Monday, September 11, 2006

 

Making filament the easy way?

I've been pondering on a way to make filament hands-free and here's what I came up with:


The left is 2-3" PVC pipe capped at both ends. Inside is a heating element, probably soldering iron or nichrome. The top has a bike pump nozzle for pressurizing, and a brass nozzle attaches out the bottom at a 45deg angle. The filament drops into the coldwater bath, and once cooled, gets bent upward to a cable spool, which is PIC controlled for a preset level of tension.

I'll bet it still needs tweaking, but it might be a way to easily extrude large amounts of 1-2mm filament.
Any thoughts on how to improve on this design?

Comments:
No problem with the idea. Big problems with the material selection.

PVC's glass temperature is only 87° C
and its melting point is 212° C. You're going to have to get somewhere near 140-150° C to extrude your CAPA at any decent rate. PVC also doesn't transfer heat very well, so you can depend on the temperatures between your heating element and the walls of your PVC extrusion barrel getting a LOT hotter than 140-150° C. Bad idea.

Replace that PVC with steel or brass and you should be good to go.
 
.. Guess I have to take another trip to the hardware store. A 18" x 2"dia steel or copper tube is gonna be hard to come by. We gotta find parts that are cheap and pretty easy to come by, a double-edged sword.

Alternatives are being considered, first idea was a steel thermos, but I don't think it will handle too much pressure before blowing out.
 
A 18" x 2"dia steel or copper tube is gonna be hard to come by.

Yes, it is, unless you go to a proper plumbing store. Mind, steel pipe would do as well except that the steel pipe that they typically sell in hardware stores is welded seam garbage that they can't even be bothered to grind down a bit on the inside. The thing you can get in a hardware store is brass in those dimensions.
 
One problem that you may have is air bubbles in the melt. If you start with CAPA granules, heat them, and squash them, air does get included. For some reason melting the stuff underwater seems to eliminate this problem - paybe it's because CAPA is so hydrophobic.

Another thing you'll find is that the melt dips where it's flowing out the nozzle with the air pressure above, and you get left with lots of melt up the sides and just air flowing out. You could put the CAPA in a bag inside the pressure vessel (microwave friendly plastic bag for high temp?) to prevent this, with the bag forming a boundary between the CAPA and the air.
 
I think that you will find that if you heat a charge of CAPA pellets and then keep the melt at a fixed temperature for about half an hour that the bubbles will rise to the surface and you won't have a problem.

Mind, heating CAPA in water might well have the advantage of suppressing oxidation of the molten polymer surface by oxygen. That could be a real plus.
 
If you're inclined to heating the CAPA in water, could you then use steam pressure to extrude the filament? If so, would a pressure cooker work? They're designed to work at pressures up to 15 psi and have an adjustable pressure valve. Is 15 psi high enough to extrude CAPA? I'm guessing not.
 
Automated filament making looks like a very practical issue, hope to see if some of these ideas work out.
 
Was just looking at your drawing again. What if, instead of placing a bike pump nozzle at the top, you placed a bike tire valve? That way, it would keep it's pressure after being disconnected. You could add or release pressure when you want to...

BTW, most tire valves, under the black rubber shell, have a brass tube the valve is built into, that could be soldered (with hi-temp solder!) or threaded onto the top of the heated canister.

If there is water involved anywhere, you'd want a safty valve somewhere on this gizmo...
 
Was just looking at your drawing again. What if, instead of placing a bike pump nozzle at the top, you placed a bike tire valve?
That's what I was thinking. Sorry if I used the wrong wording on it, but I was thinking the bike tire valve epoxied into the cap, assuming black rubber handles those pressures and temps of course.

If there is water involved anywhere, you'd want a safty valve somewhere on this gizmo...

The one reason we wouldn't want to use water is so we have free control over the temp and pressure without them being intertwined. I like the hydrophobic concept, but baking the CAPA for 30-60min before extruding should have the same effects without safety concerns.


