The extruder motor and changes needed in extruder robustness

All you need to do is make the stainless steel shaft long enough so that the heat coming down it is small compared to the motor dissipation. Somewhere in the range 50-100mm would do that I think.

---

[www.hydraraptor.blogspot.com]

Hi Vik,

I did not know furnace cement had to be fired to get strong. The best instructions I found for casting concrete made of furnace cement and perlite were at:
[web.archive.org]

In that page John Wasser said someone reported the concrete on its own was slumping, so indeed it seems that it has low strength and can probably only be used as a supported insulator.

From the specifications at:
[www.omega.com]

for high temperature operation, you have to set this brand of cement at 105 C for four to twelve hours depending on the type, to get a final set.

A set of instructions on how to fill areas with furnace cement were at:
[www.unipak.dk]

Cheers,
Enrique

degroof Wrote:
-------------------------------------------------------
> I've got a piece of 3/8" diameter bisque alumina
> on order

Update: Got the bisque alumina on Friday. It's relatively easy to machine but also pretty tough. I cut a piece, drilled a hole through the center and threaded both ends. I attached the heater barrel to it using a threaded rod coupling and warmed it up. It doesn't seem to be a very good thermal insulator, which makes me wonder if it's worth the trouble.

I would've gotten more done but I was distracted by rather nice woman on Saturday, which made concentrating on the extruder nearly impossible. Not that I'm complaining...

Hi Steve,

... it's the same with soap-stone too - the thermal isolation isn't so good as with puffy materials, but it can withstand very high temperatures (much over 1000

I wonder if something like stainless steel tubing might be viable. If I could find some with an I.D. of about 3mm, I might try that. All you'd need to do is cut a piece and thread the ends.

The thermal conductivity of alumina is high for a ceramic, actually higher than stainless steel. From the "Thermal Conductivity of some common Materials" table:
www.engineeringtoolbox.com/thermal-conductivity-d_429.html

the conductivity of alumina ( Aluminum Oxide ) is 30 W/m K. Also from that table, the lowest conductivity safe, solid, not flammable, common materials were portland cement at 0.3, gypsum plaster at 0.5, pyrex and glass in general at 1, brick at 1.3, limestone at 1.3, fireclay brick at 1.4, porcelain at 1.5, mortar cement at 1.7 and sandstone at 1.7.

Soapstone, a mixture of steatite, also called French chalk, is not on that table, and since it's a natural mixture its properties vary. There is a rough value of 2.9 at the bottom of the "Soapstone and Concrete Countertops" discussion at:
[forums.egullet.org]

For fluffy materials, fiberglass and mineral wool blanket have a conductivity of 0.04, fiber insulating board has a conductivity of 0.05. For in an between fiber hardboard has a conductivity of 0.2.

I have not worked with most of these materials, which of them are easy to use and can survive at least 300 C?

Another thing I did at one point was leave open the space normally occupied by the PTFE barrel. (i.e. drive body -> open air -> heater barrel)

In case anyone had any thoughts on doing this, I can categorically state that this will not work at all. It did pretty much what you'd expect. As soon as the filament heated up, it melted down the outside of the heater barrel.

Great thermal properties, though.

Another thought: There are 3 different transistor packages (that I know of) that have a cylindrical shape: TO-5, TO-8, TO18. Their diameters are roughly 8mm, 14mm, 4mm respectively. The heat sinks for these tend to be an upside-down top hat shape that press-fits onto the body. This might make for a cheap, pre-made barrel heat sink.

Quote

Since Ian showed Darwin can take a kilo, there is 0.5 kg leftover which could be used to add an extra 50 hours of granules for a weekend of unattended operation.

It would not be complex to build an open top grain hopper attached to the extruder with a much larger hopper to one side of the fabricator with a dose dispenser. Just drive the extruder into one corner of the fabricator so it hits the dosing device, and gets another 100g of granules dropped in. Or possibly have a long flexible wide bore hose suspended from something attached to the top of the extruder.

