bowden-esque granular extruder?

So i'm just wondering, has anybody considered a bowden type granular extruder, perhaps with some sort of valve on the end for print control?

IE, use 2 wheels to shove a bunch of plastic in whatever form (come to think of it such an extruder could take filament too) down a ptfe tube to a hot section, open and close nozzle of hot section to print?

the idea's been kicking around my head for a while now, and if nobody's tried it and run into serious speed bumps, I think i might have to try it, after properly tuning my regular extruder of course.

footnote: I've also been considering a bowden extruder mounted with a screw? this way you could tune it's height exactly quite well.

If I understand your idea correctly, it seems the granules would jam in the tube, or at least require a lot of force to keep them moving.

It seems like one of the important features of a stepper extruder is that it can pull the filament backwards and create a negative pressure that sucks the molten plastic back into the nozzle. You wouldn't be able to do what with a broken filament (stack of plastic beads); you'd only be able to reduce the pressure to zero. Additionally, it seems like the stack of beads would be unpredictable. As they shifted in the tube the distance they were traveling for each turn of the stepper would vary. Finally, to actually melt them and force them through a nozzle you'd need a beefy motor and gear on the extruder anyway.

The only real benefit to this approach seems to be that you could have a bucket of plastic beads connected to a moving extruder, instead of the extruder needing to be fixed in place.

Quote

It seems like one of the important features of a stepper extruder is that it can pull the filament backwards and create a negative pressure that sucks the molten plastic back into the nozzle.

If you manage to prevent dripping by some other means, e.g. by attaching a valve, this sucking-back function is obsolete.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

True, but it does significantly increase the complexity of the equipment. I'm never ever going to argue that throwing more money and expertise at something won't improve its functionality. I'm just curious whether or not it's a high-value addition. The whole point of the Reprap project is to do 90% of what the expensive machines do but for 10% of the price. It's a question of tradeoffs.

For example, in my mind the only reason to go to a granule extruder is if you want to print a whole lot of plastic because then the savings of not paying for the ganules to be turned into filament makes sense. It's a scale thing. If you're just going to putz around at home you're never going to push enough plastic through the machine for the savings to appear. And, if you are planning on pushing a ton of plastic through the machine, that implies you're going to (at a minimum) have the machine running 24/7. If that's part of the plan you'll get a lot more use out of improvements to reliability than improvements to anit-ooze.

If you want to be able to recycle plastic it would make more sense to just include a filament extruder in the recycling machine. That thing is going to have to be big, complicated and dangerous just because it has to shred plastic into little bits. Adding an auger and 3mm hot end will be a minor component of the project. Then you can just use that filament in your (or anyone else's) printer.

On the other hand, the technical challenge would be fun.

makeme Wrote:
-------------------------------------------------------
> For example, in my mind the only reason to go to a
> granule extruder is if you want to print a whole
> lot of plastic because then the savings of not
> paying for the ganules to be turned into filament
> makes sense. It's a scale thing. If you're just
> going to putz around at home you're never going to
> push enough plastic through the machine for the
> savings to appear.

I think the greater reason is that a granule extruder allows you to do your own compounding - getting a custom polycarbonate/ABS blend into a filament may run you over a thousand dollars, but if you have the raw granules already, you can test out all kinds of blends without going broke as there's no need for an intermediate filament step.

---

[haveblue.org]

IMO the main advantage of using filament is that it's a cheap way to get precision. You know that for a filament of cross-sectional area A, advancing it by a distance D will extrude a volume of plastic equal to D x A.

If you make a granular extruder like that, you won't be able to benefit from a geometrical relationship like that anymore. Then you will need another system to control the flow rate.

In my opinion, the most reasonable way to achieve precision using a granule extruder is to first extrude a filament from the granules, and then pass that to the print head.

Quote

For example, in my mind the only reason to go to a granule extruder is if you want to print a whole lot of plastic because then the savings of not paying for the ganules to be turned into filament makes sense.

This is true with the assumption 3 mm filament would be much easier to manufacture than a 0.5 mm one. Without that assumption, producing thick filament is just a waste.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

It just seems like a granule extruder is drifting much closer to an industrial machine than a desktop machine. I mean, turning granules into something else is an industrial process. The up-front investment and ongoing operating/maintenance costs only seem justified if you're working with plastic in bulk quantities. Even running a 3d printer 24/7 won't use up plastic fast enough to justify any economies of scale. It would be like adding a 6-stall garage to your house, but still only using it to park your minivan.