So, if we can find a large diameter copper or brass pipe, suspend and insulate it over the angled nozzle and find some way to keep the filament from buckling or breaking during extrusion, we might have a solid setup from which to make hundred-foot CAPA spools at home.
 
Don't forget heating it. These might help.

http://www.medibix.com/runsearch.jsp?manufacturer=Marathon+Heater+Inc&productsort=manufacturer,+title&category_id=16045

I'm going to call and get prices this morning.
 
I figured a 15-50watt soldering iron element would work for heating it.
 
So how do you plan to integrate it?
 
Got some prices on those sleeve heaters. Two and three inch diameter ones 3 inches wide run US$22-25 while a four inch band heater four inches wide runs $42.85.

Marathon won't sell directly but will refer you to their distributors.
 
Silly question -- does CAPA float? I'd check this myself except mine's still in the mail.
 
So how do you plan to integrate it?

I made a tiny test-extruder and mounted a hot-glue-gun element to brass pipe via very tightly packed aluminum foil. I joined both cylindrical parts with a screw-type hose clamp and it fired up great.

Something similar perhaps would work, especially if it were fairly small design.
 
I'll have to consider gettingn a couple sleeve heaters anyway, for testing.

CAPA without bubbles sinks, BTW.
 
Silly question -- does CAPA float? I'd check this myself except mine's still in the mail.

No, it's slightly denser than water (density about 1.04). It does have a nasty tendency when extruded onto a cooling bath to float by means of surface tension, though.
 
Just stopped by the hardware store on my lunch break, and I guess Plaasjaapie was pretty much right about the selection.

The best I see is 1.5" x 24" steel pipe with plenty of caps and fittings. At 1/2 full CAPA that's 28in^2. Using a seperate air tank, 56in^2.

That's ~450ft of 1mm thread or
225ft for 2mm thread per half-tube of CAPA.

I figure some 1/4" brass pipe through one end-cap for a nozzle fitting and a bike-valve on the other, both hot-soldered into place along with 4 nut couplings to hook up to a wood frame.
 
Reiyuki, if you are still looking for bigger pipe, (1.5 to 3 inches) there is rough (seam welded as Forrest mentioned) steel pipe available with massive threaded ends and caps. It's used for heating oil tanks and such, you can get short bits for cheap at steel suppliers. Usually hidden in the industrial sectors of town.
 
I appreciate the thought, Yvan Roy, but I'm trying to stick as close as possible to parts I can find at the hardware store.
The more common the components used, the easier it will be for anyone else who wants to replicate it (pun intended).
 
The best I see is 1.5" x 24" steel pipe with plenty of caps and fittings. At 1/2 full CAPA that's 28in^2. Using a seperate air tank, 56in^2.

The problem with long, thin pipes is as Adrian said,

you'll find is that the melt dips where it's flowing out the nozzle with the air pressure above, and you get left with lots of melt up the sides and just air flowing out.

There is also the problem of evenly heating such a long container. If you heat it at the bottom near the extruder you'll find that the pipe will most likely conduct heat upwards and melt the CAPA just enough to stick it to the sides without melting it enough to settle to the heated bottom.

Be warned.
 
Maybe a test/prototype could be built with a short 1.5 inch pipe, do avoid complications? I wonder if a freely sliding "plug" of some kind (a loose fitting ball?) could be popped into the pipe on top of the CAPA load, to block air from just pushing down the middle...
 
There seems to be a conundrum with that approach. If the plug is tight enough that it can actually transmit force onto the melt the air entrained in the melt won't be able to get out. If, on the other hand, it isn't that tight it won't move down the cylinder.
 
Sheesh, yes I see... What if it's just 3/4 the pipe's internal diameter, not enought to act as a piston, but enough to prevent a long narrow air well from forming in the CAPA?

I'm reminded of the old trick of placing a floating block of wood in a water filled bucket, so it can be carried without splashing upwards in the middle...
 
