The Badass Pellet Extruder

So, time for a new thread

As you might have noticed, I've been developing the motion system hardware for the Hangprinter since October. I source most of my stuff in China nowadays, so there has been also a lot of waiting involved in that. And to keep myself busy, I've been working on the extrusion end of the printer. During the summer I used the Hangprinter in two large-scale print projects, which were... let's say very educative. I spotted several issues, most of them mechanical, but the biggest issue was the material output per hour and its supply during a week long print. Although I have the SuperVulcano attached, it is nowhere near the output capability I need.

Even though I used 2,5 kg spools, I wasn't able to print overnight without disruption. And that of course accumulates in production delay. One way to deal with automatic spool change would have been the Prusa MMU2 for example. But, from what I've read, it isn't foolproof either. Combined with the fact, that FDM as technology can't be sped up much further, I came to the conclusion that it has to be something else, if we ever want to achieve speeds needed for industrial applications. So, I set a goal of approx. 2 kg/h material output and a temperature range which covers every possible polymer you could come up with. Further, I wanted to have a changeable nozzle in a size range of 1-6 mm.

And this beast was born in the lab:



Length: 550 mm
Weight: 4,5 kg
Heating power: 3 x 200 W / 230 V
Motor: Nema23 with 46:1 reduction gearbox
Temp Sensors: K-Type
Max temp: 400-500 C

It is yet to be tested. I'm waiting (surprise) for the Duet thermocoupler daughterboards to arrive, before first firing. That will happen before Christmas. Then I will be able to tell more about the performance. For testing, I'll be using commercial pellets to maintain steady quality of the extrusion, but my goal is to eventually use the 3D-printer waste material we have been collecting in our lab and to do experiments with your daily HDPE & PP plastic packages. These will be shredded to a granulate fine enough to be feded into the extruder.

The barrel and the heating elements will be insulated with a silica aerogel/fiberglass blanket, which is currently the best commercially available thermal insulator.
Check this video! It's insane!!

The flame is easily beyond the 1000 C mark and 10 mm above that, the highest reading we got was 64 C. Even that didn't feel warm at all, it was like cozy warm.

Well, and since the extruder is going to put some weight on the scale, I'm also designing a sturdy end-effector made of 1,5 mm steel sheet metal. The parts will be laser cut and bonded with glue and rivets. I've put some thoughts into getting it torsionally stable. The line guiding and every load bearing part of the printer will need a redesign obviously, but we'll get there part by part.

Let me hear your impression, feedback, etc.

Edited 2 time(s). Last edit at 12/20/2019 04:09AM by atmark.

---

company website // portfolio // project gallery // instagram

Update:
Ok, so I got the thermocouple daughterboards finally. One daughterboard serves as a terminal for two K-type thermocouples. The second board is stacked onto the expansion connector of the first board. I got a reading from the 1st sensor without big hassle, but the two others were silent. It was a struggle, until I realized that Tobben's firmware for HP4 uses the SPIO header for communication with the ODrive board. I assumed, that this interferes with the temp sensor communication and for testing I flashed the Duet with the latest v2.05 firmware.

After flashing I was able to get readings from the 1st & 2nd sensor, but not from the 3rd. Both daughterboards are definately working, since I swapped the boards for testing. Haven't figured out the mystery of the 3rd sensor yet. I also noticed that the heater bands I bought from Aliexpress are sub-standard quality. On one heater the lines are shorted to the sensor sheath... That makes the signal noisy and fries the Duet in case there should ever be a short from mains to ground.

Today, finally after fixing a defective band heater (crappy Aliexpress...), I got all three of them working. And I had also a chance to do a first test extrusion with LDPE pellet I have laying around.







Video: Test extrusion

Edited 1 time(s). Last edit at 12/30/2019 01:04PM by atmark.

Great news!

After replacing the nema 23 planetary gear reduction to 4:1 things start to roll nicely. I was already able to push the output to approx. 800 g/h
And I'm sure that wasn't the boundary yet, still aiming for 2 kg/h

Here some images from testing the extruder with UPM Formi -pellets. It is a cellulose-PLA compound with 20 or 40% of cellulose fibers.

Really nice work! I am thinkering about something similar since de filament is getting really expensive.

Don't you need any cooling of the tube in the beginning? That is something I am worrying about... any chance of sharing the parts you used for tube, screw and nozzle?

Stijn

Quote
stijnvanstijn
Don't you need any cooling of the tube in the beginning? That is something I am worrying about... any chance of sharing the parts you used for tube, screw and nozzle?
/quote]

Yes, active cooling is needed. I've already ordered heat sinks with fans, that clamps themselves on the barrel. Before they arrive I can't really test the extruder. The feed area overheats quite fast until to the point the pellets start to melt already in the hopper. When that happens no further extrusion is possible. I'm adding also a print area heating/cooling unit. HDPE/PP are tough materials when it comes to layer adhesion. If the layer beneath is already cool the hot extrusion will have issues to stick to it. Therefore I'm planning to add heating element into the air stream that is directed to the nozzle tip area. The heater can output up to 500 deg C. I hope I can quickly melt or at least soften the layer beneath before a hot extrusion layer is added on top of it.

I will publish the blueprints of the extruder once it works reliably. It works on mains voltage, so there is a real danger of lethal electric shocks. Something, I don't want to happen to anybody.
But, stay tuned. It'll be out this year I think before summer.

