Build Your Own 3d Printer Filament Factory (Filament Extruder)

1.2M4.1K667

Intro: Build Your Own 3d Printer Filament Factory (Filament Extruder)

Too long, didn't read:

Make your own 3D printer filament !
Cheap and high quality at a decent speed of 150-190 IPM ! (4-5 meters per minute)

UPDATE: Now with wiring diagram !

Long read:

3D printers are cool and they finally start to drop in price. Kickstarter campaigns like the one from QB-UP or M3D are popping up and they are finally "affordable". And with affordable I mean affordable like 200 $ and not "affordable" like 2.199$ affordable. However, once you are a proud owner of a 3D printer you will soon realize that your wallet is far from being let alone. No ! You need plastic filament of course to print those super awesome coat hooks and wheel chocks. Since the price for these filaments tend to top the actual material costs, printing before mentioned life savers is kind of expensive and could become a problem to the development of the ever growing 3D printer community

BUT FEAR NO MORE !! Some clever gents came along - Hugh Lyman with his Lyman Extruder may be mentioned here or the guys over at Filastruder.com - and saved the day ! YAY. And there was much rejoicing ! They have built plastic extruders everyone can build or buy at a decent price. However if you are a fellow Instructable.com user the first thing that should come to your mind is "I can build this by myself...and cheaper...". Building at lower costs is the nature of DIY after all.

And much more fun than putting together a premade kit, of course.

Special greetings go out to Xabbax and his plain simple but super awesome Low Cost Filament Extruder !

So how much money do I save when making my own filament ?

Good question ! A lot !

Depending on the pellets you get you can make your filament starting at 1$/kg.

How long does it take to produce 1 kg of filament you may ask ??

Using the build I describe here...roughly 1 hour. (for 1,75mm filament using ABS/PC pellets).

So, let's say on a Saturday in your next workshop session you start at 10 AM and batten down the hatches at 5 PM you could make 4-5 kg of filament, saving between 125-150 $ leaving you with lots of filament for hundreds of thousands of eggcups and phone cases and other useless needful things.

Oh yeah what about the build cost ?

Depending on shipping and local prices, I would guess around 130-150$.

Next step: List of Materials

STEP 1: Material List

Except for the electronics everything listed here can be bought at your local hardware store.

Materials:

  • 1x Wiper Motor (Ebay EU - 15€) / 5€ from the junkyard
  • 1x Auger bit(diameter = 16mm ; length = 460mm)
  • 1x PID Temperatur Controller - DC 12V version (Ebay)
  • 1x SSR-25DA Solid State Relay 3-32V DC / 24-380V AC / 25A (Ebay)
  • 1x K-type thermocouple (Ebay - like this one; does not need to be that shop :) just an example)
    -->!!! Sometimes the PID is bundled with an SSR and an K-Type Thermocouple !!!<--

  • 1x Motor Controller 20A (Ebay)
  • 1x Power Supply 12V, 240W+ (Ebay)
  • 1x Heating band (200 Watt 25mmx30mm) (Ebay)
  • 2x Fans (80mm) 12V
  • 1x Fitting 3/4" US Inch UNC --- 1/2" German Inch - 18cm long
  • 1x Water tap extension - 3/4" UNC threads --- 1/2" German Inch - 50mm long, 27mm diameter (one core thread and one exterior thread)
  • 1x End cap 1/2"
  • 1x Faucet-mounted filter - 1/2" diameter
  • 3x Steel angle
  • 1x Axial ball thrust bearing (Ebay) - Fitting exactly onto the auger bit's shaft.
  • 2x 10mm threaded rod
  • 1x Insulation
  • PTFE tape
  • Heat resistant tape
  • 3x Rocker (previously "rocket") switches
  • 1x Wooden board 100cm x 10cm x 2cm
  • Several screws and nuts
  • 2x sockets (1 that fits on the auger bit and 1 that fits on the nuts of the motor shaft)
  • Wires (two colors)

Tools:

  • Multitool (Dremel-like)
  • Saw
  • Hammer
  • Drill

STEP 2: Base Plate

Take the wooden board and cut away two pieces each 15cm in length (~6"). They will serve as a mount for the motor and for the barrel.

STEP 3: The Motor Mount

Mount the wiper motor to the motor mount and place it somewhere at the end of the base plate. See the technical drawing for an estimation.

Use the steel angles to attach it to the base plate.

The motor just has a threaded shaft. For the coupling to fit onto the motor I took a hex-nut with 13mm outer diameter and put it on the shaft. When the shaft rotates and the coupling is attached, the nut would untwist. To fix this I drilled a hole in-between the attached nut and the motor shaft and put in a 2mm steel bolt. This prevents the nut from opening. See the last picture above.

STEP 4: The Barrel Mount

Drill two holes into the other piece of wood so the flanges can be attached left and right of the board. Drill another 1/2" hole for the auger bit.