Holly Crap, I'm a little frustrated with this blog right now... Here we are, we want to make a rerrap with limited cost, and so called "Hardware store" off the shelf components. Yet at the same time the current deposition material is an expensive, pain in the but to get a hold of, polymer. There is no way that anytime in the next 10 years that I will be able to go down to my local ACE hardware or Home depot and get "polymorph". We are also planning on using electronics that are mail order items, that I have had no luck getting from ANY local electronics stores (Radio Shack sucks more than ever now).
So holy crap when you say that we shouldn't use a good product, because I can’t go down to the hardware store and buy the pipe I need to use to make it. Yet I can order the part in for nearly nothing. I think that everybody using this blog would greatly benefit from looking thru an great general purpose catalog like McMaster Carr to see what commercial components are available. There are tons of products already available to solve so many of the problems we are facing right now. I mean heck, McMaster sells great band heaters for 13.00 (Iteme #3594K95 mcmaster.com). Yes they are for 110 or 240 VAC, but what the heck is wrong with using a simple relay in addition to a transistor? But I mean give me a break, if we are going to say only off the shelf hardware store products, lets get off this polymorph crap and use off the shelf polymer filament. I can go down to the hardware store and get 10 Kg of PVC or HDPE lawn trimmer line for $10. With a product like trimmer line you already have a close tolerance pre-formed filament. The only thing stopping us from using PVC is developing an extruder head that will handle the heat. Lets face it, if we choose some alternate materials, and some new concepts it is far from impossible.
So I'm on a soap box right now and possibly not really making too much sense to anyone but myself. Don’t get me wrong, there have been tons of great ideas that have come out of trying to do the whole project with such limited resources. However I think that the time has come to figure out how this project can be made a reality, I cannot see reinventing the wheel with all the parts of the reprap in the name of being able to get the parts from the hardware store when I have to bend over backwards to get polymorph.
I started this comment intending just to make the comment about low cost heater bands available from McMaster Carr. I have found that McMaster has many products that have helped me many times while designing machines. BTW McMaster sells inexpensive ACME threaded rod and screws. So if anybody gets anything from my off the wall comments, at least take a look at some websites like McMaster that are such wonderful resources to the machine designer. Sorry if I have offended anyone with this comment or stepped on any toes.
 
Good post, jgeorge.
 
Indeed.

I suppose I got carried away with the hardware-store-only concept for building most of the thing. Cheap orderable parts should be applicable, too, espeically since 90% of the electronics have to be ordered as well.

I'm REALLY interested in the trimmer line idea, though. That would be huge if it works.
 
I can go down to the hardware store and get 10 Kg of PVC or HDPE lawn trimmer line for $10.

Can I have the telephone number of your hardware store?
 
Well that estimate on price is quite a bit off from what you can regualy get trimmer line at. That estimate was based of a huge deep discount on trimmer line when my local walmart was remodeling. they were selling trimmer line in about 200ft packs, maybe 150? for $.50. So at that still maybe only 5 kg for 10$, I didn't work out the details, but sure alot cheaper and simpler that making our own filiment.
 
This comment has been removed by a blog administrator.
 
This comment has been removed by a blog administrator.
 
I hate not being able to edit posts.

We might be on a wild-goose chase looking for COMPLETELY store-bought materials. Getting too specific could limit things for some locations as much as it'd make it easier for others. And what if not all trimmer line's HDPE?

I'd be happy to simply order preformed filament instead of make my own, if it's a good enough deal. Apparently they use spools of HDPE for thermoplastic welding.

HDPE only seems to melt at 127°C or so, well below the max of 250° that the PTFE components of the printing head limit it to.

You can get entire pounds of 4mm HDPE filament here
 
//I can go down to the hardware store and get 10 Kg of PVC or HDPE lawn trimmer line for $10.//

(smacks forehead) I never even thought of it. And I was just using a lawn trimmer the other day. I might have to play with some when I get home from work.
 
It might be worthwhile to see if that firm might be willing to special order spools of CAPA. They're down in Santa Barbara and I'm going to be on the ground in Santa Barbara in a few weeks. I think I'll give them a call this morning.
 
Post a Comment

Links to this post:

Create a Link



<< Home

This page is powered by Blogger. Isn't yours?

Subscribe to
Posts [Atom]