I think the best course of action is to try and get a functioning granule extruder sperate of the Darwin, just for making filament, first. After we get one or to that can be made easily(the less milling an lath work needed the better) and from Darwin made parts, we can attempt to combine the two of them. This way we can try getting one to function properly and then try and then later try and limit the weight of extruder.

Edited 1 time(s). Last edit at 04/14/2008 11:19AM by Kyle.

Quote
ronanwarrior
I think the best course of action is to try and get a functioning granule extruder sperate of the Darwin, just for making filament, first. After we get one or to that can be made easily(the less milling an lath work needed the better) and from Darwin made parts, we can attempt to combine the two of them.

To me, the logical iterations are:

1. Standalone recycled plastic/granule extruder developed - this might not be fabbable with a Darwin, but I think that requiring other technology (a CNC mill, for example) is not too great of a detriment in this case. 3mm filament will become a commodity, and if you have one or two "recyclers" in a particular neighborhood/region that will probably serve the needs of that community well enough, if they are properly managed.

2. Port this design to a RepRappable one - this will allow everyone to have a recycler of their own, further streamlining the process. However, I'm not certain that designing a reliable RepRappable recycler is a trivial task.

3. Shrink the design and allow for it to be integrated as an extruder head on the machine itself - I'm not sure whether this is necessary or not. It may be just as easy to have a standalone machine continually recycling, and then just use the filament it generates in a relatively light-weight and simple filament extruder, like the one we use now.

The whole point of having a granule extruder, whether to make 3 mm filament or to use directly, is to be able to recycle plastic.

The question I see is that none of the plastics that are being considered currently are out in the world in big enough amounts to justify a recycling scheme.

I thought it was also because it is much easier to buy PCL and PLA in granules than it is in filament.

---

[www.hydraraptor.blogspot.com]

All you have to do to get PCL and PLA into 3 mm filament is to buy some sacks and have 'em shipped to New Image. They charge about $3-4/lb to convert it into filament. The problem is that in order to avoid getting killed on shipping costs for getting the plastic to New Image you need to move quite a bit, like a palate of the stuff. That means a fairly substantial front end cost and inventories of the filament and resin.

It's a revenue flow problem, not an economic problem per se. Once you get a few hundred Darwins running, the problem will evaporate.

With recycling, otoh, you not only need an extruder but also something to chop recycled plastic into small enough bits that it will go into the extruder. That isn't nearly as easy as it sounds. Plastic is a right bastard to grind.

Edited 1 time(s). Last edit at 04/14/2008 02:46PM by Forrest Higgs.

I was trying to come up with cheap ways to recycle plastic into 3mm filament. Had this vague idea of taking an electric deep fryer and drilling a hole in the bottom. The nice thing about them is that most have an adjustable thermostat and are made of cast aluminum (easy to drill).

Not sure how to force the plastic out consistently, though. I've got a soft-serve ice cream maker that has a mixing blade with an auger in the middle. Maybe something similar.

I forget who had the idea, it was pretty recently. I think that it was Forrest. I think he mention that you could use a weight and a surface area on a piston, and then you'd be able to know the pressure coming out of the nozzle would be constant.

---

Jay

Thats how the ASTM measures viscosity in molten plastic, more or less.

Commercial plastic recycling does not grind but chops the plastic into granules size chucks using a fast spinning blade like a hand held liquidizer blade, must try ours when the wifes out

---

Ian
[www.bitsfrombytes.com]

I will deny all knowledge of any specific event if asked, but they blend their own plastic accessories quite effectively!

Vik :v)

Yes, it blends!