What is the screw diameter? I am working on a gate valve principle for the nozzle, so that there will be no oozing. Maybe we can help each other out a bit!

I am looking at a 16mm screw and barrel now with a compression rate of 2,5 were I am trying to implement the system in!

The screw diameter is 16 mm, with a compression rate of 2,5. So, I think we're talking about the same screw here.
Your gate valve idea is interesting, since there will be oozing/retraction issue if you don't address this matter. Have you done already extrusion tests with the valve? I thought about a valve too, but I haven't decided yet if and which type to use. I'm having some concern about the sealing, because the pressure at the nozzle are can easily go to 250 Bar. You will also need a strong servo to control the valve.

So there are some spring valves available which actually open at 200 bar, but I am not sure if there would give a constant flow. I would rather solve this matter mechanically where there is more control. Adding another stepper / servo to open and close the valve is something where I am not a big fan of due to the complexion of it.

I just had this Idea and working om some drawings right now. I will actually will try a 20mm screw maybe because I need some more space (inside of the screw). I actually didn't know the pressure could be so high at the nozzle, I thought it would be much lower, quite concerning for what I had in mind. If I finis my drawing I will send it to you and we can have a look at it.

So there are some spring valves available which actually open at 200 bar, but I am not sure if there would give a constant flow. I would rather solve this matter mechanically where there is more control. Adding another stepper / servo to open and close the valve is something where I am not a big fan of due to the complexion of it.

I just had this Idea and working om some drawings right now. I will actually will try a 20mm (compression 2.8) screw maybe because I need some more space (inside of the screw). I actually didn't know the pressure could be so high at the nozzle, I thought it would be much lower, quite concerning for what I had in mind. If I finis my drawing I will send it to you and we can have a look at it.

Ok, so I got the heatsinks and they fit perfectly onto the barrel. But, as I want to have uniform cooling on both sides of the barrel, I had to saw pieces off the heatsinks so they fit in opposition to each other. For now they are hold in place by aluminium plates, but I might finish it with a hinge on one side and a tightening clamp on the front side. I will also swap the 5V fans for 24V equivalents. On the back side plate I will attach a case for the thermocouple daughterboards. Next thing to figure out will be cable management for the heater band power cables. It would be nice to have one central junction box for all cables.

Next week I will test the extruder again, this time with the feed section actively cooled. So, I'm aiming for an endurance test to see how everything works in continuous run. Temp rise, material flow, etc.

---

company website // portfolio // project gallery // instagram

Time for an update!

The Badass Pellet Extruder has got a electronics enclosure for the Duet thermocouple daughterboards and for the power cable junction. This is where all the cables coming from the control board will be terminated. Now it starts to look like a neat package. Next thing on the todo list will be an anti-ooze mechanism for the nozzle.

---

company website // portfolio // project gallery // instagram

Hello atmark,I assume it is already in production yours pellet extruder, do you have a clip with it working?

Well, the extruder isn't quite yet ready, since we have put much work into the printer during spring/summer. But we are getting close. There aren't that much of issues to be solved before we get to print.

Here's the latest clip of the extruder & pellet transport system working

---

company website // portfolio // project gallery // instagram

You have done a great jobb, all my appreciation! Did you think about selling one ? what would be your target price?

Quote
atmark
Well, the extruder isn't quite yet ready, since we have put much work into the printer during spring/summer. But we are getting close. There aren't that much of issues to be solved before we get to print.

Here's the latest clip of the extruder & pellet transport system working

Very impressive. You should take a look at Dirk van der Kooij's work. He also has quite an interesting extruder!

---

http://www.marinusdebeer.nl/

Quote
DukeNukem
You have done a great jobb, all my appreciation! Did you think about selling one ? what would be your target price?

Yes, we are planning to sell the extruder setup with or without the pellet dispenser system, so you can use either a big hopper and some magic a la gravity, or if you prefer to print huge without intervention, then you should go with the auto dispenser.

For the extruder we are aiming for around 1500-2500€ (I realize that this is a quite large range, but for the prototype we used cheap chinese components that I wouldn't want to suggest to anyone. We will swap for high quality components from trusted suppliers) That will reflect in the price, but in the long run you will appreciate the better quality, safety and endurance.

Feel free to DM me, if you would like to participate in our beta program. We will offer the extruder for a reduced price to the participants.

---

company website // portfolio // project gallery // instagram

Hello Guys. Congrats Atmark!!
Very nice your venturi solution. it seems light!
As I wrote you privately in the past, we also have a pellet extruder that is working properly and we are doing tests with different pellet materials.
You can check our website if I'm not disturbing "www.rev3rd.com".
I'm attaching a video about it. If someone is interested in our solutions, can write me, making a reference to this post.

PELLET EXTRUDER RX-M10 REV3RD IN ACTION

PELLET EXTRUDER RX-M10 REV3RD IN ACTION 2



Kindly Regards
Alessandro Torre

Edited 4 time(s). Last edit at 01/06/2021 08:45AM by AlessandroTorre.

---

Alessandro Torre | Founder & Mechanical Engineer
a: REV3RD s.r.l. | Via Carlo Pisacane, 4 | 72015 Pezze di Greco - Fasano (BR)
e: alessandro.torre@rev3rd.com | w: rev3rd.com
m: (+39) 338 866 89 30