Filament Extruder - Convert pellets to filament

So, like most people, I've been surprised at how much filament costs, compared to raw plastic. For example, 2kg of Makerbot plastic is $110 shipped, but 1kg of raw ABS is $15 shipped. As some of you probably know, this has been identified by others as a problem too. Desktop Factory has put out an offer of $40,000+ prizes for anyone that comes up with a machine that converts pellets to filament.

[desktopfactory2012.istart.org]

Clearly, I am not alone in my desire to use ABS pellets! So, I set out to build a machine to convert these pellets to filament. Here's what I came up with:




Right now, I extrude at about 10 inches per minute, though that can be sped up. I'm kinda taking things easy for now and learning about it. The total power draw is about 40 watts, and at 10 inches/minute, with 1.75mm filament, I'm looking at about 24 hours to extrude 1kg. So, 1kg works out to about 1kwh of electricity, or 12 cents where I live. Not bad! I'm extruding at roughly 185degC. This is a little harder on the motor than a higher temperature, but it's all I can get to right now. I need a higher power heater, something more like 60 watts.

I was concerned that the filament would not be consistent. I measured the first 4 feet at 6 inch intervals with a caliper. Here's what I found:

1.8mm
1.76mm
1.8mm
1.78mm
1.76mm
1.79mm
1.77mm
1.79mm

Which works out to 1.78 +/- 0.02mm. Here's some of the filament I made:



I haven't tried printing with it yet, but I don't think it'd be a problem.

What do you guys think? Open to feedback or comments! I'm pretty excited about this, can't wait to play with it some more!

I'm thinking about selling kits or assembled extruders. Would anyone here be interested?

Very impressive, you should take a look at the cross sections and make sure they are circular. But if you are getting such consistent results then it must be assumed that it is.
Your motivation for this is the price of ABS filament? There are a few sources that are cheaper than 15USD a kilo of spooled filament (granted they are Chinese suppliers), perhaps you can find an even better price for the raw pellets?

There are quite a few of these machines coming out which is good to see, make sure you learn from the other designs and incorporate them into your own. I think we would all be interested in making our own filament, but probably not at such an early stage. If you can prove ultimate consistency and accuracy that rivals professional systems (much like a RepRap ) then I think a lot of people will get on board. If it pans out more like FilaBot, then you will lose credit very rapidly.

Good luck!

yydoctt Wrote:
-------------------------------------------------------
> If it pans out more like
> FilaBot, then you will lose credit very rapidly.
>
> Good luck!

What went wrong with FilaBot?

FilaBot is a kick starter campaign that was funded with hopes of making recycled feedstock in January of 2012. as of now not a single machine has shipped almost 11 months later. to be fair it is complex to make a simple machine to do this, and it was given that money would go to develop the machine. it sounds like they are making progress and have a goal of shipping by years end. they also have a working prototype.

I think the main quarrel is their lack of updates (it must be months between choosing a grinder and showing a near-final prototype, there should have been updates between), I did not revisit their site when posting and see that they are once again making progress!

Looks very nice, I've been wanting to build one to recycle failed prints or PP plastic items.
I looked into the FilaBot but honestly the creator seems to be somewhat of an arrogant self absorbed high horse rider, letting people wait months and months on an update and never replying to emails.

The main blockage for me to keep me from diving head first into a project like this is the availability of a good grinder/shredder. I saw a "mini shredder" on YouTube, but I do not have to resources to cut machine tool steel, and buying it for $300 for the ginder alone is a bit ludacris. I actually wonder how much the FilaBot will end up costing, grinder and extruder.

---

--

Charles S.
Software Engineer
Prusa Mendel I2, RAMPS 1.4, Marlin 1.0 R2, Pronterface, Slic3r

Quote

The main blockage for me to keep me from diving head first into a project like this is the availability of a good grinder/shredder.

This could be a stupid idea, but what about just using a food blender?

crispy1 Wrote:
-------------------------------------------------------
> This could be a stupid idea, but what about just
> using a food blender?

Reminds me of Will it blend?
I'm pretty sure their blender would handle any failed prints!

Anyways, if the mini shedder was this thing, it's total overkill for shredding plastics. That one is a scaled down version of the ones that can shred car engine blocks... It will happily shred metal but will produce way too big bits for feeding into a tabletop filament extruder. I believe that the small filament extruder designs all require granulates that are no larger than 2 mm or so.

Quote

I believe that the small filament extruder designs all require granulates that are no larger than 2 mm or so.

Sounds about right. I can't think of why a decent blender with a sharp blade shouldn't be able to shred plastic down to that size, even if it took a few minutes.

But there are limitations to using recycled plastic to make new plastic. Eventually you burn off all the volatiles and the plastic is not so plastic any more. Injection molding houses typically do not exceed a ratio of 15% "regrind" plastic mixed in with new plastic because the thermal and mechanical properties degrade so much.

