I like the idea of using a granule extruder. It eliminates one step in the process, and makes it much easier to recycle on-site. However, I see a problem with producing the same quality filaments we get from commercial houses. The granules or pellets start out with air gaps between them. When compressed and heated, the air gaps will turn into bubbles in the filament. If we go straight from granules to extruded 0.3mm these will become gaps in the stream, rather like extruding a dashed line rather than a solid one.
To fix this, you need to add some way of removing the air while still keeping the plastic in place.
One way is to use a vacuum system while the granules are still solid. With a fine mesh on one side, and an auger or gear pump forcing the granules through, and something to plug the opening (like more granules being pushed in) the air can be separated from the granules so that later, when they are heated and compressed it will be solid plastic. The advantage of this approach is that the mesh does not need to be ultra fine to prevent the plastic from squeezing through. The disadvantage is that a vacuum pump adds power drain, cost, and complexity to the design.
The second way is to pass the compressed, melted granules and air pockets through tube, funnel, auger that has a porous surface that the air can pass through but the molten, pressurised plastic can not. The best would be an auger moving the plastic through a porous tube, so that the auger would churn the plastic and get all air pockets up against the porous surface to allow the air to escape. The advantages would be simplicity and lower power consumption, but the problem is finding a porous surface in the size and shape we need that will have holes big enough to let the air out, but small enough to keep the plastic from oozing out or even plugging the holes. It would also need to be a low stick surface, and I am not sure how to get the really tiny holes needed in large enough numbers in PTFE.
The problem of excluding the air is solved in commercial extruders by a simple heated auger. The molten plastic has a more difficult time escaping (especially past all the non-molten granules) thus forcing the air back out the hopper end, and leaving only plastic on the business end. The auger also pressurizes the plastic for extrusion.
This has been done, with varying degrees of success, by various people in the reprap community. I think adrian built one out by wrapping heavy copper wire around a rod, but the heat conducted back up, and melted the plastic still in the hopper.
I think a ceramic auger would work better, and I think one cold be made by the following process:
1. Print a plastic auger in two halves, at the same time.
2. Remove the build surface, and place in the bottom of a watertight box, with a paper/cardboard/plastic/whatever wall between the halves.
3. Pour full of plaster. (Drywall compound, plaster, same thing)
4. Remove the plastic bits etc, and take the two-piece plaster mold to any ceramics shop. They'll pour it full of slip, let is set an hour, remove the auger, and let it dry.
5. Finish the "greenware" auger, removing the mold marks etc.
6. Have the ceramic people fire it for you.
7. If needed, repeat for the "tube" the auger fits in? Maybe just use a piece of pipe?
You'll need a very flat print bed for best results...
Hi, I might be able to help some with a granule extruder. Looking over that past thoughts
it seems like you want a hot melt zone, but a cool granule pushing zone and not a large transitions
zone because the transition zone can get stuck whenever turned off. So that makes me think of heat rising also,
and the idea to pull air out of thin slots and/or holes in a ring around a vertical tube that granules are descending in will keep them cool and unmelted in the granule flow zone during operation and during turn off and cool down.
Another thing to consider is making the pushing force stop start instead of continuous, so you could get a positive granule moving effect, yet have some simplicity and ease of undoing any meltdowns. I imagine a pusher with a plunger that goes down to close a hole in a the bottom of a funnel full of granules, pushes some to get melt and flow, then comes up past the hole closing point, where more granules fall in, and the cycle repeats. It's easily possible to jam such a contraption, so some experimenting is needed. A plunger about 15mm with flat bottom pushing into a tube top with diameter 20mm might push 4mm granules well and not clog much. The plunger edge should be thin, about 1mm, then become less above, maybe a 5mm diameter if made of metal, perhaps looking like an IC engine exhaust valve.
How do you keep plastic from dripping? Do you just let it flow -- that's the way most reprap output looks to me so far..Kind of lumpy.
Without the naturally fast flow regulation of plastic filament, perhaps output flow control is needed? That
makes me think of a needle valve where the needle comes in from the side so it cannot get frozen in plastic coming through the nozzle. The nozzle needs uniform heat so it does not freeze up, so it could be brass or bronze. The needle can be thin stainless so it conducts low amounts of heat and lets the nozzle get hot and unfreeze.The valve body between the needle and nozzle needs to be insulating some for low heat drain on the nozzle. Stainless steel sheet metal might be good for that.
Thinking about small size for use on a light weight reprap printhead, what if the granule pusher is sized to push one granule at at time of commercially available sizes, ( they seem usually to be from 3mm to 5mm, sometimes larger)? Is it important to regrind your own plastic instead of dealing with a recycler? In my city we have a nearby plastics recycler that cleans grinds, pelletizes plastics so it's not important for me to do that myself...
So, if you want a single plastic pellet pusher, it could be a shaft motor that is small and light -- it might still need vertical height above, but that seems compatible with reprap designs I've seen.
There are always some projects going on with the devolopment of a granule extruder, beceasue of chaos properties of open source development, I will post this to notify you about the most recent (first quarter 2012) projects regarding granule extruders, including mine
I would like to notify you about my project, which is partially bases on research such as this forum topic and other simultanious projects: