet printer. maybe this helps on the stringing retraction issue. micro optical type switch or nano like adjustable switch control to adjust layer height and retraction.. just ideas...
This is exactly what I need. I have been saving all my PLA trash from supports, failed prints etc, since I started 3d printing, waiting for this product.
It is "less" easier because more steps are involved in producing filament than just directly printing with pellets. However, nothing is holding you back to produce Filament with the Pellet Extruder, if that is what you want to do. The point is that you have full freedom of choice... 😀
Well done! A good enough print as shown in the video is insane. The possibility for community collection and reusing plastic is through the roof. There are a lot of nonprofits in Buffalo NY, that currently have 3d printers that would be interested in this project. Filament is the biggest cost right now which blocks us from legitimate local manufacturing. Since the injection molded parts are cheaper than just the filament, its hard for 3d printing to make sense This project could bring us closer to - buy nothing world I think a lot of 3d printing print farms are looking closely at this too! Could make their prices much more competitive and possible to break into more product categories, including more simple shapes.
Run this beside the printer with a 1.75mm nozzle. Run the output through a bowden tube to a standard direct-drive extruder. If you set the pellet extruder as a standard second extruder in "ditto" mode with the correct steps/mm, you'd be getting filament on demand for the "primary" extruder with no changes to the printer necessary. Heck, you could use it as a pure bowden extruder and have zero worries about the weight of the entire setup.
How is that different than an inferior filament extruder? What benefit would it bring? Filament extruder are slightly more complex as they typically require sensors to ensure the filament is always the correct size, without it the resulting filament would be inconsistent reducing quality or causing clogs. Why not buy a filament extruder to produce your own filament from cheap pellets?
@@hellothere6627 The difference is this would be done on-demand, and would be compensating for inconsistency instead of trying to make 100% consistent filament. This is akin to using sensors and processing power to compensate for low-cost parts the way Bambu has done successfully. Large, heavy, stable durable machines existed with mechanical perfection giving near perfect accurate printing. Bambu just measures the imperfections and compensates for them in real-time to generate results.
@@greenboy3d I would love to see it. The only limitation I initially see is that you would want to break up extrusions into small chunks, such that it's not making one measurement adjust for a very large movement. A post-processing script could break up any long extrusions. If a long extrusion was defined as a single extrusion that uses more than one (small) pellet's worth of material, it would theoretically be adjusting flow at the same resolution (or better) as pellet volume.
If the quality and cost benefit speak for themselves, I really really don't think supporting standard nozzles is something you should focus on. The moment this thing releases - if the major 3d printing channels "get" it and are able to reproduce the benefits you show (material choice, cost, flexibility) then it's a winner. But, if the end of those videos they say "but, I've not found the quality to be usable for the kinds of prints I do" its a loser. Speaking personally, my opinion would be swayed by this kind of thing. I want a third party to do strength tests and quality tests before I buy into a whole ecosystem. All this is to say - focus entirely on quality and reliability. Nozzle compatibility seems like a far far worry from that. Multiple melt zones, flow rate sensors, that seems like the next step, but you won't get adoption if the quality is under the acceptable threshold for many people today.
It is "less" easier to produce Filament because more steps are involved in in that process than just directly printing with pellets. However, nothing is holding you back to produce Filament with the Pellet Extruder, if that is what you want to do. The point is that you have full freedom of choice... 😀
I think this is the way to go especially for large printvolume printers since the printmechanism is bulkier and the material is way cheaper. Mach weiter so, 👍
One idea for making inconsistent scraps into consistent prints... You know how some people tried 3d printing airsoft 6mm bb's? Yeah, print those at 50% scale with the smallest of scraps to get semi consistent pellets of 3mm
An option to improve quality with recycled plastic would be to use a 1.75mm “nozzle” to constantly extrude something that resembles filament, then regrind it into pellets, as CNCKitchen showed with his Artme3D
@@greenboy3d maybe after a time, but I think you could counteract that by introducing some virgin plastic each regrind. I don’t have any experience with pellet extruders, I’m just going off stuff I’ve read online. Your project looks really cool though, I’d be really interested in trying it.
I would worry about off-gassing for using the various shake methods while the pellets are in the extruder. I thinnk a greater focus on post-painting and recycling already mostly off-gassed materials makes more sense.
I don't know much about it but from what I understood the print quality is caused by the pellets inconsistency, maybe it is fixable with a bigger heating nozzle, or maybe a semi-melting tube before it reaches the main nozzle. I am aware of the dripping problem it creates since there is nothing to hold the plastic from flowing down but maybe a custom nozzle could handle it? I loved the video btw
I think the best solution for the nozzle would be using a larger thread for the proprietary nozzle. Say, M10 or something. (Ideally with a taper lock at the top so molten plastic doesn’t enter the threads) This way, the system can use proprietary nozzles or standard M6x1 nozzles with an adapter bushing.
If the meltzone isn't needed anymore, you could use E3D-style nozzles (because the have a smaller head than the old MK8 nozzles) and shorten the threads a bit on the lathe. This way the meltzone decreases, but you can still use all the normal cheap nozzles as a quick replacement in a pinch.
Next printing with various materials and explaining where to buy . Adding cf an gf. Developing: injecting color on the fly by design. Auto drying the pellets on the fly
IIRC the pellets won't run themselves through the extruder at a perfectly constant rate. You'd have the same problem using filament with a non-constant diameter.
@@ski3091 so I suppose this type of extruder is not good for prints with tight tolerances or small details but is excellent for large structural objects that you would need a lot of?
Like others mentioned, the pellets or granules don't have a perfect constant flow rate. I have some ideas to solve this problem in the future, but for now it is what it is :)
High volume extrusion + large format(build volume)? Robotic arm or gantry For molds. Like CEAD. Thank you. Because then you can use carbon fiber and vacuum bag setting resin- infusion.
There is no "standard nozzle" anyway. Most of them are M6, true, but not as a standard. If we have to chop them off or recut threads then that's that. Even a custom order at mail-order-cnc isn't that expensive anymore. Quality is more important than a non-standard part which is easily manufactured.