[www.youtube.com]

(And so does glass marbles ( [www.youtube.com] ), and a lot of other things. )

I'm in the process of rebuilding my extruder (yes, again). Current state is:

- PTFE barrel is 1" dia and 3" long
- heater barrel is also longer to allow more of it to be threaded into the PTFE
- still using a pipe clamp around the PTFE
- cut a new extruder body out of a plastic cutting board (the old one was warped)
- wider PTFE barrel allowed me to screw it directly to the extruder body
- ran a very long drill bit through the entire assembly to ensure proper alignment
- used bare nichrome by coating the heater barrel in JBWeld, wrapping the wire, then more JBWeld
- wrapped the whole heater barrel, all the way up to the pipe clamp, in insulation

Preliminary tests have been good. The heater isn't getting as hot as I'd like. I may need to shorten the nichrome slightly. The good news is that the top of the PTFE barrel was cool to the touch.

Next step is to get the temperature up and re-profile the thing.

Currently distracted by real-life stuff, so it may be a while.

would a really long woodscrew with the threads machined or ground down in certain areas work for an auger?

It might well, if you can keep it centred.

I wondered about coach bolts also?

Vik :v)

A note for those of you looking at furnace cement and other refractories. The general rule of thumb with them is to use the lowest possible temperature rating you can. I'm not certain how the strength of the material varies, but the thermal conductivity goes down with reduced temperature rating. Higher temp refractories may last longer, but you need a thicker layer to insulate to the same degree.

For the current design: the extruder's flex drive to screw connection breaks all the time. I've had it wear out four times now. I'm really looking into a better fix? Anyone tried anything that works well for longer? I'm thinking of epoxy instead of solder, but I'm certainly not sure whether that will do it... I'm also wondering why the drive needs to be flexible, since my filament seems to bend without any problems with this radius. My plastic is HDPE, thought I'm going to try ABS once I have it. Is it an option to remove the flex drive out of the equation?

---

Regards,

Erik de Bruijn
[Ultimaker.com] - [blog.erikdebruijn.nl]

I don't have any problems now with the connection, either soldered, of JB Welded, but the wire itself wears out eventually. I am on my fourth piece of cable I think. If it is soldered it is not too big a job to replace it, as I had to do last night.

You don't need it for HDPE, ABS, PCL or PLA. They all will bend enough. I intend to revert to the previous design of the extruder, they will be the next parts I print. All that changes is the motor bracket and the coupler I think.

If it works I could make you the parts if you want.

---

[www.hydraraptor.blogspot.com]

I have been using the non cable design from the beginning and it works well the biggest problem I have had though is is the connector from the motor to the drive screw my first ones were using flats on the screw the new one I got from Zach, has a nut on the screw. This has worked fine for ABS, CAPA and HDPE with no issues. The flex cable needs to be used on the latest geared down extruder.



Bruce

> If it works I could make you the parts if you want.

That's a very nice offer. I'll try something out of wood that may work well. I don't want to take up much of your time because I believe you're doing very important work! Not that I don't have big plans for RepRap myself... on the other hand, if you make me an extruder and I get this one working as well, I'll have RP parts for an extruder ready for someone else. I intend to build RepRaps for friends anyway. For now I hope you can put time in getting the same print quality with the Darwin design or changing the Darwin design as much as needed to get there. Your quality seems comparable to commercial FDM already. That a huge accomplishment, I'd really like that to be the default print quality for a RepRap that anyone can have! If you make a Darwin you'll have HydraRaptor available as CNC machine again. Nice to have some fabrication diversity, I'd think.

---

Regards,

Erik de Bruijn
[Ultimaker.com] - [blog.erikdebruijn.nl]

I just Got thinking a bit about the current flex drive system that we have and one of the bigger problems is that the JB weled and solder can't hold the wire in place so what if first the the drilled hole in the drive shaft was tapped and given left-hand thread. So that as the wire spins it would naturally force it self deeper into the hole. Having the holes threaded would allow for backlash without having to have the solder or JB weld taking on that stress.

I was looking around for some left handed taps and found this place [www.victornet.com] so left handed taps do exist.... well this is just an idea. I would love to personally test it out but have yet to get started on my reprap.

... another idea: - crimp short copper-tubes at the ends of the wire, then you can easy solder the copper-ends ...

Viktor