Homemade Pellet Extruder for 3d Printer

This is listed in one of the challenging projects I have ever made. During my normal build I sometimes failed in achieving the final result but in the end, accomplished that but this is the case in which I don't know I failed dozens of times and I think this is what happens when someone starts inventing new. I have been working on this project for the past one and half months and at the start, I thought it would not be a difficult project but it made my life hell during this period. I don't even remember exactly how many times I failed but I don't want to get beat up by those obstacles. Now the origin of this project starts back one year ago when I built 3 in 1 CNC. At that time my brother said what if we made an extruder that was capable of using waste plastic and hereafter one year I am able to make a pellet extruder. Now I don't have enough time to play much with this tool because I am busy supervising construction work but in the upcoming week there are definitely some videos coming out onto the channel page, so don't forget to watch them. Another main reason to build this project is to directly use the old failed printed parts to reuse them. Now one thing I have to mention is that this is definitely a difficult project to make. I also wanted to use the waste plastic bottles for my 3d prints to make planters and other useful items. One thing I definitely wanted to mention is that if you have the right tools like a lathe and milling machine the execution will go a lot smoother and easier. During this build, I always find that lathe would be the essential machine to make these kinds of projects. There are so many phases where I can utilize lathe to avoid multiple parts construction and make them with one single body. In that way, I think I am able to reduce the weight of the extruder to some extent. Now as far as the nozzle is concerned I didn't have the right tools to make a nozzle that's why the size of the hole I able to drill is 1.5 mm which is definitely thick but if you wanted to make planters and other decorative items in which layer height doesn't matter it won't be a big issue. At first, I made a design that I also machined onto a piece of brass but in the end, I have to scrap it and start over from zero.

The main reason for the failure is that I underestimated the pellets and the amount of force they required to extrude. This Instructables didn't have any design because it didn't go the way I was going with design at first and then I just went without any design and failed a few more times and then finally it's done. The main motive is also to tell you the points you need to consider while designing a pellet extruder for your 3d printer. Now I definitely did a couple of tests and found that with the material having a smaller granule size like 1.5-2 mm size there would be no problem in attaining the speed of 50 mm/sec but with the thicker material like 3-4 mm I can't able to go more than 24 mm/sec. In the vase mode with the same bigger material, I am able to manage up to a speed of 30 mm/sec. Now definitely many factors still take place while extrusion like material temp, the temperature of hotend, motor torque, etc. But what I found is that you can't achieve fast speed without adding much weight. Now if I increase the hotend length so that material starts much longer in the heater block and the Design is also a converging type to feed the material then it will definitely improve a bit more. So overall said all the failure which I faced in building this pellet extruder I am going to share that with you. Now if I start making a video over this project including the failure then without a doubt it will definitely cross more than an hour and it also increases the editing time. So I put all that information in these Instructables so that if anyone is interested in building one there are many points he/she needs to consider. The overall weight of this extruder is 1860 gm. It's definitely on the heavier side but my 3 in 1 CNC didn't face any issue of weight during the prints. I knew there are many experts sitting over here so I strongly urge you to share your feedback and tips regarding this extruder. Now without further wasting any more time let's build this bad boy.

For projects like these, I think you need to have tools as many as you can for a machine shop. My design was good but the tools I am using are not that great. If you have a better tool than this then I think you can make a much better version than this. I want you to make your version by considering few points mentioned in this Instructables.

1. NEMA 17 geared stepper Motor
2. Spider coupling 10 mm to 8 mm
3. Aluminum pipe 1" dia
4. Copper Block 32X32X40 mm ( but recommend to chose a bit bigger in length)
5. Aluminum strip 8mm and 3 mm thick 32 mm wide for making the heat sink
6. M10 stainless steel rod with M10 washer having OD 23mm. for auger screw( it would be good if you go with a wood auger drill bit)
7. Allen Bolt M3 15mm (3pcs for heater block )
8. M3 20 mm long for the fan (8pcs)
9. M4 25 mm Long (4pcs for the Connecting motor bracket to pellet feeder
10. M3 20 mm long for connecting motor to the motor frame
11. M5 12mm 2 pc for connecting heater block to the pellet feeder
12. M10 10 mm long bolt for nozzle
13. 100k thermistor
14. 50W ceramic heater 2 pc. (i use 24v)

.