Both mounting boards need their center opening to be aligned to each other so the auger / coupling / shaft-axis can rotate freely.

Fasten the flanges with two pieces of the 10mm threaded rod. The rods must be left long enough so they can be screwed to the auger "kickback protection". 10 cm is good enough. They can be cut to size later on.

This will get clear in the next step.

STEP 5: Auger Kickback Protection


When the auger bit turns and hauls the pellets a lot of pressure builds up. In the worst case this could damage the worm drive inside the wiper motor. To counter that problem, we need a kickback protection. This is simply done by a sturdy steel angle and an axial ball thrust bearing.These ball bearing withstand alot of force applied to them.

It works like that: The auger pushed back due to its "backward" turning attitude. Because of its taper the auger's shaft pushes against the axial ballthrust bearing which itself pushes against the steel angle. The coupling between the auger and the motor should always have a little clearance. So that no force is applied to the motors shaft.

Now place the steel angle with the inserted rods at a distance to the barrel mount so that the auger's shaft sticks out for about 3-4cm (~1.5"-2").

The pictures should explain it as well. Moreover I have made a short video that should illustrate it as well. The dimensions of the parts might differ from the ones you have access to. So exact measurement might not help you very much, but the pictures should give you an idea how it should be put together.

STEP 6: The Barrel and Auger Bit

Barrel:

Smooth out the ends and the seams of the pipe so the auger bit can rotate freely.

Before cutting an opening into the pipe screw it tight onto the flange and mark the upper area and remove the barrel again.

Take your multitool and cut out the marked area at the end of the pipe where the pellets should fall in. Wind some PTFE tape around that end of the pipe. This should prevent the pipe from turning with the augers movement. Remember the motor is very powerful and if there is some friction between the auger and the pellets, the pipe easily turns another 4-5 mm even if it was fastened with a monkey wrench.

The threads on the flange and fittings are not made for perfect 90° angles. So the fitting/barrel might stand in an oblique angle. To fix this take some washers and place them under the flang where necessary.

Take a square piece of wood and drill a hole lengthways for the pipe to run through. Now drill another hole orthogonal to the "pipe channel" so that a bottle can fit tightly. Now just cut the block in half for easy dis/-assembly.

Auger-bit:

The auger might be too long so you need to cut off its tip with an angle grinder.

The auger bit should reach up to the heater. See the pictures above.

STEP 7: The Auger-motor Coupling

Take a 5cm (2 inch) piece of a square steel that fits into the ends of the sockets (about 12mm edge length).
Put the coupling on the auger bit and attach the motor to the motor mount.

The coupling should now fit nicely in-between.

Alternatively you could use a spark plug socket instead of the two sockets. But therefore the distance between the motor-mount and the auger/barrel-mount needs adjustment.

I went with the above mentioned method because I did not have spark plug socket at hand but I will try this with the next build.

STEP 8: The Nozzle

Nozzle diameter:
Depending on the material you process the diameter of the hole in the nozzle will vary and finding the right dimension is a process of trial and error. For ABS/PC blend pellets with a melting point between 240-280°C a 1.5mm hole perfomed well from my experience.

Breaker plate:

Take the faucet-mounted filter and cut it into a 1/2" diameter if needed. This will act as a breaker plate. What this breaker plates does is mix the molten plastic and retains dirty (which should not be there of course) and eventually small bubbles that could occur in the melting process. This helps smoothing the plastic pushing through the nozzle.

Be sure there are no chippings or strands ! You don't want to ruin your printers nozzle !

Take a washer, place it inside the end cap and put the DIY breaker plate on top.

STEP 9: Band Heater and Temperatur Probe (K-type Thermocouple)

Drill a 2mm hole near the front of the water tap extender for the thermocouple to fit in.

Strip the thermocouple wire to length. It should just be as long as needed.

Push the band heater on the tap extender. It should sit around the end of the extender.

Next take some PTFE tape and wind it around the thread of the tap extender. This prevents the molten plastic from squeezing through the thread.

Fix the thermocouple with some heat resistant tape.

Then put on the nozzle from the previous step.

Next, take a 10cm long piece of aluminium tubing with a diameter of around 1cm and place it in front of the nozzle using some rigid wire. This gives the filament a nice curl when cooling.
Thanks Xabbax for the idea.

Now wrap the insulation around the heater so that the nozzle is covered as well.

STEP 10: Cooling

The front of the nozzle and the motor needs some cooling.

The filament is still very hot and soft when it exits the nozzle. To prevent it from stretching too much from the affecting g-forces when falling down, cooling is very important. The more you cool the better you can control the diameter of the filament later on.

Although the motor builds up some heat and the fan helps to keep it cool.