A question that can't get out of my mind is what people would if they had a nozzle clog, which can happen on any 3D printer due to some particle blocking the nozzle hole. If the "special nozzle" is not easily available globally because only I am selling, then this might be a problem. Any ideas?
@@greenboy3d I don't understand. First of all, any metal nozzle can be cleaned with solvents depending on what clogged it. Acetone works for most polymers, some need something else. The nuclear solution is carbonizing the polymer by burning it off. What are the specific requirements of this special nozzle? Does it need to be especially short? How short are we talking? 2mm? 5mm? 10mm? If it's just chopping off most of the threads of a readily available nozzle then anyone with a saw, sandpaper and a vise can do it.
I honestly think I could live with the reduced print speed if it prints consistently or great. but even right now it could be used in a tool changer? I also have an idea: i'm developing a pressure sensor for hotends that can optimize the flow and calibrate on the go based on hotend pressure, you could use one and then do a closed loop servo style hotend!
This is really interesting. I wonder if you could make some GCODE that would use your pellet extruder to make spools too with a small addon. Seems like it would be a pretty handy thing to have around in general. Your screw style pellet extruder might actually be ideal for people's DIY filament recyclers, turning older printer parts into filament recyclers is of interest to a lot of people.
I think the nozzle should be optimal for the process however if you can make it so the nozzle is easily swapped so the user can choose, that might be best. Seems like this system would be great for prototyping where you may not be worried about print quality however I would question whether the print quality is good enough for dimensional accuracy.
@@greenboy3d I would like to see that you are financially rewarded for all this work; make sure you charge something for your design. I know the Chinese will copy it quickly probably but I hope you can be rewarded.
Love this project and can’t wait I have an old Ender i could dedicate for this awesome project! I think having a quick swap that can change to other machines.
The main problem I see is this: Filament prices are coming down. The cheapest PLA pellets I could find were about 10€ per Kg, the cheapest filament is 12€ per Kg. That means the filament is now so cheap that my biggest printer uses more money in power than filament.
![Hunxho - Ball Forever [Official Video]](http://i.ytimg.com/vi/nki9ZTF9Uls/mqdefault.jpg)
[Update October 13th]
Working very hard right now to organize the shipping of the Pellet Extruder and other tasks related to Greenboy3D like organizing a supply of pellets for everyone and a pellet extruder wiki, which is why I sometimes have trouble answering everyone immediately.
Most parts for the extruder have already arrived and have been processed & packed for shipping, however the pellet extruder screws are still in manufacturing since this is the most complicated and costly part of the pellet extruder. I believe that the first units will be shipped out by the end of October.
Due to the additional organizational tasks related to the supply of pellets, there is simply more work that needs to be done. This includes staying in contact with many companies selling different types of plastics & watching out for good offers, requesting samples, testing them, negotiating prices, organizing the transportation of tons of plastic pellets, and so on.
These tasks also take up some of my time, which would otherwise have been spent on the Wiki/Knowledge Base platform and the guide videos.
Just a couple days ago, for example, I ordered and paid for 8,600 kg of PLA pellets, which will likely be transported to my location from Slovenia next week.
2500 kg PETG pellets are already in my possession, but more of other types are needed like:
ABS/ASA/ABS-GF/ABS-GB
PET
TPU/TPE (various Shore levels)
PA GF/CF
Additionally soon there will be a new update about the ongoing progress (including the one about pellets) 🙂
If you want to get some additional pellets along with your order then that is possible however you need to consider that, while the pellet extruder will be shipped for free there will be additional shipping costs for pellets.
However I will keep you updated regarding the pellets.
The only thing I am not sure about is whether I should also get these extremely elastic pellets I showed in one of my videos, after doing more research it turned out these were not Shore 5A, but Shore 30 (with out the "A") which is significantly below Shore 0A meaning even more elastic and flexible than Shore 0A
The only problem is that these pellets are quite expensive compared to other plastics and could only be offered for prices like 12,99 USD/EUR per kg (way higher than the other plastic types like PLA, which will probably go for 3,5 USD/EUR)
Should you have more questions or concerns please feel free to contact me anytime - I am here to help.
Hi
My idea for coloring is using different color inkjet bottles they feeding directly from side your extruder simply using stepper motor pumps. So you can run the color(s) slow or higher speed, or in a mix of different colors into the extruder.
@madison66 hmm🤔
@@madison66 a similar thought occured to me also. Realistically, for any innovative 3d printing system to try and take on the giants like prusa, creality, or bambu labs, then it needs to have an answer to multi material or multi colour systems with a user experience at least half way as refined.
Automated side feeding of colourant into the feeder could be part of that.
@@madison66 theoretically you would only need red blue yellow black and white to make any color combination you could use two or more print heads to achieve full color 3d prints
@@muatok9904 same way like an inkjet printer does, so there's no need of rocket sience
I loved your research and the video - but I’ll be blatantly honest about the print quality: it’s not there yet. The Benchy has VERY inconsistent layer lines. On a commercial printer I would have returned it. The question is: are the inconsistencies generated by your pallet extruder? Or is it the mechanics of the printer that are suffering from a heavy print head? I would love to see it on a stiffer printer / maybe it gets good on a printer with a more rigid gantry, linear rails or wider gantry extrusion… more tension on the belts.
Please, don’t dismiss the importance of print quality - people are willing to pay more for better quality filament exactly because they want smoother prints.
Keep going the awesome research!
The issue is caused by the inconsistency that plastic pellets in general have. It's like having a Filament spool with fluctuation a diameter between 5% but because of this it costs also 5 to 10 times less.
This issue can be solved in the future for example with a pressure sensor in the extruder barrel or a small AI camera monitoring the extrusion rate and real-time-micro adjusting it, which is why pellet extruders will have in the future basically the same print quality as filament printers. But this this kind of technology still needs to be developed...
If I make enough sales, I will try to develop this technology with extrusion measurement myself, since it would be my dream to bring a huge innovation to the world of 3D Printing... Wish me LUCK🙂
@@greenboy3dyou could try an esp32-cam or 2 with a macro lens focused on the hotend, as a bonus you can detect blobs before they cause issues.