I believe the reason is that the blade would dull very quickly. Think about it - knives on wood can't be any gentler, or think about why a weedwhacker uses nylon instead of steel blades. Well that last one is many things (danger, ability to sell consumables) but if a steel system was super effective and lasted a while, you'd at least see it on professional models. That makes me sad, cause I like the blender idea alot. I think a more sturdy rock-crushing/pulverizing type system could be used without failure heavy duty use.

Just need to use a Blendtec blender! [youtu.be]

@greenman100: Will you show us pictures of what's inside the extruder?

Have you already thought about publishing the plans of how to build it or at least a BOM? I am interested in a cheap DIY open source filament extruder which is easy to make.

---

Detlef

Excalibur Hotend

reprapclean32.png

reprapzone.blogspot.de

crispy1 Wrote:
-------------------------------------------------------
> Just need to use a Blendtec blender!
> [youtu.be]

WOW.

>> Is this really new? I thought after all the the showings of videos out there (Lyman and filabot and now this post) that someone released a BOM by now. I've built one too, although its designed for high viscosity pellets. It was very very easy to build, it just took many iterations of learning. I also am not sure that the auger is the best approach since it moves way too much mater too fast and doesn't allow the proper space for temperature to reach the center of the nozzle, this material could look mixed but have the molecular chains mixed in a very heterogenous way (i.e. crumbles or breaks along random flake-like boundaries). To compensate, you'd have to really constrict the end nozzle, slow down the mass, and increase the temp.

If I didn't say this already, I've very impressed by the OP's work and progress. These things are great, and only a few months ago people were saying it is impossible. Shows you want $40,000 can spur (and they haven't even awarded anything).

theodleif Wrote:
-------------------------------------------------------
> @greenman100: Will you show us pictures of what's
> inside the extruder?
>
> Have you already thought about publishing the
> plans of how to build it or at least a BOM? I am
> interested in a cheap DIY open source filament
> extruder which is easy to make.


Soon! I want to finalize the design first.

Looking forward to this...

And with some professional builds of this machine out there we could start to see filament cost drop a little out there. :-)

Beta testing has begun! Here's the details:

[www.soliforum.com]

Is this really needed? The cost of the supplies will come down dramatically with volume. The whole reprap community isn't using the volume to get good pricing yet but will as we continue to grow. ABS pellets are roughly 3 dollars a kilo shipped to your door from a FOB. The problem is that you have to buy 2-5 million pounds a year to get that pricing ;-) Once we have the demand our currently suppliers will get bigger volume discounts the more repraps we make

A kilo of filament for $15 or less? Care to tell me where to buy it?

yydoctt Wrote:
-------------------------------------------------------
> Very impressive, you should take a look at the
> cross sections and make sure they are circular.
> But if you are getting such consistent results
> then it must be assumed that it is.
> Your motivation for this is the price of ABS
> filament? There are a few sources that are cheaper
> than 15USD a kilo of spooled filament (granted >
they are Chinese suppliers), perhaps you can find
> an even better price for the raw pellets?
>
> There are quite a few of these machines coming out
> which is good to see, make sure you learn from the
> other designs and incorporate them into your own.
> I think we would all be interested in making our
> own filament, but probably not at such an early
> stage. If you can prove ultimate consistency and
> accuracy that rivals professional systems (much
> like a RepRap ) then I think a lot of people
> will get on board. If it pans out more like
> FilaBot, then you will lose credit very rapidly.
>
> Good luck!

I can get ABS pellets for $2.50/lb shipped to my door, quantites of 10lbs at a time.

Why wait for hyopthetical pricing to come down? If you use even 15 lbs of filament over the course of your RepRap's life, an extruder has paid for itself.




WICK12345 Wrote:
-------------------------------------------------------
> Is this really needed? The cost of the supplies
> will come down dramatically with volume. The
> whole reprap community isn't using the volume to
> get good pricing yet but will as we continue to
> grow. ABS pellets are roughly 3 dollars a kilo
> shipped to your door from a FOB. The problem is
> that you have to buy 2-5 million pounds a year to
> get that pricing ;-) Once we have the demand our
> currently suppliers will get bigger volume
> discounts the more repraps we make

He's not counting shipping, which is ~$15/kg from China.

waitaki Wrote:
-------------------------------------------------------
> A kilo of filament for $15 or less? Care to tell
> me where to buy it?

Hello!
I has just sawn your machine, has you finally begun to sell the kits/machine assembled?? I am very interested in this, since I have a local 3d printer business.
Glad to contac you!

Hi there,

We're fufilling the ~770 Kickstarter rewards right now, but will start taking orders soon! You can email filastruder at gmail dot com to hop on the waitlist - we'll notify you once we're ready.

Thanks!