If you watch the video then you might think that there are not so many parts in this build but even though It had few parts, those individual parts took a lot more time to finally prepare them for final use with the tools I have. If I start telling you from the start then this blog will be like mashed potatoes because there are so many failures and if any area you find any part doesn't show how I made and already present then assume it to be already made in failed attempts. During this project if I made something then I proceed to test it but I didn't record that phase because most of them are failures, so I am going to start as I made the video so that it's much easier for me to write this post and also easier for you to understand as well.

As I told you in the intro part I didn't make any Design for this project but for your sake of understanding, I am posting a fusion file that might help you to understand. The file does not contain all the elements because some of them like the auger screw are difficult for me to build so I am not going to include that in this design. I am not going to share the failed design as there would be no use in using them. So let's start this project.

This is the bottom part of this extruder assembly and also to be considered an important part of this tool. At first instance, it might look very simple in design but to execute it you really need to pay attention because without proper markings you can't make it right.

During this project, I made 5 different heater blocks and all had something different either in dimensions or in the drilling pattern or the heater placement and many other things too. The final piece has dimensions 32X32X40 mm. Now there are so many reasons why I go to this one and now I think it would be better to make it a bit more longer in length.

In the beginning, I thought to melt such a big mass I needed multiple heaters and to go with a round heater block but as soon as I completed drilling the holes and the heater cartridges were becoming tiny loose that started to create the first issue. The first block is 1 inch round stock having a length around 50 mm. The problem with the round stock is that there isn't enough material to drill and tap the holes for making a squeezing mechanism for the cartridge and if those heaters weren't able to make a good connection with the walls then it won't be able to transfer heat much more quickly when it needed to melt the upcoming plastic. So because of that, I have to discard that design and then I chose the aluminum, and instead of the round I go with square one having 4 heater cartridges but due to the wobbling issue of my drill press, I drilled a slightly larger hole which ended up discarded. Then in the next copper block, I made so many different strategies which definitely made after testing the heater block that it starts to get leaked out because I drilled so many holes to hold the stainless-steel shaft and due to less precision I drilled holes that went all the way inside the main chamber and as soon as material enters into the chamber and screw squeeze it. It starts to come out from different holes. So I have to try again and use brass again for the hotend. This one works but the entire setup has to be banished because the pellets are bigger and not able to fit up in my first design. So the fifth one which is the final is the successful version of the heater block which works awesome.

.During the manufacturing of the heater block the first thing you need to remember is how you are going to arrange all the components. Once it's ready I made a mark on one of the faces. Now few things matter here, the diameter of the heat breaker which is also going to feed the material into the heater block, the size of the ceramic heater cartridge, and placement of the thermistor. In my case, the OD of the pipe is 18.75mm so I made a marking according to that. First I colored the face and then marked the center. This is going to be the main chamber of the Heater block. For the cartridge layout, you need to keep the heater as close to the main chamber so that heat transfer doesn't take much time. The material also plays an important role in this. Usually, materials like copper and aluminum have higher thermal conductivity but brass have half of these materials, but the disadvantage of aluminum and copper is that they are not easy to work with and often clog the bits and mills. Slow feeding and proper cooling are absolutely mandatory for these soft metals. For the ceramic heaters, I chose two opposite corners. Since I have to press the outer wall to squeeze it so that they can grab the cartridge tightly and be able to transfer the heat much faster. Once this is done I mark the center punch and start the drilling work.