Imo another option might be double extrusion, with a pellet extruder extruding into a bowden tube and the filament being picked up by a regular hotend, that allows you to use light to measure the diameter, and to use a way lighter moving assembly.
(Also depending on print speed you could have a shell that's solid but the core still above 60C, which means more flow rate.)
Yet another option (imo the easiest to do currently) would be to use an idex system and do the outer wall on a regular .4mm nozzle, then fill in with the pellet extruder on a 1.2 or 1.6mm nozzle.
The great thing with that solution imo, is that you don't need good quality pellets, and can use whatever color recycled plastic in it and still have an uniform print.
And also the large diameter nozzle means that you don't need the heavy pellet extruder to move fast.
@@greenboy3d look into extrusion screw yours is too loose and no enough tight and maybe to short too
I have access to literal tons of both engineering and commodity resins for free that's thrown away from my workplace. I think you 100% need to pick a nozzle that's optimal for pellet extrusion more so than maintaining broad compatibility.
What place u work for?
Absolutely. This gives so many benefits overall, a small compromise is no problem.
Thank you for your feedback🙂
What would you like to see next?
1kg PETG cost here 6-9 usd per kg. That is so cheap, if it would be free, I would not print more. In the company we also put hugh amounts of PA6 and POM in the junk
Special nozzle all the way. If a standard platform for pellet extruders eventually comes out as a result too, it just means this will have set the ground for that standard! If there is demand, standardization will come, and people will prefer the performance gain and the eventual affordability.
My 2 cents on your question about "standard nozzles": not critical, especially if they are suboptimal for this extrusion technique. Most important is a pellet extruder with optimal heating & retraction results. So if that means a different nozzle type, so be it.
Fantastic work by the way ! 👍 Been saving a ton of PLA parts in the hope of recycling them into new prints. But nothing out there seems affordable in terms of time , money or reliability. Most require recycling into filament first, which feels like a dead end for consumers: too many issues, too time consuming, somewhat expensive.
Your approach that just requires careful grinding seems the most promising to me so far. And seems good enough for my prototyping workflow for functional parts: many iterations, no need for perfect surface quality.
Thanks for your 2 years of effort on this front !
I agree on this. CNC Kitchen even tried to build a high flow nozzle out of a regular brass nozzle and a copper insert when HF nozzles were a novelty. So already manufactured nozzles can be adapted "quick and cheap" with a round copper insert by just press-fitting I guess, and new manufactured nozzles just with one hole less! I am sure that with the visibility you are earning through your videos, you can achieve something like that pretty easily. However, I would let the nozzle compatibility window open.
Agreed - for PLA the time and effort isn't worth it - aesthetic prints like cosplay items, busts, toys etc. wouldn't have the look or shape consistency needed, and the functional parts may not have the tolerances needed - it could be great for printing support or infill perhaps but then that introduces a whole other series of challenges in slicing or IDEX integration. I would recommend this extruder for printing thermoplastic elastomers that are too soft to turn into filament, or perhaps make printed parts from materials other than standard and widely available filaments.
Thank you for your great input :)
A question that can't get out of my mind is what people would do if they had a nozzle clog, which can happen on any 3D printer due to some particle blocking the nozzle hole.
If the "special nozzle" is not easily available globally because only I am selling it, then this might be a problem. Any ideas?
@@greenboy3d depends on your goals, I guess. If it is to make money on nozzles then indeed it could become an impediment for buyers. If not, publishing the design specs will get a long way toward copycat nozzles appearing on aliexpress or ebay. With many biz models in between (partnering w 3D printing accessory distributors on each continent, selling through specialized online stores that have figured out worlwide buy & ship like CNC Kitchen's, etc).
BUT the nozzle is only a tiny concerning piece in my opinion. How about other proprietary or uncommon parts of the extruder and providing spare parts for repair ? Like the extruder screw, or its housing ? Screw extruders do NOT have a great reputation on the hobby scene. So, as a buyer I'd be less worried about the nozzle (I'll likely find a way to unclog it) and more about whether I'd be able to get spare parts if anything breaks down ?
Bambu Lab might be an example to ponder. The highly specialised parts in their printers initially turned many people off for fear of ending up with an expensive unrepairable dead paperweight. But their policy of selling most parts online at a reasonable price went a long way to assuage these fears. Maybe something can also be learned / copied from that ?
I love your attitude of just getting things done rather than getting stuck in making the prints perfectly clean 😃
And I love you comment ❤
What would you like to see next?
@@greenboy3d I'm quite interested in more details on shredding the plastic to make it workable, or is it really as simple as just throwing parts in a kitchen blender and sifting the result? I would also like to know more about your 2-year journey (what major hurdles did you encounter and how did you tackle them) to get at this point, and on how to get my hands on the result to be able to play with it.
@@Kurckie maybe another extruder just a little bit bigger where you can throw in bigger junks and then making consistent pellets not just dust...
0.0.o
Idea: try adding some new pellets to your shredded waste, somewhere between 10-40% of new pellets. This is normally done in recycling as it increases quality of the end product by a lot. And 90% of recycled material will still safe a lot of waste
Thank you for your interesting input 🙂
What would you like to see next?
I mean, 100% recycled material or 90% recycled material should safe the same amount of waste, since you are propably using more waste than you produce (if over 90% of your prints are waste, you are propably doing Something wrong)
Put me down as "don't care" about nozzle compatibility. Nozzles are cheap, they last a long time, and if a special extruder needs a special nozzle, I can understand that. As long as I can GET them, or I can modify standard ones to work. Are you thinking of tapering the inner profile? What would make a standard work better on a pellet extruder?
I'd think something that's internally threaded to eliminate the extra unneeded melt area.
A question that can't get out of my mind is what people would do if they had a nozzle clog, which can happen on any 3D printer due to some particle blocking the nozzle hole.
If the "special nozzle" is not easily available globally because only I am selling it, then this might be a problem. Any ideas?