STEP 11: Electronics

Now that most of the mechanical parts are set and done it is time for installing the electronics.
But before, take a piece of wood for the front enclosure and arrange the 3 rocket switches, the PID controller and the motor controller's potentiometer and fix them with some hot glue.

Main power
Connect the power cord via a rocker/t switch to the power supply (Ports L, N and Ground).

PID temperature controller
Connect the PID temperature controller via rocker switches to the power supply.

Solid State Relay & Band heater
Connect the 12V ports of Solid State Relay to the PID (Port 6 and 8)
Connect port 1 of the SSR to the 220V (EU) /120V (US) port (Port L) of the power supply.
Connect port 2 of the SSR to one of the band heater ports.
The other free port of the band heater is connected to the N port of the power supply.

What does the SSR do actually ??
The band heater is a 220V part but the PID only runs on 12V. Therefore the SSR connects the 12V PID with the 220V heater. The PID powers the SSR on and off if needed. When it is on then 220V are connected to the band heater and it gets warm. If the relay is off, the band heaters isn't connected to 220V and ergo is powered down. The idea is to control a high power device (Heater) with a low power device (PID).

Motor controller

Connect the motor controller via a rocker switch to the power supply. Then connect the motor to the motor controller. Use the pinout for the 2nd speed setting of the motor. The pinouts differ from model to model and you first have to find out which pins are for which speed setting.
The two fans are connected to the same ports as the motor is to the motor controller.

Wiring diagram
I am not sure if I am allowed to post the wiring diagrams for license reasons so I will link to the respective websites.


1) Filastruder wiring diagram

2) Filabot Wee wiring diagram (scroll down)

3) Here is a link to the Sestos PID I used.

STEP 12: Extrusion Settings and Setting Up the PID

Different materials need different extrusion settings.

For pure ABS a temperatur of 190°C is about right.

PLA requires less heat and ABS/PC blend needs higher temperatures like 260-270°C.

The Sestos PID is able to autotune to the desired temperature.

PID controller setup:

To enable the autotune function press "SET" for 3 seconds.
You will now see "HIAL" on the display. Now use the DOWN button until you see "Ctrl" and adjust it to "2". This is the number for the autotune function. Press "SET" again until you see the temperature readout again. Just after all the EP1-8 options. Set the desired temperature using the up and down buttons and wait until the display stops flashing (~10-15 minutes).

Activate the motor and let the extruding begin. You have to play around with the speed of the motor.
From my experience setting the potentiometer to half speed @ 270°C for ABS/PC performed very well.

572 Comments

What is a German inch ??????????
Those Germans invented the metric system & litterally get offended when you use imperial measurements.
3/4 inch = 19.05mm
1/2 inch = 12.70mm
This could mean anything from a M2 to a M24

This whole thread switches between inches & mm. (Very confusing)

Germans didn't invent the metric system, and they don't get offended by people using other things, but go off.

A German inch is called a Zoll, and plumbing pipes are sold in Germany with those measures for diameter for various reasons including compatibility with piping that was in service before metric adoption, similar to how a lot of plumbing pipes are sold in Australia with inch sizing for diameter when we use metric for pretty much everything, so we would have 1/2" BSP gal tube in 6 meter lengths for example rather than saying 12.7mm, changing the actual sizing of the pipe would create incompatibility with older plumbing and it was easier to continue to refer to it by the old measure and put the metric diameter in brackets.

The German author provided the sizing that pipes are sold in for their own country and also the sizing for the US because those are the tube/pipes needed for the project and aside from those pipe size references and cutting a filter to suit said pipe, and drilling a hole to suit the auger that fits said pipe, everything is in metric here but with some conversions added for the metrically challenged Americans. It's not at all confusing unless you want it to be.

After all these years I'm finally deciding to get off my butt and build this as I have lots of plastic from waste printer parts from when I used to repair printers that I want to recycle. I also have some scavinged thermistors and I am wondering if there is any way I can use one of them instead of a thermocouple.
Although I've posted in my profile as "hobbyist", I have 30+ years experience as a process engineer in extrusion, on machines ranging from 3/4" to 4.5" diameter, with a wide range of different plastics.

This instructable is great as far as covering the drive, mechanical & electrical aspects. However, this, and other similar designs suffer from the same fundamental design flaw: a drill bit is not a feed screw! A true extruder feed screw has, along the length of the barrel, feed, compression and metering sections, with progressively shallower clearance between the root of the screw and barrel. This is the key to stable melting and output.

I've experimented with one of the low-cost kit designs, with the included auger bit as well as other types (hammer drill bits, which look more like a real feed screw) - collected a lot of data (~1000 points) - measuring not the filament diameter, which is affected by cooling, but the weight/unit time exiting the nozzle. My conclusion is that even for an "easy" resin like ABS, output variability of +/-10-15% is expected. This can be accommodated somewhat with the flow setting on 3D printers, but jams are inevitable on the high end (> ~1.9mm on my Prusa).

