Evolution of the Perpetual Plastic Project Extruder
We intend to be a completely transparent organisation and we have been showing and talking about our technology with anyone interested since we started. We believe that there is more and faster return on investment in sharing knowledge then in hiding it. Having sprouted from and being part of an open-source community requires that we actively give something back. Therefore, by popular request, Laura has taken the time to document the evolution of our extruder designs. For each iteration we have listed the obstacles we encountered and the improvements we made.
* Source files: The step files we have right now are kind of messy so stay posted for updates. Feel free to contact us about details. We promise that we'll always respond but we can't promise when. (-:
Extruder 0.0 [March 2012]
When we first had the idea for the project we didn't even know if it would be possible to recycle cups and create filament out of it. That's why we made the most simple extruder imaginable. It consists of a tube with internal threading and a threaded rod that pushes the material through by turning it. For a nozzle we used standard plumbing parts that have a standard compression fit on the tube. We heated the nozzle with a torch. Unfortunately the fit wasn't tight enough and the nozzle came off.
improvements: none. first version.
+ : easy to make. standard parts.
- : nozzle falls off, speed of extrusion (meant for testing)
Extruder 0.1 [April 2012]
We kept the design the same and replaced the nozzle with a nozzle that we turned ourselves on the lathe. With this extruder we were able to create our first filament from old plastic cups and we made our first print with it. We also found out at this stage that the plastic shreds have to be dry to prevent the water from causing bubbles in the filament. We had proven that our concept would work. The next step would be to create an extruder that could extruder material continuously instead of having to reload it by turning back the rod.
improvements: compression fit nozzle replaced by custom nozzle
+ : first print was made with material
- : material has to be dry, extruder has to be reloaded by turning back the rod
Extruder 0.5 [June 2012]
This extruder design was inspired by Bigelow Brook Farm's plastic extruder, meant for creating plant growing beds. At the time, that was one of - if not THE only DIY extruder design that was shared on the internet.
We used an 18mm auger and a small DC motor that was controlled by Arduino. For heating we used NiChr 0.5mm wire that we insulated by hand with Kapton tape and coiled around the tube, powering it with an ATX computer power supply.
The small DC motor wasn't powerful enough and failed halfway through testing, so we continued testing by hand cranking the auger. This actually worked very well because the torque one is able to deliver by hand can compensate for irregularities in the shredded plastic input.
improvements: automated process; hopper, auger, motor, electrical heating
+ : automating the process
- : motor not strong enough to handle irregular input
Extruder 1.0 [July 2012]
For the final design to be integrated in the installation we decided to keep the extruder hand cranked. This eliminates electronics that are needed and compensates for a large motor that would be needed for extruding recycled material. We thought that making the extruder vertical would help the flow of material by gravity.
Because the crank had to be on ergonomic height the nozzle ended up a bit low. This complicated the catching of the filament because we had to get on our knees to retrieve it and it was less visible from a visitor point of view. Another design flaw was the tolerance on the gears and the auger which causes the auger to grind the inner tube a bit.
improvements: hand crank, vertical orientation
+ : integration in the interactive installation, less susceptible to electronic failures
- : catching the filament, tolerances in the system
Extruder 1.1 [July 2013]
A year down the road and some minor adjustments later we gave the extruder an overhaul again, also in the cosmetic sense: we used acrylic that we had previously also used in our printer and shredder. It just looks better. But the functionality was also improved: we put bearings in the auger enclosure to prevent grinding and the orientation was changed back to horizontal to facilitate the catching of filament.
Because the flanges of the auger housing were welded this caused some deformation in the metal thereby again creating tolerances on the auger and housing. An idea was to create the flanges from 1 part in order to eliminate the need for welding.
improvements: horizontal orientation, looks, bearings,
+ : catching the filament, looks the same as the other machines now
- : tolerances in the system, maintenance requires removing many nuts and sections
Extruder 1.2 [January 2014]
To better know at which temperatures we were extruding we switched to using heating elements (160W). The temperature is measured with a type K sensor (M6 x 12mm). For a constant feed of material we used an 28Nm strong motor at 8rpm.
Moving the heating elements more towards the front allowed us to experiment with the effect of heating the sections differently. This wasn't possible with the wire. Although the motor now gave us a nice and constant rpm we noticed that the flaw lay in the homogeneity of the input. If the plastic shreds aren't small enough they don't get transported by the auger immediately.
We noticed during our event on 14 Feb that the interaction with the installation is much less when the extruder is automated: then it becomes only a matter of watching the machine. We'll continue to develop the automation of the recycling process in order to create recycled filament spools but for the installation we will focus more on the interaction and feedback part in order to achieve better output when operated by hand.
improvements: heating elements, temp sensor, strong motor
+ : one person can make filament, temperature feedback improves filament quality
- : automation takes away from interaction in the installation, we need to add speed and emergency stop control to the motor