@greenboy3d the idea to modify a stndard nozzle from @BrightBlueJim would be the easiest way in terms of accesability. Even when you break multiple ones during the process of modifying or if they clogged faster you can get a inexpensive replacement. The tapping tool could be a custom cadfile ready to cnc manufacture via different big manufactures, like JLC for example, same for special nozzles.
@@greenboy3di mean if you’re the only one selling it it could be a problem, but if you can solve the distrubution option it wouldnt be an issue imo. You could look into if modifying a nozzle thats readily avaliable is an option because then people could just get that nozzle and modify it themselves if buying one from you isnt an option.
@@greenboy3d I work in blow molding. We use stainless steel screens and place them after the extrusion screw. This catches contaminants and solves your clog. Your only option after that is solvents when removing your nozzle.
I am in absolute awe. You are ABSOLUTELY correct. There is a huge market that falls under the 'Good enough' category. It's not even bad print quality. I'd MUCH rather have a bunch of good enough prints made from all of my plastic waste. I will be ordering soon.
Thank you for your appreciation 🙂
If you have any sort of questions, feel free to contact me anytime. For example, through email or Instagram.
I also suggest you to join the Greenboy Discord Community Server where you will encounter more than 1300+ like-minded people that discuss daily all sorts of things about pellet 3d printing > discord.gg/UuFDbFj6FC
Honestly the majority of new printers have proprietary nozzles, so if an optimal nozzle can be close in price, i dont see it being an issue. Really interested in this for an option for my sunlu s8 with a bigger nozzle to print my failed prontd
A question that can't get out of my mind is what people would do if they had a nozzle clog, which can happen on any 3D printer due to some particle blocking the nozzle hole.
If the "special nozzle" is not easily available globally because only I am selling it, then this might be a problem. Any ideas?
@@greenboy3d Good point indeed. Humans on all the sides of the planet.
may be you go as some success company do - make a cheap option and a pricy one. So you will also not let your interest go away for some ideas.
Cool to see progress on this! Reg.nozzle: why not just make custom nozzles that use m6 threads? You maintain compatibility, and you can also get the characteristics you want
A question that can't get out of my mind is what people would do if they had a nozzle clog, which can happen on any 3D printer due to some particle blocking the nozzle hole.
If the "special nozzle" is not easily available globally because only I am selling it, then this might be a problem. Any ideas?
@@greenboy3d Make many or get a manufacturing partner. If you want this to catch on it needs to be accessible.
@greenboy3d well, it's not like you can buy normal nozzles at the local super market either, i think as long as you maintain compatibility with standard nozzles, it should not be an issue, otherwise you can sell them in packs of 10, that gives people alot of time to order a new set when they need it.
I suspect that the custom nozzle is going to be all about having as little fillament volume as possible to avoid oozing, right? In that case, it might be even easier to clear the nozzle, and it would make it even more a non issue :)
@@TurboSunShine there is a screen going somewhere at the end of the screw catching debris
this makes utter sense when recycling, you reduce one step and won't reheat the plastic again saving it from losing it physical properties
What would you like to see next? :)
From tests I've seen reheating doesn't really reduce properties by a significant amount, the main issue with recycling is contaminants.
@@satibel I think so is for metals and well, many other materials that melt at high temperatures, so i suppose it should be a similar story with plastic since it's so similar to those just in a much weaker and lighter (and cheaper) form
@@ERIK-457 yeah for metals the main issue is that most of them are alloys and you can't precisely control the quantity of each metal, so you have higher variance on properties with recycled metal compared to metal that is made from known sources.
This channel ought to have 500k+ subscribers. Good job at pawing the way for pallet extrusion on consumer grade printers.
Thank you for your kind words ❤
What would you like to see next?
I think a possibly better option is to look at making filament yourself. If you can make filament then you can use any off the shelf printers or parts. Splitting it into separate steps is likely better.
That also removes the print quality concern.
Another option may be to put the pellet extruder on top of the printer and have it create filament that goes straight into a Bowden tube down to the hot end, although this would likely not be as good as making the filament separately.
Supporting standard nozzles will bring more people in, so in the short term that's probably the better plan, but the next step will be to develop custom nozzles and setups that work better.
Thank you for your great input 🙂
What would you like to see next?
It's better to have something that works. If it doesn't work it doesn't matter if it uses a standard nozzle.
In this case, in terms of speed, I think a bed-slinger might actually perform better than a CoreXY. On a CoreXY the heavy toolhead would affect the resonances of both axis, which would limit accelerations. On a bed-slinger the heavy Y axis performs worse than the X, bottlenecking the X axis speeds. But in your case, both axis are heavy, so they don't limit each other, and because they're separate, it's probably going to be easier to deal with their respective resonances. But that's just my theory.
True :)
By the way, what would you like to see next?
For recycling it's typically done by mixing the old plastic into a new batch.
You could try how well the system works at different ratios of old and new.
Ideally ofc you would want to shred it to similar granule size, but the hopper and screw system should be able to handle varying sizes.
Would like to see some tests on this. Recycling old plastic inhouse can be a huge saving overall.
For the nozzle question, I think you should aim to have optimal nozzle for the printing and not be chained by universality.
If the system becomes widely used then it becomes the new standard :)
very interesting input 🙂
Thank you
I used to work at Tektronix, and there was a shade of color the used a lot on plastic molded parts that we called "Tek blue". I learned one day that they made this by mixing medium blue, white, and black RIT fabric dye (the dry powdered version) to get this color. Of course, those were injection molded parts, so I don't know how that would work here.
Thank you for your very interesting input 🙂
What would you like to see next?
I would agree with most others that a more efficient nozzle would be preferred over a standard one.
A question that can't get out of my mind is what people would if they had a nozzle clog, which can happen on any 3D printer due to some particle blocking the nozzle hole.
If the "special nozzle" is not easily available globally because only I am selling, then this might be a problem. Any ideas?
Nozzle clog can be solved by buying another one. Profit.
Sell a 3D design of the nozzle for a substantial amount, or just publish it for free for the street cred
I calculated 112g/h = 38.3 mm^3/s, am I correct? Calculations below:
Assuming PLA density is 1.24g/cm^3:
112g * 1.24 = 138cm^3/h
138 / 60 = 2.3cm^3/min
2.3 / 60 = 0.0383cm^3/s = 38.3mm^3/s flow
It would be cool to see you perform the orcaslicer max flowrate test on different nozzle sizes. It makes it easier to see the actual limits of flowrate.
