Changes that need to be made to the extruder

For insulation what about asbestos don't know if it will machine but it is a very well understood material been used since Roman times

---

Ian
[www.bitsfrombytes.com]

I think machining asbestos kills you doesn't it?

---

[www.hydraraptor.blogspot.com]

I just realised, my post about trying to extrude aluminium through an insulator is nonsense. The AL will conduct the heat past the insulator.

---

[www.hydraraptor.blogspot.com]

Won't aluminum will erode the nozzle, regardless of the material?
Probably more efficient to do lost wax casting.

Maybe you should consider building it up like MIG welding?
[en.wikipedia.org]

or this stuff:
[www-rpl.stanford.edu]

They're all a bit beyond us right now.

Laser sintering is probably the best way of doing AL I think.

---

[www.hydraraptor.blogspot.com]

nophead

Thats is pretty funny. I think that the only way to prevent the heat transfer would be to heat the aluminum up fast enough so that you could extrude it at a high enough speed that the aluminum is gone before the heat gets transferred along the drive train. Kind of the way spitting fire works. If the performer left a slow stream of alcohol, they would get burned, but if the performer gets the alcohol moving fast enough, than the fire does not have enough time to move the distance to the performer. (The heat, *fire*, has a constant transfer rate, if you are able to overcome this rate by moving the filament faster, extrusion is possible)

---

Jay

Yes, but what happens when you want to stop? One end of the AL is at 660C, the heater is off but that heat will travel back up the stock. It will reduce in temperature as it travels though.

So we need to do a calculation with the thermal conductivity and specific heat capacity to work out how fast the heat will travel to see if it is practical.

---

[www.hydraraptor.blogspot.com]

Hi nophead,


... then back again to soapstone - heatresistant until 1600

Edited 1 time(s). Last edit at 01/21/2008 01:17AM by Viktor Dirks.

What if after extrusion you lift the extruder and snip off the filament above the heater barrel? No heat to marginal heat could be transferred anymore. Especially if you use soapstone as vick is suggesting.

---

Jay

nophead Wrote:
-------------------------------------------------------
> Yes, but what happens when you want to stop? One
> end of the AL is at 660C, the heater is off but
> that heat will travel back up the stock. It will
> reduce in temperature as it travels though.

Would it help any if the aluminum were discontiguous? Say you used aluminum powder and gradually dropped small amounts into a heating chamber. You wouldn't have any pressure behind it, though. Not sure if the weight of the molten metal would be enough to get it to extruder.

well if you add a plunger and insulate the plunger shaft enough, a constant pressure could be applied.

---

Jay

Cheap! Wrote:
-------------------------------------------------------
> I've been extruding HDPE, and I need keep the
> temp. near 250C. The thermister (I am told) is not
> that accurate at those temps. I am told that I
> might see temps as high as 280C before the
> thermister reports back a change. What are you
> using to achieve that +- 3C in your extruder?


The generation2 electronics have been improved since we tried this in Houston. I have not tried the improvements... but I understand this was mostly a software / logical issue. The Thermistor is OK, but how we read it was a bit of a problem in the 200+ range.

And the recent development of a gear for the motor driver may allow us to use pressure more instead of heat. At least that seems promising to me... as HDPE seems to extrude better when I physically shove it down the extruder

// Stephen Gutknecht
/// Canyon Lake, TX

Just make the extruder ceramic and extrude aluminum like we do HDPE now. No worries about heat as long as all the parts are heat tolerant to a greater extent.

Demented

>And the recent development of a gear for the motor driver may allow us to use
>pressure more instead of heat. At least that seems promising to me... as HDPE
>seems to extrude better when I physically shove it down the extruder

With my extruder the limit is not down to the motor, it's the grip on the HDPE. I.e. if it jambs the motor does not stop. What happens is that the thread cut into the HDPE strips and I have to push it until a fresh bit of HDPE starts to bite.

