Why are filament spools so expensive compared to pellets?

I don't think there is much quality difference between cheap and expensive. I am quite happy with the filament I've got, which is pretty cheap. It performs similair to the more expensive filaments I've gotten. I don't think a dual head is going to improve that.

Also I don't drink much milk. Not that I can actually buy milk in plastic bottles in Holland. I guess it's time to start growing my own corn :p

---

---

reprapworld.com
PLA, ABS, PETG multiple colors (3 and 1.75 mm) €16,49 / kg
Megatronics 3.0 €79.99 / Minitronics €37,19

It is only the sparse infill layers that are not critical on the volume of plastic. Solid infill layers are more critical than the outline.

---

[www.hydraraptor.blogspot.com]

I think the most viable option is making an own dedicated filament extruder since one machine could be used to provide filament for multiple repraps.

Lets face it, no matter what you do the pellets will be the cheapest option no matter where it comes from. (industrial, recycled, or selfmade) it will always come in some sort of granular form. A granular extruder will always be more complicated, heavier and more expensive unless one has a genius concept on how to extrude the pellets one by one.
Multiple extruders could work one after another while others are reloading.
But any mechanics I can come up with that could accomplish that would be overly complicated, but maybe someone has an idea.

On the "cheaper, less quality" filament option: I don't think you could obtain anything of this in the near future...

From my perspective the thread has done its purpose. It's pretty much clear now what I suspected in the first place. Too many merchants have to large profits. (Or not if you buy and sell any goods yourself )

Edited 1 time(s). Last edit at 07/06/2011 07:36AM by ElectricMucus.

ElectricMucus Wrote:
-------------------------------------------------------

> heavier and more expensive unless one has a genius
> concept on how to extrude the pellets one by one.
>

ooh, I Have now!

ablainey Wrote:
-------------------------
> ooh, I Have now!


Well, thanks.
I actually thought of a piston driven mechanism either fed using air pressure or gravity. The problem is even if you squash one pellet while melting you could still get air coming out of the nozzle.
And we would have to completely get rid of stringing and ooze if we run two or multiple extruders in rapid succession.

The air bubbles could probably removed by providing an escape path for it and a lot of try and error,

but for the stringing... I have no idea.

Edited 1 time(s). Last edit at 07/06/2011 12:22PM by ElectricMucus.

ElectricMucus Wrote:
-------------------------------------------------------
> ablainey Wrote:
> -------------------------
> > ooh, I Have now!
>
>
> Well, thanks.
> I actually thought of a piston driven mechanism
> either fed using air pressure or gravity. The
> problem is even if you squash one pellet while
> melting you could still get air coming out of the
> nozzle.
> And we would have to completely get rid of
> stringing and ooze if we run two or multiple
> extruders in rapid succession.
>
> The air bubbles could probably removed by
> providing an escape path for it and a lot of try
> and error,
>
> but for the stringing... I have no idea.


Greats minds and all that. I was also thinking a piston but the feed pellets being in a vacuum chamber to stop any bubbles.
Basically a sealed hopper with an auger at the bottom (basic meat grinder type set up) a vaccum pump drawing out the air. The pellets are forced along a tube which pre heats them. At the end of the tube is a small ram piston which smashes the pellets down into a tappering die tube. The die tube is heated and the ram is driven by motor and crank. The tappered die is cooled at the lower end.
Cheap and slow but it should work.
Or the setup could be turned 90 degree so the filament comes out the side.

My worry is that there may be lamination in the filiment where the pellets are merged. Say from surface contaminants like oils. So the pellets might need a solvent wash and dry before going in the machine? Or the die tapper could be wider to smash a few pellets at a time.

ablainey Wrote:
-------------------------------------------------------
> Greats minds and all that. I was also thinking a
> piston but the feed pellets being in a vacuum
> chamber to stop any bubbles.

The problem with vacuum pumps is: they are expensive. But if you design the system in such a way that you only need one pump even with multiple extruders it might be worth it. and I am unsure if there is an economic and quiet solution.

> Basically a sealed hopper with an auger at the
> bottom (basic meat grinder type set up) a vaccum
> pump drawing out the air.