A bit of an overestimate on flow rate unfortunately - if the PLA is more than a gram per cm^3 then 112 g should represent less than 112 cm^3. I think the calculation should be as follows:
112 g/1.24 g/cm^3 = 90.3 cm^3
90.3 / 60 = 1.50 cm^3/min
1.5 / 60 = 0.025 cm^3/s = 25 mm^3/s
Still better than many standard nozzles, but not matching top of the line high flow. I think beyond this, the cooling rates will be a bigger limits but with klipper input shaping the weight of the extruder may be much less of an issue than with previous firmware. Still a very impressive project in general!
I'll look into the "orcaslicer max flowrate test", might test it this way next time.
Thank you for your very interesting input 🙂
What would you like to see next?
For the nozzle, maybe go with a CHT nozzle, since it has three channels in which the molten plastic flows through, meaning it might have more consistency in the heating and flow. As for the small discrepencies in the melting process, maybe add some small notches on the side of the extrusion screw so that any trapped air could be potentially allowed to bubble out or maybe as a way to control a bit of pressure.
As for dying your batch, could try going with alcohol dyes and do it in the same format of mixing it in an enclosed jar/container.
Thank you for your very interesting input 🙂
What would you like to see next?
I like the idea, however I think print quality is a vital thing to consider, especially from those of us who make a living off of making prints. Especially on dimensional accuracy, print quality is something needed.
The retraction explanation makes perfect sense when you consider that reversing the conveyer screw in a pellet extruder is functionally doing the same thing as a "retract" move on a filament extruder; namely, they both are reducing the pressure inside the melt chamber. Which is the real goal of a "retract" move anyways.
Some of the problems you're running into are also issues in small extruders.
I.e. if the extruder screw is running at 60 rpm, expect an uneven "squirt" of resin every 1 second out the end of the nozzle/die.This is why metering pumps are often used at the end of an extruder (to tightly monitor flow).
The 3d filament printer brilliantly solves this issue inexpensively by using filament.
The filament feeding system is acting as the gear pump.
Another issue will be safety. If you build up pressure behind molten plastic that has frozen off at the nozzle, watch out.
Randcastle has made a great miniature extruder for decades to learn some of the inherent issues.
Thank you for your interesting input :)
What would you like to see next?
You’re designing an entirely new hotend. A custom nozzle is not going to be the deciding factor in whether people choose your design or not.
Very good video and so happy to see your progress. I have 40 years of high production high precision injection mold making and molding experience. You are dead on in the process you outlined and are mastering the rheology of thermoplastic flow . The only difference in the outcome is that you are not able to get the density of an infection molded part because you do not have a packing phase in 3d printing. As a result you may have aligned molecules and a grain like wood in printing that has some advantages and disadvantages. But with carbon fiber we may be able to produce a similar strength . Not sure yet all the doors you are going to unlock but accolades to you this this an amazing project. But I can tell you that the injection molding press industry has already figured out how to do a lot of what your doing they just have not figured out how to 3d an extruder for 3d printing
Thank you for the Feedback :)
Yea, there a definitely some additional challenges involved making a "mini" extruder for 3d printing
About the nozzles: could one not just cut away most of a standard nozzle with a hacksaw, leaving just enough threading to screw it in? This would eliminate the entire part where plastic is melted and leave just the final conical shrinky bit needed for printing. Given that this extruder isn't likely to be acquired by anyone non-technical, I think having "SAW ALL THE THINGS" as part of the build process isn't too demanding a step.
you are right on your conclusion but yet it would still make things more complicated for everyone...
because if you would do it this way and optimize the pellet extruder for a shorter threat than people without this modification couldn't properly use the pellet extruder which is a problem :(
@@greenboy3d I was thinking more like - have a standard unmodified plan for a normal nozzle that everybody can follow, but mention that this is a possibility so that anyone willing can saw the needless bit off and enjoy a little extra effectiveness.
@@fallingwater ok
Honestly, if the nozzle is open source, it wont matter if you make it specific or universal, people can come out with their own to help improve etc.
This is an awesome project and you're doing the community AND the environment a huge favor. Keep up the good work 👍💪
I’m all for recycling and reducing waste, also I like saving money, so I love it!
Most prints just don't need good quality. Stuff like hooks, holders, storage containers and so on work just the same even with a +-1mm tolerance, so the reduced surface finish is no problem.
Besides printing metals this is by far the best development in the 3D printing space since years. Kudos to you, keep up the good work.
this should be watched by all big 3d companies. Whoever makes this reliable and with quality will win everything.
I once read an article about an Indian university student who couldn't afford a 3D printer and so he designed and built his own, the only things he bought were the arduino chip to build his control board and the steppers, he was also having trouble affording filament and so he also designed and built a pellet extruder as he could get the pellets for pennies.
I have a few old bedslingers that I would love to convert into pellet printers!
Let's not forget that pellet 3D printing is also better for the environment. You're cutting out the entire step of shipping to the filament producer, extruding filament, spooling it, and packing it. I'm not a type of "oh my god the environment is everything and we have to drop everything to protect it"-person, but it's still nice when something is more efficient.
I do wonder, what's causing the visible layer lines? Is it the relatively inconsistent extrusion rate?
660g is world record for sure, you can achive less with robot joint motors but they are expensive (BLDC+Harmonics+step/dir input)
Thank you for your very interesting input 🙂
What would you like to see next?
I have collected around 8-9 kg of PLA waste in the course of 14 months and building an extruder to use them to print functional parts will be a much better use for them instead of giving them for recycling. I am very excited for the design files to come out and know where I can source the screw from so that I can recycle my waste into useful parts!
Thank you for your warming words ❤
What would you like to see next?
For recycling 3D prints, CNC Kitchen got the best results by repelletizing them. He was making filament but the basics were:
1.) Shred
2.) Sift
3.) Repeat till consistent
4.) Form filament
5.) Pelletize this first batch
6.) Form final filament
Clearly you can skip 6 and probably combine 4 and 5 to be a single operation.