For drilling the holes I started from the center one which is the main chamber. I use a 3mm drill to drill the hole all the way. My drill doesn't make a straight hole, but at the other end, it deviates 2mm. After that, I drilled two holes for the ceramic heater. For the heater, you don't need to drill a complete hole. Since our heater cartridges are 20mm long and block length is 40 mm therefore a 30 mm hole is required in this case. Having a higher thermal conductivity I found that heaters have no issue in heating the material to the required state. Also, the heater wires are crimped that's why it's important to secure them also and for that placing them 10 mm down from the top is a good location for this. Usually, ceramic heaters come up in a size of 6 mm which you can normally find over the internet and I am using two of them having the power of 40 watts each. During my tests, I found that two heaters and more than sufficient if your heater block is insulated from outside, and with my setup, I am able to reach up to a temp of 300° c in 3 min. 20 sec. Definitely, it's not a fast heater but it's capable of retaining the heat for a longer duration. I found that gradually drilling the hole from smaller to larger one gives a much better wall finish. For the center hole, it's highly recommended to have a tapered hole like a converging hole so that the material heats up well. I tried with a step drill but it didn't work as I wanted. In a 40 mm long block the first hole is kept 18.25 mm up to a depth of 10 mm and from here after 5mm for the 16mm, 5 for 14mm, 5 for 12 mm, and 5 for 10 mm hole is drilled with a normal drill bit. Definitely, it's not good compared to the converging hole but I have no other option, but I strongly urge you to go with a lathe or some other tool so that you are able to have that converging type hole. If the hole has these steps formed inside it then the motor needs to do a bit more effort to push the material. Since the stainless steel pipe and the hole didn't have much of a difference that's why I decided to go with a heat and press fit.

Once the hole for the heater and the main chamber drilled I start the process of grabbing them to their place during the printing because if the heater block starts to heat even if there is a precise hole drilled for the heater cartridge after heating that hole going to expand and that precise fit gonna lose and heat is not going to flow effectively. In my previous attempts, I used a hacksaw and jewelers saw but they took a lot more time and cutting is not going to achieve straight so at this time I made a fixture and clamped the grinder to its place securely. The grinder is a variable speed grinder and I am using it at the slowest speed. I made a riser block so that I am able to lift the block at the center mark which I marked on the center block and then insert the block into the wheel to make a cut. During these cuts, I kept on cooling down the copper block and using all safety measures. So this carefully or if you have a milling machine then this setup is not needed for you. After an hour-long work, I am able to achieve cuts on both sides.

To pinch the block to hold the heater cartridge I made a mark that lies in the center of that cut and then drilled a hole for the M3 thread. One thing I recommend is that in the area where you have to apply force for clamping or tightening mechanism, in that area it's much better to use a bigger screw to avoid thread stripping or you can also use thread inserts.

A thermistor is a sensor that is going to sense the heat and tells the firmware. Now normally the thermistor is installed closer to the nozzle but in my case, I decided to install it in the middle of the entire block. Being a copper it's going to distribute heat evenly so whatever the temperature is it remains the same in the entire block. To hold the wire to its position I also drilled and thread another hole for the M3 bolt. The thermistor is closer to the ceramic heater therefore it's not going to take too much time to read the temperature.

Once the copper block is ready it's time to attach the heat breaker or the stainless steel pipe. The reason for using stainless is because its thermal conductivity is very poor and it does not transfer the heat very easily to the heat sink area. For that, I place the stainless-steel pipe in icy water for 30 minutes and in the meanwhile I clean the copper block and start heating it. The expansion is not too High but with a change in dimension of both the materials, it's definitely going to make a big difference. With the help of a torch, I heat the copper and then place the pipe over it and smash it with a hammer. Later on, after trimming it to the required length I press more with the help of a bench vice.

Once the heater block work is finished I start the work onto the pellet feeder, although the nozzle hole is still left I continue in the upcoming section. For the feeder I learned from my previous attempt, I have to increase the size so that I am able to feed the bigger pellet also. I am using a 1-inch pipe having inner dia around 18.75 mm. I cut two pieces from the aluminum pipe one is going to be the main chamber in which the auger screw is going to rotate and the second one is that through which pellets are being fed into the main chamber. In my previous attempts, I found that with a diameter of 13 mm the pellets get locked from each other and it's difficult to print until you poked something into the chamber but I found that with 18 mm pipe it's not a big deal. The pellet feeder pipe is going to be welded at an angle of 45 deg. to get a better slope and faster material conveying into the main feeder.

I also decided to make a heat sink onto the pellet feeder because if any amount of heat travels from the heater block to the main pellet feeder and brings the temperature to the melting stage then the entire system is going to be clogged up. For that, I am using a 32 mm wide strip both 1⁄8" and 3⁄8" in thickness. The dimensions of the heat sink are going to be around 50 X 50 mm. There are a total of 6 strips going to be cut down and two of them are 3⁄8thick and the remaining 4 are 1⁄8" thick.