The problem might be that once the piston is released you are sucking air back though the nozzle along with the molten material. (Just look at mercury barometer...)

> The pellets are forced
> along a tube which pre heats them. At the end of
> the tube is a small ram piston which smashes the
> pellets down into a tappering die tube. The die
> tube is heated and the ram is driven by motor and
> crank. The tappered die is cooled at the lower
> end.

It might be difficult to control the heat that the pellets do not melt prematurely and clog the tube... but otherwise it should work.

> Cheap and slow but it should work.
> Or the setup could be turned 90 degree so the
> filament comes out the side.

Would probably be best, since the mechanics sort of demand it..

> My worry is that there may be lamination in the
> filiment where the pellets are merged. Say from
> surface contaminants like oils. So the pellets
> might need a solvent wash and dry before going in
> the machine? Or the die tapper could be wider to
> smash a few pellets at a time.

I don't think this will be necessary, at least in with. The way I thought of it is to always have a pool of molten material at the bottom and add new pellets on top over time, squash them a bit, add more and so on.
The advantage of a wider piston would be that the frequency of where the extruder would have to be stopped will be lower.
The optimum might be that a full piston can draw at least one outline or possible a full layer of your average object.

I have been trying to find how commercial machine get rid of the air and so far I'm not sure.

Does anyone know or is the air removed by the screw pump process itself?

---

Make your Mendel twice as accurate.
[www.thingiverse.com]

In all the bigger pellet fed machines that I have worked with start to melt them in the hopper. So I guess if you do it that way air bubbles are not a big issue. For something at this scale. I am not too sure the best way would be.

---

[mike-mack.blogspot.com]

Madkite Wrote:
-------------------------------------------------------
> I have been trying to find how commercial machine
> get rid of the air and so far I'm not sure.
>
> Does anyone know or is the air removed by the
> screw pump process itself?


I always thought so...
at least I know of no other way.

The spaces between the screw threads are chambers where the material gets compressed and compacted while air is passed along the opposite side. The pressure is higher in each chamber so the air has a pressure gradient to follow.

I like that idea. Have you made any attempts to make a working prototype?

zekebird3 Wrote:
-------------------------------------------------------
> I like that idea. Have you made any attempts to
> make a working prototype?

Not yet, but which idea are we talking about?

It is possible that vertical supply chains give a vested interest for filament to be charged at a higher price. Basically monopoly systems by price fixing would be an example.

If you were a plastic manufacture that made filament, and you also made chess pieces, dishes, and legos, would you want a cheap filament on the market? And many of the big corporations discuss issues with each other on such topics. There would be a resistance to expand that market to lower price inside that industry.

Making a home made Plastic Filament from pellets seems really easy, and cheep. Could be used to recycle any ABS plastic by remelting also.

A hopper with two sections, bottom section vacuum with pellets, the other part pushes down on pellets with a press or weight.
Use a reversed bicycle pump to make the vacuum
A cone under hopper with a stir capability, heated to melt pellets, inside cone have slats of copper to help transfer heat to inside of cone.
A copper tube under that to heat a little higher to get good viscosity.
A nozzle that puts out 3mm strand
Then another copper tube that makes sure it is 3mm
And finally a cooling tube.
Wind it all on a fishing reel like winder.

The hopper could be any pot, cut the bottom out to make a washer to make two sections in pot, use a bicycle tire for the edge of that gasket after adding some ring around the cut out section to make the surface area from that washer and inside of pot higher surface area, or make one out of sheet metal. Use the glue they use to stick automobile tires to rims to hold rubber to that washer.
Some mixing tube could be put through with a seal through the middle of that washer down into the cone where plastic is melted to get color consistent from color parts put into hopper.