Nice. For recycled plastic the bulk density will be different. Building a small pelletizer extruder would be more consistent. I worked as a tool maker in an extrusion plant for 10 years. There are ways to make it more consistent but the complexity goes up.
Thank you for your very interesting input 🙂
What would you like to see next?
Please note that if you have quality pellets (clean) you can produce filament on professional liine for 4-5 eur/kg. For instance TPU 65 shore D costs 3,8-4,8 euro kg now (Resinex, Ravathane 130 D65 NATURAL), regranulate 2 eur. Total is 6 euro kg while in store you pay 40 euro.
Love your work keep doing a good job improving our 3D printing community I plan on using this design on a personal project in the future.
Thank you for your warming words ❤
What would you like to see next?
spray painting raises cost and time quite a bit.
I have a question: why does using pellets reduce print quality?
I recommend you to watch my First video "Pellet vs Filament 3D Printing" ruclips.net/video/AS898H9F04s/видео.html
I explain the reason for this in it
I just filled the form 😌
In short, for me it would be all about reducing waste from prototype prints: I would shred them and re-use the plastic.
Most of the time, quality doesn't matter that much when prototyping iterations especially if you add some more tolerances if needed.
Thank you for your input 🙂
What would you like to see next?
@@greenboy3d I'd like to see how the shredding process could be optimized, maybe with finer particles, filters and multiple iterations to reduce variance in extrusion.
I'm guessing that pellets are a convenient form factor but not made with tight tolerances. And maybe it's possible to increase the print quality by shredding everything, including pellets themselves into finer particles with tighter tolerances.
You should check out the quantum delta 3d printer by Kilian Gosewisch, instead of moving the printhead it moves the entire bed on all axis and the extruder stays stationary, that way you can eliminate the size and weight constraints of your extruder.
Nozzle compatibility across filament printers makes sense, but it does not have to be that way for a pellet extruder. It makes more sense for the pellet extruder that uses a different extrusion method to have a different nozzle. Then from there, hopefully all future pellet extruders would have the same nozzle compatibility.
Maybe if you just keep the same threading, it can be up to the user to decide if they want to use an easily available filament nozzle, or a pellet nozzle.
You've got a point here :)
Any wishes for the next video?
I feel that its worth mentioning that spraypaint isnt a great thing to be melting down, depending on the paint it could be very toxic.
Additionally, when mixing pigments into plastics the formulation is tricky as the pigments can easily ruin your material properties...
Yeah also pre pigmented pellets exist that aren't that expensive.
I agree with him... It's a great idea to try to recycle the filaments residue, even if the prints doesn't get perfect
Great video! really hope this gets adopted in a larger scale than tinkerers.
Thank you for your kind words ❤
What would you like to see next?
PLA pellets can only reliably sourced at $10-$15/kg. The cheapest PLA filament is $15/kg… last I checked. 0-30% cost savings was not 300% (missinformation factor 10x to infinity)
You just don't know where to buy pellets.
@@greenboy3d so where do you buy pellets?
@@pyalot going to make a detailed video about that in the future
My personal opinnion on the nozzle type is to stick with the current, widely available nozzles and once its widely adopted, offer specially designed nozzles for improved print quality.
I would preffer that because of 2 reasons:
- It would get a lot more people interested if they can retrofit their printers
- They would have as may interchangeable parts as possible
Moving on to the specialized nozzle:
- It would offer to the people who initially are focused on print quality and not function a reason to recyle and be more sustainable
- The hobyists would also benefit from having something to upgrade to
This is an awesome idea. It would be really nice to be able to actually make our own spools from pellets. Im sure you've already posted a video on why you didn't do that here.
I can try that out, should I? :D
Hey, this is great! I was looking for a convenient pellet extruder. I currently recycle old plastic and create my own filament, but that is much more work than directly printing the shredded material.
this is what big filament doesnt want you to know
😶
Nozzle compatibility isn't important if it comes at a cost of performance.
What would ultimately be nice is to design it to function with a breakout board, and be compatible with toolswap plates such as the Hermit Crab or WhamBam Mutant for Ender 3, or be able to easily mount it to a toolchanging printer. That's a much nicer quality of life, to be able to quickly swap between a pellet extruder and filament extruder.
What would you say are the top 5 most popular toolswap plates?
I dont like too much all the "green" theme, but i like the idea of self-sufficiency, i might use recycled plastic for low quality prototypes and only use top quality expensive filament for things that i want to be more permanent and of higher quality, i am a fanatic of quality by the way
Neat video, I love the amount of work folks put into FDM printers just to get a new way to print or not waste materials. I am not sure I would go through the trouble of doing all this, just to save a few bucks in materials, but I feel the pain of wasted materials as well. Resin printers create so much waste with support materials only being used and tossed. I love that you took and Ender 3V2 and did this, those things are like Honda Civics, cheap and easy to mod.
Love to see that project continuing... any estimation on when this will be available for testing? Would love to adapt one of those to my voron-toolchanger 😁
Probably in the next couple of weeks
What would you like to see next? :)
Thank you for answering my question in your previous video about retraction!
no problem :)
Be I have a flash-forge adventure 5 m pro printer . It is a big improvement but still has some things that can still improve on
It’s the best high speed for the price but it has triggered lower printer prices from the competitors
Try a little brass funnel that will fit inside the nozzle. Or match up.
Good idea :)
I love what you got here. The problem is layer lines. I would like to see this mod on something really good with that. If it's your system then open-source it to see how you can compress it. Current superchargers come to mind with compression.
Personally, I do not enjoy tinkering... that's why I went with a Bambu. However, this seems promising for prototyping. Keep pushing the envelope. That's the only way progress is made.