Once the material is cut down then I move onto my belt sander and start deburring the edges and make everything smooth. Since there is a pipe needed to be welded at an angle of 45 deg that's why I need to make a notch in one of the pipes also. For that, I use a 1-inch side of the belt sander and use it to grind down the 45 deg side of the pipe. After this it's ready.

Once the material has been cut down and cleaned up I start the work onto the welding the pieces together. To weld the pieces I need to drill a one-inch hoke in all of them. So after marking the center, I hold them in vice and then with the help of drill and hole saw made a one-inch hole in all of the pieces. The pieces come a tiny bit lose but since I am going to weld them, therefore, they are able to transfer the heat very easily to these fins and keep the chamber cool during the printing process. After that, I deburr the holes and clean them in acetone. From thereafter they are ready for welding. The first piece is the thicker one, and by keeping the piece perpendicular to the pipe I made a complete weld. From here after I use 3mm carbide rods which were in my drawer and use them as a spacer. I am not good at aluminum welding so onto the thin strips I use only three tack on one side and 3 on the opposite side to make it secure. I think this is plenty enough to transfer the heat and then cool the entire feeder area.

For this, I need to drill down a hole in the main chamber so that pellets can enter inside the chamber and be able to feed into the heater block. I manually hold the piece into hand and then with the help of paint make a mark and later on drill down a hole. Because of paint on this area, I get absolutely worse welding joints but with the grinding, it's not gonna bother. Once the hole has been drilled hold the piece into the vice and makes tack weld first from thereafter making a complete weld. Later on, with the help of carbide burr grind it flush.

Once the feeder is completed I start the work onto the motor mount. The design is pretty simple and the material I am choosing here is a 6mm thick aluminum sheet. Being aluminum it's also going to act as a good heatsink and I think able to keep the motor cool also. I marked the dimensions onto the plate and then with the help of my table saw I cut two strips both have different dimensions one is 43mm wide and the second is 55 mm wide. Two of these strips are needed to be cut equal to the length of the motor so that I am able to make some space to tighten or loosen the coupling. Because of the odd motor design there, I need to install two plates having holes in the center but both having different diameters. The front one also needs to drill some holes so that I am able to fasten the motor to the mainframe. The holes are kept a bit oversize so that I have a slight wiggle room for errors. There are also two more square pieces that are cut down which are going to join the motor mount to the feeder area but in one of them, u also need to make a hole for a 10 mm bearing. Having OD 26 mm.

In my few failed attempts the frame I was using was 3d printed but I noticed that as soon as the motor starts to turn the entire frame gets rotated and also creates a crackling sound. So that was also the reason for making it with aluminum and I think it looks more professional than the printed version.

Once the material has been cut down it's time for the prep work. I cleaned the pieces with the help of acetone and also deburr all the edges. First I made a tack weld to two sides one is bigger and one is smaller. I clamped it to a bigger piece of aluminum to make it stable during the welding process. To prevent the over-tightening of the motor mount I place a piece of thick paper so that after shrinkage there is enough space left so that I am able to freely insert the motor into the frame. I just make a tack weld while the motor is inside the frame. Make sure not to overheat the motor and avoid keeping it inside the frame during welding.

After that, I grind the sides onto my belt sander and make everything flat and flush. There are few fish eye holes left due to welding skills but I leave them as it is. In the area where belts can reach I use a file to remove the material and make it completely flush Side which is going to be the front side onto I also made a hole so that the wire didn't have any problem coming out of the mounting.

As I told you earlier there are going to be two plates which are going to be used for connecting the motor mount and the pellet feeder. I marked the center and drilled hole for the clearness of the coupler and in the second one made a hole to fit the bearing. Along with this I also marked the hole location on both the plates because later on, they are going to be used for a bolted connection for easy disassembling. There is the third plate in between both the two assemblies which are going to act as a barrier and prevent the auger screw lifting movement which I noticed in my first few attempts. If that thing happened then all the force went straight in the motor and it has to bear two forces one is rotation and the second is this pushing force. So this piece is absolutely mandatory. I welded a piece to the motor mounting assembly and a second to the feeder. On the feeder side, I use the bearing to lift the feeder so that the bearing remains completely flush to the surface and that the barrier plate doesn't interfere with the rotation. Once the welding work is finished with the help of file and flap sander flush smooth the surface.