A stand-alone machine might make sense, particularly if it can be made cheap enough. Also, it would make sense if you were building larger objects, and therefore using more plastic faster. But .5mm at 25mm/sec is only 5 cubic mm/sec. That's only .5 Kg of ABS per day (roughly) and even that is assuming 100% uptime.

Just doesn't seem like it's worth the extra complexity.

makeme Wrote:
-------------------------------------------------------
> It just seems like a granule extruder is drifting
> much closer to an industrial machine than a
> desktop machine. I mean, turning granules into
> something else is an industrial process.

Have you seen the granule based rapid prototyper Anna Bellini built as part of her thesis? (page 206: [dspace.library.drexel.edu]) It's most definitely tabletop size, and most of the bulk is actually from the positioners she used. Granule extrusion isn't that complex - web4deb proved that quite handily: [web4deb.blogspot.com]

---

[haveblue.org]

Yeah, I read that paper, and I've seen the growing media extruder, and I even saw some group over in Europe (don't remember what they were called) who had a little funnel on top of a moving granule extruder.

I still think it's too impractical. And the primary reason is still based on the amount of plastic being pushed through the nozzle more than anything else. What's the point of all that added cost and complexity when you can only process a half a Kg per day on your best day? And that's assuming you're using your 3d printer as a production machine, not as a prototyper. And at the moment the only people who are going to buy products made on a desktop 3d printer are people who want parts for another 3d printer. Although, I did see one guy who claimed that these things sell rather well: [www.thingiverse.com]

Again, it makes sense as a stand-alone addition to a 3d printer, particularly if it can handle the array of particle sizes that would be generated from recycled plastic. Maybe you could even mount the granule extruder on the wall above the printer and feed the filament right into the top. Just have it extrude filament at roughly the same speed. At least then you wouldn't have to deal with 1) pushing granules through a tube or 2) moving around a big hopper full of plastic and you could still use the two machines separately.

I think our concepts of 'desktop sized' may be skewed a little as that term can cover quite a range (and I've admittedly been playing a bit of devil's advocate). I seem to recall that Stratasys FDM machines were once advertised as 'desktop', though at several hundred pounds, I really can't take that seriously...

For utter simplicity, compactness, and ease of use, no - I don't think you can beat a filament-fed extruder, especially for people looking for a simple turnkey solution.

But I still think granule extruders have merit, especially for slightly larger machines (still well within the 'desktop' range, however). One big reason is that such a system bypasses the vast majority of Stratasys patents, as they are all filament based. Not a big concern for hobbyists, but if somebody wanted to commercialize a machine, it would offer a bit of protection.

Back to the original question - I don't think a Bowden style granule extruder could be made to work, but the flexible tube for delivering resin from a stationary hopper to the moving print head would be great. If you placed a flexible rotating auger in the tube, you could easily move granules along. Recently I had to do some major repairs on a color laser printer, and discovered that this is exactly how toner powder is transported from the toner cartridges to the developer units.

---

[haveblue.org]

Oh, yeah, that's cool! I've never seen a flexible auger before. [www.dynametconveyor.com]
I wonder how well they work when they're moving around constantly.

Maybe you could build a big box for holding lots of granules, and mount a Mendel on top of it (desktop height). Then a tube could pull granules up from the box while following the print head around. You'd still need a heavy extruder to actually mash the granules through the nozzle, but if you're buying granules in bulk anyway you probably don't mind.

A reporposed CNC machine would be well suited to this converstion. And anyone with one of those would be more likely to be doing enough volume to make the conversion worthwhile.

Perhaps it moves away from a bowden extruder but what if one provides a rigid smooth hollow bore to extrude 3mm through with pressure just enough to provide a consistent just in time (JIT) supply for the conventional extruder.

The granular feed would be fairly slow and some form of active cooling might be required.
You would also be heating the plastic in 2 places.

Maybe coupling your filament supply and working end so closely you can adjust the feed of your primary extruder to the actual diameter of your granular extruder.

just a random thought...