I'm still thinking about, if not actively working on, a way to make this little machine more functional. The sticking point is the much higher cost of of a CNC-machined feed screw, many times the price of a kit or Ian's DIY approach.

(I know I'm late to the party)
So, obviously CNC machining feed screw would produce an end part with much higher precision. But have you thought about casting the feed screw?

3d print the screw in high resolution resin then just sacrifice it in an aluminum castijg with investment casting processes.

We're not building a high precision machine here. It should definitely be good enough for government work.
What about using something high temperature material and just turning a regular drill bit into feed screw. By progressively covering the screw until the end.
Or, sanding down a drill bit so that it can go thinner along the length. Then use that drill to drill a pipe.
Keep lining up the drill bit with kapton tape. Layers upon layer
I am totally with you in this regard. But an auger is the closest part I could get when building on a budget. And those home kits just use an auger an sell it with sky high price tags which IMHO is... You get the point. Using a real feed screw would also require to exchange most of the parts of my machine and effectively the outcome would be a totally new machine.
What I wanted to show with this instructable was that it is paradox to buy such a commercial machine because the soul of this hobby is DIY for the sake of DIY. Not to buy ridiculous expensive kits.
No reason why a real feed screw couldn't be made to fit inside your design, just needs to have same L/D and drive end to match the auger bit. I admit I took the easy way out with the kit, but at only about 2x cost of the pure DIY route. It was certainly an interesting project, and for me became a challenge using my professional knowledge to see what it could do as-is, and then with various hacks. Not to be too discouraging, but those seeing this design and thinking "I'll never have to buy a roll of filament again" (or, especially, "I can chop up my scrap prints and make new filament") need to understand its fundamental limitations.

There's another way, probably cheaper than a CNC-machined feed screw, to make a screw (still from a drill bit) & barrel (still from a standard pipe) that can generate compression that I'm looking into. Requires machine shop work, but simple lathing and wire EDM (not as exotic as you might imagine). Will keep you posted.
As I was watching the video , I was thinking about the differences between a drill bit and feed screw. Did you make any progress with your thoughts on it?
Hi, where to place the heater? Whether near the nozzle or mid of the pipe? And what temperature needed for PLA pellets?
Hi, I had done this extruder and the extrusion comes really well, but my problem was the diameter of the extrusion was varying in sometimes. So I didn't get the uniform size of the filament. So how can I overcome the thing??
Why use the wiper motor? Should you possible use other motors as well? For example a drill motor?
I ended up trying to build this machine, and after putting it together I had no success whatsoever. My barrel is slightly longer and I am not using a thrust bearing. Additionally, I am using blended failed 3d prints, so any of these 3 things may be causing my issues. My 2 main issues are that I'm not getting nearly enough torque out of the wiper motor, and the band heater struggles to get it up to temp. By the time the tip is up to temp the rest of the pipe is hot and I get no movement from the motor. I don't know what I was expecting after trying and failing to build an augur-fed extruder in the past but I would appreciate any tips.
Thanks,
Ian
I haven't made this, but I would say you should make sure that the augur can turn freely. This is brought up in step 3 of this similar instructable: https://www.instructables.com/Homemade-3D-Printer-...
Ensure that there are no burs around the end of the pipe, or around the opening where the plastic is being fed. If this isn't the issue, I would suspect you need the thrust bearing or a different motor. Also be sure your motor is being controlled with Pulse width modulation or something similar. PWM is important because it gives less power by sending bursts of full power, meaning the motor will receive more torque than a constant but lower amount of power.
The other thing I can think of causing issues with the torque is the plastic being fed. It should be fine to use failed 3d prints, but the smaller the pieces, the better. You don't want fibers getting wedged between the pipe wall and the augur, and you obviously don't want to block the filament. I hope that helps.
I ended up trying to build this machine, and after putting it together I had no success whatsoever. My barrel is slightly longer and I am not using a thrust bearing. Additionally, I am using blended failed 3d prints, so any of these 3 things may be causing my issues. My 2 main issues are that I'm not getting nearly enough torque out of the wiper motor, and the band heater struggles to get it up to temp. By the time the tip is up to temp the rest of the pipe is hot and I get no movement from the motor. I don't know what I was expecting after trying and failing to build an augur-fed extruder in the past but I would appreciate any tips.
Thanks,
Ian
Clever stuff :)

Does this machine have to use pellets? By that I mean can I recycle old prints?

Thanks

James
Drill, tap and use a standard print nozzle (brass) that way you can get a range of output sizes cheap, they go from .1 to 6mm and that way all you have to do is pick a size smaller (filament expands as it comes out) and purchase 1 letter drill, or pcb drill (cheapest as you get a set) and start at the smallest setting and drill out until you get your required size. Also works for those that use 3mm filament.
More Comments