Later on, onto the marked position, I drilled some holes and then tapped them with an m4 thread tap. Onto the motor mounting assembly, the holes do not contain threads.

To mount the entire assembly I need something along the sides so that I am able to mount it onto the CNC. For this, I use a 30 mm wide strip of aluminum and then weld it along the sides of the motor assembly. During the welding, I placed it onto a flat piece tack weld first and then made a complete weld at the back from there after making a complete weld onto one side. During the welding, it's good to have separate brushes but I kept contaminated the surface with other metals and I think that is also the main reason for bad welds. So if this is happening to you make sure to have a separate dedicated brush for cleaning aluminum. Later on, I cleaned up the welds and made them uniform and used 150 grit sandpaper to even the scratches.

My initial plan is to hold the piece with a grub screw but I noticed that the amount of force made by the motor is pretty high and often with that force the heater block pops out from the feeder, therefore to counter that force I welded a metal strip around the pipe and then fusion weld with tig. Now off camera, I made some groves so that it's easy to enter the heater block in and out. Since that needs to be attached with a screw that's why I need to make some threads in the feeder area. At first, I was using an M3 screw but during further testing when threads failed I had to use the bigger M5 screws and from then everything is fine. Along with the bigger screw I used some brass washer custom made for this to reinforce the strip. I have to do this work because I underestimated the motor force. But glad it failed and I get a chance to share this with you. That's why in my earlier statement I said it's good to use thick bolts if there are possibilities and areas under tremendous force.

Being a completely custom-made extruder the nozzle also needs to be custom made, but due to lack of tools, I decided to go with my method. Since I didn't have a lathe and broke many .7 and .8 mm drill bits I decided to have the hole is around 1.5 mm because that size I am able to achieve with my drill. Now during this, I made many different nozzles having different lengths but during this, I showed only one but the construction process is the same. Earlier I was using a regular m6 hotel but I noticed that due to the big inlet that the tiny opening of the V6 hot end gets clogged because it suddenly shrunk to a smaller size. So to counter that problem I decided to make my own with an M10 Allen bolt. I drilled the hotend and tapped it with an M10 thread. Then I filled the cap with welding and then started grinding it. Once it gets grind down I give it a regular nozzle type shape and also the one which is easier to tighten with the help of a single wrench only. Off-camera I made different inlet nozzles with the same outlet of 1.5 mm dia but I noticed a major difference in them. The one with an inlet diameter of 3 mm pays much more pressure onto the motor compared to 6 mm. And in the 10 mm dia bolt I found that if the nozzle inlet opening is 6mm then it's good compared to one with a 4mm inlet opening. Once I drilled the hole at the threaded end with the help of a smaller drill I made another to the facing end but they definitely don't coincide with each other. I think if they made lathe that would give extremely good results.

I have to say this part I really love the most. I never made an auger screw but after watching a couple of videos I thought that I could do this job. At first, I did this with a smaller shaft of 5 mm but due to the force some of the washers weldings got broken and ended up clogging the path and the feeder is no longer able to feed the material. I also ordered a wood auger bit because it's not available locally and it was a 14 mm dia bit. I thought I could grind it a little bit and fit in my previous design but after watching each part fail I dumped that idea and made a bigger feeder with a bigger screw which was the end result.

For the screw, I used m10 washers and a 10 mm stainless-steel shaft which was laying in the shop. In my test, I found that the auger bit dia of the washer needs to be slightly bigger than the shaft, in my case I found that a 12 mm screw is sufficient enough to give me a good travel distance.

I first drilled a 12 mm hole in the washer which I am going to use. Then I cut them from one side with the help of Hexa blade. After that, I hold half the portion into the vice and another half with the locking plier. Then with the help of a gentle twist, make a twist in the washer and this will also stretch it a little bit. I think for a much better explanation it's good to see the video of how I made them twisted and kept checking them by inserting them into the vice. If it comes loose then stretch them a little more. Or you can also measure the opening also to maintain consistency.