👌
I think significantly better print quality is more important than full nozzle compatibility. However it would be great if you could maintain some level of compatibility, for example by keeping the M6 thread. Then the pellet extruder would work best with the custom nozzles but still function ok-ish with a common nozzle. Some compatibility would prevent complete downtime and allow the early adopters more ways to experiment - an easy one would be shortening common nozzles as from what you're saying a stubbier nozzle is closer to ideal.
method 4 was genius. this project is fire, not something i would ever have use for but so incredible
Could you technically print pellets with the shredded prints so the size is consistent? As for the standard nozzle question, I'd love to see the difference in print quality with a non standard nozzle
Depends on how consistent the shredded prints are. If you have for example cylindrical shaped shredded plastic, and you plan to mix it with round shaped pellets, then you might get even more inconsistency... But yes you can...
pellet extruder is number one priority, but its great that you thinking of an options
I really like what you are trying to do. Even if the up front cost is most likely going to be more, the saving on the filament is huge. Making custom colors is interesting too. Are you using regular spray paint?
Yes, regular acrylic based spray paint. But like I mention there are multiple ways and spray paint types...
By the way, what would you like to see next?
SAW VIDEO 1 AND NOW THIS!! Thank You! Can a pressure lift switch be added between the nozzle head assembly and the actual carrier mount guide rail? maybe this would be fore a different design. say 0.6 nozzle layer eight travel is 1.2. the stepper feed goes till the head is lifted ill the trigger head lift switch is tripped or lifted to that height, then the travel steppers are engaged to conduct gcode shape. maybe be do a reverse stepper feed code for retraction. Head should relax and lift switch switch off for next travel movement and no feed. Maybe CNC kitchen or Dr. Dflow. Dflow did a very large scale pellet printer. maybe this helps on the stringing retraction issue. micro optical type switch or nano like adjustable switch control to adjust layer height and retraction.. just ideas...
Thank you for your very interesting input 🙂
What would you like to see next?
This is exactly what I need. I have been saving all my PLA trash from supports, failed prints etc, since I started 3d printing, waiting for this product.
Working on it :)
Maybe it is easier to use pellets to make your own filaments for the 3D printer?
It is "less" easier because more steps are involved in producing filament than just directly printing with pellets.
However, nothing is holding you back to produce Filament with the Pellet Extruder, if that is what you want to do. The point is that you have full freedom of choice... 😀
This is really impressive! You just revolutionized the 3d printing industry on your own 😮
Thank you for your warming words ❤
What would you like to see next?
Well done! A good enough print as shown in the video is insane. The possibility for community collection and reusing plastic is through the roof.
There are a lot of nonprofits in Buffalo NY, that currently have 3d printers that would be interested in this project. Filament is the biggest cost right now which blocks us from legitimate local manufacturing. Since the injection molded parts are cheaper than just the filament, its hard for 3d printing to make sense
This project could bring us closer to - buy nothing world
I think a lot of 3d printing print farms are looking closely at this too! Could make their prices much more competitive and possible to break into more product categories, including more simple shapes.
Make an adapter for your propriety "nozzle" that has multiple SKUs for every nozzle thread type. Boom done. Really cool idea man.
Run this beside the printer with a 1.75mm nozzle. Run the output through a bowden tube to a standard direct-drive extruder. If you set the pellet extruder as a standard second extruder in "ditto" mode with the correct steps/mm, you'd be getting filament on demand for the "primary" extruder with no changes to the printer necessary. Heck, you could use it as a pure bowden extruder and have zero worries about the weight of the entire setup.
Interesting idea :D
Would you like me to try that out in of my next videos?
@@greenboy3d Yes please! Great idea!
How is that different than an inferior filament extruder? What benefit would it bring?
Filament extruder are slightly more complex as they typically require sensors to ensure the filament is always the correct size, without it the resulting filament would be inconsistent reducing quality or causing clogs.
Why not buy a filament extruder to produce your own filament from cheap pellets?
@@hellothere6627 The difference is this would be done on-demand, and would be compensating for inconsistency instead of trying to make 100% consistent filament.
This is akin to using sensors and processing power to compensate for low-cost parts the way Bambu has done successfully. Large, heavy, stable durable machines existed with mechanical perfection giving near perfect accurate printing. Bambu just measures the imperfections and compensates for them in real-time to generate results.
@@greenboy3d I would love to see it. The only limitation I initially see is that you would want to break up extrusions into small chunks, such that it's not making one measurement adjust for a very large movement. A post-processing script could break up any long extrusions. If a long extrusion was defined as a single extrusion that uses more than one (small) pellet's worth of material, it would theoretically be adjusting flow at the same resolution (or better) as pellet volume.
If the quality and cost benefit speak for themselves, I really really don't think supporting standard nozzles is something you should focus on.
The moment this thing releases - if the major 3d printing channels "get" it and are able to reproduce the benefits you show (material choice, cost, flexibility) then it's a winner. But, if the end of those videos they say "but, I've not found the quality to be usable for the kinds of prints I do" its a loser.
Speaking personally, my opinion would be swayed by this kind of thing. I want a third party to do strength tests and quality tests before I buy into a whole ecosystem.
All this is to say - focus entirely on quality and reliability. Nozzle compatibility seems like a far far worry from that. Multiple melt zones, flow rate sensors, that seems like the next step, but you won't get adoption if the quality is under the acceptable threshold for many people today.
Thank you for your very interesting input 🙂
Any idea for flow rate sensors?
@@greenboy3d Perhaps temperature loss or pressure on the nozzle could be a proxy for flow rate measurement?
Would the cost of making a filament extruder using pellets and recycled prints give you the best of both worlds?
It is "less" easier to produce Filament because more steps are involved in in that process than just directly printing with pellets.
However, nothing is holding you back to produce Filament with the Pellet Extruder, if that is what you want to do. The point is that you have full freedom of choice... 😀
I think this is the way to go especially for large printvolume printers since the printmechanism is bulkier and the material is way cheaper. Mach weiter so, 👍
Danke :)
Was würdest du gerne als nächstes sehen?
Great job on this video! From the production to the information. Fantastic.
Thank you for your warming words ❤
What would you like to see next?
One idea for making inconsistent scraps into consistent prints... You know how some people tried 3d printing airsoft 6mm bb's? Yeah, print those at 50% scale with the smallest of scraps to get semi consistent pellets of 3mm
An option to improve quality with recycled plastic would be to use a 1.75mm “nozzle” to constantly extrude something that resembles filament, then regrind it into pellets, as CNCKitchen showed with his Artme3D
But wouldn't this worsen the material properties through extra thermal degradation?