I have also found that the temperature makes little difference to the amount of force you need between about 160C and 240C. Below 160C it rises rapidly as you approach the freezing point. Much higher temperature (when you set light to it) does make HDPE drip like candle wax so it does continue to get less viscous, just not very rapidly.

What is important about the temperature is whether one layer welds to the next. If the nozzle is moving quickly you need to extrude at (Tm x 2 - Ta) when Tm is the melting point, i.e. 130C and Ta is the temperature of the previous layer. If the object is large enough it will have got back down to room temp so I currently extrude at 130 x 2 - 20 = 240C. You can get by with lower, e.g. 200C if the object is small enough that you do the next layer before it has cooled down, or if the nozzle is moving slow enough that it conducts and radiates extra heat.

I am hoping a heated table will let me drop the temperature and reduce warping. E.g. keeping the object at 80C would let me extrude at 180C.

---

[www.hydraraptor.blogspot.com]

nophead,

Perhaps on the first few layers that would be true. However, once you get above the table a ways, the temperature is sure to drop off and you are back to the problem first encountered. I think just solving the general physics of how the warping occurs would be a "better" solution. Admittedly, this involves a shit-ton of work and is in no way easier, just cleaner in an abstract sense.

Demented

I don't think there is a likely to be a solution to the warping except physical restraint. This is exactly the problem you have when welding steel when the filler cools it shrinks and "pulls" the welded item with a surprising amount of force.

This is a very well understood problem and the only solutions I'm aware of are physical restraint with bracing or welding form both side at the same time (difficult for us to do!)

I think nophead's current solution with a chopping board is a very good approach and if the support material could also be HDPE with a lower weld strength then it could possibly be able to do overhangs. I also in no way underestimate the software challenges to achieve this.

---

Ian
[www.bitsfrombytes.com]

Welding from both sides... good idea! Print 5 layers, remove the part, flip it upside down, print 5 layers, flip it, etc. Bit more work, especially the alignment, but it should stop warping.

The best way to do this would still be using a raft, and a box like jig to fit the raft.

--Blerik

How much the temperature drops off will depends on how good a conductor HDPE is versus the convection/radiation cooling. The fact that the table is hot will cause hot air to rise all around the object and the newly deposited layers are adding heat as well.

It might not work but it's easy to do so it's worth an experiment.

Computing the warping would be theoretically possible, but I am not sure compensating for it by extruding an inverse warped shape is always possible. Even the simple case of making a rectangular block would require support material to get a flat base.

Edited 1 time(s). Last edit at 01/21/2008 06:00AM by nophead.

---

[www.hydraraptor.blogspot.com]

Far easier to put down a paste of aluminum-powder and some liquid or wax like binder an then and sinter the whole object in a furnace.

Regarding laser-sintering of metallic powders, I believe the metal particles are coated in some kind of laser-susceptor (absorber, so that the metal heats up) or a flux perhaps, to prevent oxidation (low speed or 'my god, the printer is on fire'). I don't think it is as simple as spreading out a bed of powdered aluminum.

Hi Sebastien,

... i'm sintering and melting gold-filled paste on goldpads with a selfmade PWM-driven 4Watt-diodelaser - it's steaming and fuming heavy, when drying and then the gold went fluid and connect to the underlying pad ...

It's a neat chemistry for the adhesion-additives for good contact of the dispensed droplet to the pad when heating, as the droplet is nearly 80 microns wide and only 30 microns in height and sometimes tend to form a perfect sphere instead of staying down on the pad :-/

When only sintering to a foam-like structure, then it's no problem to make it so - then you have coated particles and melt only the coating, not the core ...

But when you want to have a strong and homogenous structure and melt the metall to solid, then the process isn't so stable and you have much shrinkage through the vanishing fluid too.

Anyway, i'm experimenting on it, but actually i have only limited materials (gold and platin) and limited time too on job -- and only a 1Watt-diodelaser at home ...

Viktor

Hi all,

... here i have a sample of milled soapstone:


It's very dense quality, usable for kitchen-tables and stairs.