Once I prepared a bunch of these washers then I started the welding process. One by one I aligned them and then welded them together. Once it's done. I secure the top and the bottom spiral to the main shaft and after that. With the help of a hammer and chisel make them look uniform. I knew that I could use an auger drill bit but I thought this would generate much pressure compared to a wood auger because of these spirals. Once the bit seems to be uniform I start the complete welding process.

Once the screw is made it's time to give it a look of a real screw. First I started grinding it onto my belt sander since the surface is flat it will keep the entire screw flat as possible it could be. Then I hold it inside the drill bit and then with the help of foredom and angle grinder start shaping it. And I myself get amazed to see how well this came out. One thing I found out after using it to polish the entire screw. That will help to pour the material without any friction and hopefully be able to save some of the motor energy. I always think that a better option to align these kinds of screws is a lathe. That will maintain a minimum gap to the wall. One thing you also noticed is that there is a hole in the shaft and a pin is placed inside it, I would suggest that if you are able to find a pillow bearing that would be a much better option compared to this pin setup. I have to install the pin onto both sides of the bearing in order to avoid any moment of screw towards the motor. Which I saw in the previous tests.

For easy bed levelling, it's easy to have a sensor attached to the extruder. Some use proximity sensors but I always prefer bk touch as I am used to this. For the mounting system I used a 1⁄8" thick aluminium strip and bent it in l shape. I measured the distance of the heater block along with the nozzle and then cut the strip according to that. Then I drill holes to attach bk touch and also threaded some to mount it onto the heat sink. I am really happy to see that the entire system is pretty sturdy enough with that welded frame. I think making an aluminium frame is definitely a good approach for me in this build.

The thermal conductivity of copper is high, if it gets heat up faster then it also cools fast, so to avoid that I need to make a jacket. For that, I use fiberglass cloth with glass fiber wool. I measured the length needed to wrap the heater block and then with the help of Kapton tape sealed it from all sides. Now the fiberglass jacket is ready and believes me this is worth the effort.

Once all the pieces have been ready I cleaned all of them in warm soapy water to remove first and oil and then start the assembling process started from the motor first. The entire thing made pretty straightforward that even if you made it, it doesn't take much time to understand. I didn't figure out the exact location of mounting holes that's why I didn't drill them yet and that's why the sides look plain. The entire design was made in such a way that you definitely don't have a problem screwing or unscrew the things.

Once everything is assembled before start using the extruder you need to Tune the firmware because being a custom-made heater block system needs to know what type of block you are using and how much time it took to heat up. For that particular thing, you need to tune it. For that

M303 H1 S200 can be used.

Here H stands for the heater number. Usually, the bed heater is numbered H0 and so on. S stands for the temperature you wanted to raise. Like S200 means that temperature goes up to 200°C.

Once you click enter it starts the tuning phase and heats the heater. During this time the system notices how fast the temperature rises in a second and how fast it cools down. Once it's completed use M500 to save it in your firmware directory.

Along with this you also need to tune the steps. For this, I use a trial and error method. It definitely took time but I think it worked for me. I start by considering the dia of the manhole as a filament and during this, I kept the steps 800. Then I made test prints having filament sizes of varying thickness. The one in which flow seems quite consistent I consider that one which is 6mm and then start reducing the steps until it starts extruding the filament having a width of 1.5 mm. Now I reached 635 steps but still finding the dia to be 2.1 mm, which is .6 mm more than the required but I think I am finally able to manage this sooner or later but if you have any better and much more accurate method to tune the steps than definitely let me know.

At first, I didn't use the fan but as soon as I started using the extruder I think it would be a necessary thing to do that's why you saw me using fans onto the heat sink area. These are 50 mm fans and very effectively doing their job.

This is definitely being listed in one of my toughest projects but I am happy with one thing that I am able to make a working prototype of the extruder. During this build, I always thought that it would be great If I had access to a lathe, and I think the project looks much better than this. But I think keeping that aside I finally conquer my dream of building a pallet extruder which I was planned a year ago when I made a 3 in 1 CNC. Weight might be an issue for many machines out there as it's around 2 kg but my machine handled it pretty perfectly. The result was definitely not like this in the first attempt and I don't want to demoralize anyone by showing that. The final verdict is that yes you can make a pallet extruder on your own. Now have a look at the print and the final shots and I catch you in the next one.