Attach a copper shaped cone to bottom of hopper, adding in a car gasket or high temp plastic to insulate heat. Copper cooking funnel would work, or make one from sheet metal.
Add fins across inside of copper cone to increase heat transfer.
Put heater on the copper shaped cone
Attach copper tube to bottom of cone
Put heater on that also

Attach some valve that can be turned on when proper temperature and pressure is reached to assure consistent product
Attach another copper tube with 3mm inside diameter to make sure it is 3mm exactly heated also to make sure strand is consistent
Attach a thermal insulator also 3mm
Then Attach a final cooling copper tube, make it so that it takes a few bends at the end, that force will create a back pressure on filament up into valve making sure all filament is solid without thinner filament sizes in parts of strand. By adding any form of back pressure on final output area that will insure the entire 3mm section of heating tube, and cooling tube is solid and diameter of tube is full making a consistent size product.

Wind it on a reel, although reel should not pull object out, and possible could actually be used to add back pressure on filament if bending tube to add back pressure is not used. Making the filament do some of the work of turning a reel could add some back pressure also.

Either a pot with bottom cut out, or a sheet metal hopper with a couple supporting rings around outside could be used for hopper
The cone could be a copper cooking cone, or make your own by wrapped sheet metal.

If apparatus cooks at below 260 celsius JBWeld could glue items together
Or the items could be bolted together, gaskets can be created by cut out from gasket sheets sold in various auto shops, or higher temp rubber or plastic could be used for gaskets.

Seems really easy, if I had a few hundred dollars I would build one.


Robert Eastwood.

Or more accurately, if I had a 3d printer, and few hundred dollars, I would build one.

Then again, if I had a few hundred dollars, I would also build a web server with a peer to peer ability to share the many diagrams that should be available on Internet to make various items. There should be diagrams of 1000s of item schematics sold for a few pennies each download, and accessible from search engines from any Internet connection.

And not just toys but anything people use, with links to any non plastic parts added in,
As an example someone making a tape dispenser would download the 3d model, and have a link to a site that sells the little metal cutter piece at end of tape piece that would also be needed, and a link to any glue that would be needed to build a part. Or a link to how to cut a soda pop or tin can to make that piece.

And by having a few pennies charge for each item downloaded, instead of selling the product, the diagram would be sold by some cheap download price. The transportation cost of products would be replaced by the transmission of diagram to make object. Note that is effectively removing transportation cost, and for inanimate plastic items effectively a form of teleportation of an item

Interestingly the 3d printer with accessible schematics, is effectively the same concept as a teleportation device.

Robert Eastwood.

Couple of Questions:

Is Air less dense than the molten ABS or PLA?

If so and the Auger was vertical would the air not try and rise out of the melt chamber? Obviously viscosity would come into play.

It is but with very viscous liquids a vacuum is needed to get air bubbles out.

---

[www.hydraraptor.blogspot.com]

But does the pressure at the nozzle not create enough deltaP to encourage the air back to the top?

Or is it that the melt chamber is held in vacuum?

I don't know how the air is removed in commercial screw extruders. When casting silicone for example a vacuum is use to remove bubbles. I think positive pressure would have the opposite effect.

---

[www.hydraraptor.blogspot.com]

Call me crazy but i didnt really think the pricing on the filament was crazy high. I manage a 20+ machines CNC machine shop and purchase plastics all the time mostly for parts going into the food industry. Plastics are a very expensive material to start with compared to metals, and when you add one extra operation to get from bulk plastic pellets to high tolerence filament the price is probably more than doubling. Thats before it even reaches the hands of the distributors who buy large volume and stare at it on their shelves until it sells. Its marked up at least 3 times and if you are buying from a dealer rather than a direct distributor its 4 times marked up. Everyone in the chain must make a profit and they all serve a purpose.

There are only two things i can think of that would bring the end price down. If the market increases in size substantially thus increasing production for the manufacturers and sales for the retailers. When you can sell more units you can afford to sell them at a lower and lower cost. And if you can cut out one link in the supply chain, but that takes a lot of revenue up front to put into inventory and most people in the "hobby" industry dont have access to that kind of capital.

nophead Wrote:
-------------------------------------------------------
> I don't know how the air is removed in commercial
> screw extruders. When casting silicone for example
> a vacuum is use to remove bubbles. I think
> positive pressure would have the opposite effect.


Why would you post you are unaware of how to remove air from the process after I already explained how to do that.

Bob Eastwood
Robert Eastwood

Sorry I missed that. Where did you post how air is removed from commercial screw extruders?