@@greenboy3d maybe after a time, but I think you could counteract that by introducing some virgin plastic each regrind. I don’t have any experience with pellet extruders, I’m just going off stuff I’ve read online. Your project looks really cool though, I’d be really interested in trying it.
One other neat benifit you might be able to exploit. Because you aren't bound by fliment sizes, you could scale this up to crazy large flow rates.
That is true :)
I would worry about off-gassing for using the various shake methods while the pellets are in the extruder. I thinnk a greater focus on post-painting and recycling already mostly off-gassed materials makes more sense.
I don't know much about it but from what I understood the print quality is caused by the pellets inconsistency, maybe it is fixable with a bigger heating nozzle, or maybe a semi-melting tube before it reaches the main nozzle. I am aware of the dripping problem it creates since there is nothing to hold the plastic from flowing down but maybe a custom nozzle could handle it? I loved the video btw
I think the best solution for the nozzle would be using a larger thread for the proprietary nozzle. Say, M10 or something. (Ideally with a taper lock at the top so molten plastic doesn’t enter the threads) This way, the system can use proprietary nozzles or standard M6x1 nozzles with an adapter bushing.
Would melting spray painted pellets in your extruder not eventually gum up your extruder?
If the meltzone isn't needed anymore, you could use E3D-style nozzles (because the have a smaller head than the old MK8 nozzles) and shorten the threads a bit on the lathe. This way the meltzone decreases, but you can still use all the normal cheap nozzles as a quick replacement in a pinch.
Hmm.. That a good idea :)
Next printing with various materials and explaining where to buy .
Adding cf an gf.
Developing:
injecting color on the fly by design.
Auto drying the pellets on the fly
Whats the cause of the poor print quality? Is it just a lack of dialing in the settings? Or is there a mechanical reason causing it?
IIRC the pellets won't run themselves through the extruder at a perfectly constant rate. You'd have the same problem using filament with a non-constant diameter.
@@ski3091 so I suppose this type of extruder is not good for prints with tight tolerances or small details but is excellent for large structural objects that you would need a lot of?
@@ski3091 gotcha, thanks for the reply
@@ipodtouchiscoollol Sorry, I don't know enough about the process to answer that
Like others mentioned, the pellets or granules don't have a perfect constant flow rate. I have some ideas to solve this problem in the future, but for now it is what it is :)
High volume extrusion + large format(build volume)? Robotic arm or gantry For molds. Like CEAD. Thank you. Because then you can use carbon fiber and vacuum bag setting resin- infusion.
What would you like to see next? :)
while i get that the point is not print quality, it would still be nice (assuming 3mm pellets) to quantify extrusion consistency
You can get quite good extrusion consistency with some pellets that have the right size and shape...
But What would you like to see next? :)
That's really well miniaturized indeed, but i would still make filament first ...
Also for recycling?
There is no "standard nozzle" anyway. Most of them are M6, true, but not as a standard. If we have to chop them off or recut threads then that's that. Even a custom order at mail-order-cnc isn't that expensive anymore. Quality is more important than a non-standard part which is easily manufactured.
A question that can't get out of my mind is what people would if they had a nozzle clog, which can happen on any 3D printer due to some particle blocking the nozzle hole.
If the "special nozzle" is not easily available globally because only I am selling, then this might be a problem. Any ideas?
@@greenboy3d I don't understand. First of all, any metal nozzle can be cleaned with solvents depending on what clogged it. Acetone works for most polymers, some need something else. The nuclear solution is carbonizing the polymer by burning it off.
What are the specific requirements of this special nozzle? Does it need to be especially short? How short are we talking? 2mm? 5mm? 10mm? If it's just chopping off most of the threads of a readily available nozzle then anyone with a saw, sandpaper and a vise can do it.
I am still waiting for a chocolate 🍫 demo 😅
Don't worry it will come :)
I honestly think I could live with the reduced print speed if it prints consistently or great. but even right now it could be used in a tool changer?
I also have an idea: i'm developing a pressure sensor for hotends that can optimize the flow and calibrate on the go based on hotend pressure, you could use one and then do a closed loop servo style hotend!
Can you write me an email regarding your presure sensor please to kristian@greenboy3d.com :)
I am interessted and also willing to pay for it
@@greenboy3d Sure!
You won me at the 8:41 . Now I want a 3d printer just to be a brat.
😀
Makes me want to build an extruder. Then just buy pellets and colors. Would be awesome
What would you like to see next?
This is really interesting. I wonder if you could make some GCODE that would use your pellet extruder to make spools too with a small addon. Seems like it would be a pretty handy thing to have around in general. Your screw style pellet extruder might actually be ideal for people's DIY filament recyclers, turning older printer parts into filament recyclers is of interest to a lot of people.
I think the nozzle should be optimal for the process however if you can make it so the nozzle is easily swapped so the user can choose, that might be best. Seems like this system would be great for prototyping where you may not be worried about print quality however I would question whether the print quality is good enough for dimensional accuracy.
You can you make it work with normal nozzles and start making a custom nozzle after the release of the extruder
Thank you for your very interesting input 🙂
What would you like to see next?
Should get a Nobel Prize for this.
Thank you for your warming words ❤
What would you like to see next?
@@greenboy3d I would like to see that you are financially rewarded for all this work; make sure you charge something for your design. I know the Chinese will copy it quickly probably but I hope you can be rewarded.
Love this project and can’t wait I have an old Ender i could dedicate for this awesome project! I think having a quick swap that can change to other machines.
Thank you for your kind words ❤
What would you like to see next?
I recently got a Magneto X printer, and your work seems tailored to using it as a test bed, since the linear motor can move some serious mass quickly.
The main problem I see is this: Filament prices are coming down. The cheapest PLA pellets I could find were about 10€ per Kg, the cheapest filament is 12€ per Kg. That means the filament is now so cheap that my biggest printer uses more money in power than filament.