I milled it with a 3mm hardalloy-mill without any coolant at 24000 U/min - the (old) mill didn't seems to wear out.

Next i have to mill a bigger plate with over 40 inserts for firing parts in an oven - then i'll use a new mill and can test, how long it lives ...

So i can make any 3D-form in soapstone, it can be lathed, drilled, cutted and so on.

Maybe it's a good alternate for PTFE-parts, when you need higher temperatures and good accuracies in dimension - soapstone stays hard until over 1500

interesting. i'd love to see a redesign of the clamp simply in soapstone. we could probably do without the soapstone at all. i wonder if emachineshop can handle it?

Hi Zach,

... i didn't have a lathe and my greatest mill is 3mm in diameter and 10 mm cutting length, as most of my machined parts are very, very small

But maybe i can do the trick with the mill-lathe, as in this video: [www.youtube.com]

This week my millhead for the home-CNC should arrive, so i can 'play' with some sideway-technologies ...

Viktor

I got some (144) soap stone bars from an art supplier and I just knocked this up with my watchmaker's lathe.



It is the softest thing I have ever machined, more like machining soap than stone. I think it may be too weak to use as an extruder but it I am planning some heater tests so I can evaluate it's thermal properties.

I think Victor's example must be a lot harder, you could not use this stuff for a counter or a stair. Where did you get yours from Victor?

Edited 1 time(s). Last edit at 01/28/2008 04:14PM by nophead.

---

[www.hydraraptor.blogspot.com]

Hi nophead,

... i ordered my parts here: [www.speckstein-winter.de]

It seems to be a very dense/heavy/hard-grade-soapstone from Brasilia or India.

It looks like normal dark-grey stone and has polished a touch/finish like granite, but i can carve it with a knife ...

Viktor

My sample is so soft I can carve it with my fingernail! Still it might work. I have 143 more tries to get it right

---

[www.hydraraptor.blogspot.com]

Hmm, I just found out it's thermal conductivity is 13 times higher than PTFE

---

[www.hydraraptor.blogspot.com]

Hi nophead,

... yes, the thermal conductivity is higher than with PTFE, because it's much denser.

Prior my selection of soapstone i found some CNC-macineable ceramics with tifferent composition, strength and thermal properties - [www.moeschter-group.de]

I ordered some samples of DOTHERM-700 for milling-experiments and got some samples of other types too ...

Some of the materials are softer or with lower thermal conductivity, but it's much expensiver than soapstone and for my applications it wasn't so a differencce, so i selected soapstone.

I think with an optimized layout there should be no problem to use it for the hot parts in the extruder-head.

Maybe i'll design some variants for my IR- or laserdiode-sintering, but this can stay a while ...

Viktor

Regarding metal extrusion, you might want to check this out: [engr.smu.edu]

As for redesigning the extruder, why not make it a lot simpler and not use any rotating parts? Just use a copper pipe, put plastic in it, put a nozzle on one end and a connection to an air pump on the other, heat up the end with the nozzle and have the air pump push the plastic out.

In fact, it might even be possible to print out the entire air pump. Most aquarium air pumps are just a 2 valves, a membrane attached to a magnet, and a coil of wire.
[www.candyfab.org]
All things that reprap could easily make.

The advantages of this design is that it can extrude a lot more than filament plastic. In fact, it might be able to extrude plastic powders. You could grind up misprints into powder and recycle them. You could also use it to extrude pastes by turning the heat off.

The only problems I see with this design is the fact that it has to be refilled for large prints and getting it attached to Darwin so the heat conducted by the pipe doesn't get to any of the plastic parts.

I have read through some of Vik's old blogs on cement/concrete. One thing I missed was a final writeup of the advantages and disadvantages of concrete vs PTFE. Concrete seems like a great material to work with and easily acquired anywhere in the world. The disadvantage seems to be weight and tensile strength.

So why did reprap go with PTFE vs concrete? Will concrete rise again?