---

[www.hydraraptor.blogspot.com]

nophead Wrote:
-------------------------------------------------------
> Sorry I missed that. Where did you post how air is
> removed from commercial screw extruders?


[forums.reprap.org]

Create a vacuum to cook the plastic,that removes air before cooking, apply a press after reaching temperature that pressure being above 1 atmosphere, then open exit valve, and have back pressure on filament and there is not air bubbles.

vacuum out air, melt plastic, apply pressure, open valve. No air in mix. Back pressure on filament would help also, a feeder to limit exit speed.

Although honestly using a screw to push liquid is not as effective as pushing it with pressure, think tube of tooth paste. The screw device basically tries to create pressure anyways, why not pressurize the cooking pot.

I haven't tried it, but I have been told that if you apply pressure to molten plastic with air it tends to tunnel trough and find the exit. It needs a piston to work I think.

---

[www.hydraraptor.blogspot.com]

nophead Wrote:
-------------------------------------------------------
> I haven't tried it, but I have been told that if
> you apply pressure to molten plastic with air it
> tends to tunnel trough and find the exit. It needs
> a piston to work I think.


And the post I posted explains how to make that press system. The pressure is applied by a press in a cylinder after air is vacuumed out and plastic is cooked. The pressure is not applied by air, as posted above in this thread, I explained it is done by gravity or press force on the molten plastic.

http://en.wikipedia.org/wiki/Plastics_extrusion

Good overview description of what happens to the air on wikipedia.

Making a custom profile auger will be the challenge.

How about as a mad idea:

Rotary piston pump ie several cylinders that feed a pellet and squeeze it into the melt chamber?

Would be a pig to get working.

If you make the cylinder say 1L then you could extrude about 1kg in one cycle and that would make a reasonable amount to use on a printer. So unless you want a continuous cycle to be able to make enough to sell you don't need the complexity of a multi-stroke machine. Just a heated cylinder and a piston.

---

[www.hydraraptor.blogspot.com]

nophead Wrote:
-------------------------------------------------------
> If you make the cylinder say 1L then you could
> extrude about 1kg in one cycle and that would make
> a reasonable amount to use on a printer. So unless
> you want a continuous cycle to be able to make
> enough to sell you don't need the complexity of a
> multi-stroke machine. Just a heated cylinder and a
> piston.

I have a spare Pneumatic cylinder and 5 port valve about that size, I wonder if the o-rings are silicon? - if so the whole thing could be heated and with a nozzle extruded, ahhh it all sounds so easy, I expect it's not.

---

[richrap.blogspot.com]

nophead Wrote:
-------------------------------------------------------
> If you make the cylinder say 1L then you could
> extrude about 1kg in one cycle and that would make
> a reasonable amount to use on a printer. So unless
> you want a continuous cycle to be able to make
> enough to sell you don't need the complexity of a
> multi-stroke machine. Just a heated cylinder and a
> piston.

Still have the air problem, so you wouldn't get consistent flow.

As they list in the wikipedia, why use a heater, just use the cylinder pressure to heat the plastic.

Actually it says

Quote

if an extrusion line is running a certain material fast enough, the heaters can be shut off and the melt temperature maintained by pressure and friction alone inside the barrel.

I doubt the friction from pushing raw nurdles down a cold barrel is going to get them melted in any manner of timely fashion. You probably still need or want a heater to get the process started.

Quote

nophead Wrote:
-------------------------------------------------------
> If you make the cylinder say 1L then you could
> extrude about 1kg in one cycle and that would make
> a reasonable amount to use on a printer. So unless
> you want a continuous cycle to be able to make
> enough to sell you don't need the complexity of a
> multi-stroke machine. Just a heated cylinder and a
> piston.

Still have the air problem, so you wouldn't get consistent flow.

Air bubbles have a tendency to climb up in a liquid, so an exit hose at the bottom of the cylinder will receive bubble-free liquid.

Quote

As they list in the wikipedia, why use a heater, just use the cylinder pressure to heat the plastic.

Unless I'm missing something, this works with a gas, but not with a liquid.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY