Plastic welding rod produciton line

follow up to: [forums.reprap.org]

I contacted the company selling these machines and got a little more information.


Capabilities:
Default configuration produces filament from 2mm-8mm diameter in a variety of shapes (round, square, triangle, etc) and there are several other molds available for larger or smaller diameters and shapes.
Processes PP, PE, and PVC pellets (only these listed, but a few others are probably possible)
It's extrusion screw is "45/30" (what does that mean?)

Dimensions: 15m*1.5*2.5 (L*W*H)=56.25m3

Price: USD 25,538 (wow... that's a small house... or a decent car...)

I also got some photos of a sample 4mm rod. (looks good!)



If there's anyone really rich, with an empty warehouse handy.....

Otherwise, is is possible to rig something like this ourselves?
Of course it must be smaller, cheaper, and have fewer capabilities.

direct link:
[xfjx.en.alibaba.com]

Feel free to contact them yourself. Speaking Chinese might help. The contact I've been talking to seems to be using google translate...

---

David

[www.waynemachine.com]

wow, that stuffs really big...hmm, bad news for us home filament makers...at least says how difficult the process is gonna be. Blah.

Demented

Demented Chihuahua Wrote:
-------------------------------------------------------
> wow, that stuffs really big...hmm, bad news for us
> home filament makers...at least says how difficult
> the process is gonna be. Blah.
>
Keep in mind that Reprap has already brought the cost of 3D printers down from $30K to well under $1K and the cost of printed material from $30/in^3 to about $0.15/in^3.

We just need to think about this filament problem for a while. There is very likely a way to do it cheap.

True. Just because it's currently manufactured by very large, expensive machines doesn't necessarily mean that that's the *only* way to do it.

I suspect a lot of the need for a massive machine is to be able to do it quickly for an economically viable industrial process. We only use filament at a snails pace so we can afford to do it much slower, so we don't need a high pressure.

Edited 1 time(s). Last edit at 11/26/2008 06:01PM by nophead.

---

[www.hydraraptor.blogspot.com]

From having extruded some 3 mm filament back in 2006, I think the chief problem that we will be facing is assuring diameter consistency. It ain't that hard to extrude, but keeping diameter to 0.1% is going to be tricky.

0.1% is pretty stringent, I would be very happy with 1%.

An optical measuring system and feedback?

We can either make accurate filament or measure it on entry to the extruder and compensate when extruding.

---

[www.hydraraptor.blogspot.com]

I've seen some filament machines on the tv the other day, they made synthetic turf ... nevermind,
the point is, they said to ensure constant filament diameter they extend the warm and soft filament when it comes out of the extruder, that way the bigger and softer parts stretches more than the thiner parts..
warming the filament and repeating that multiple times they made some 0.1mm filaments.
Don't ask me about the accuracy, it was about synthetic turf in some swiss stadium (Stade de Suisse, i guess) so they didn't had too much details.

Maybe that'll work with bigger diameters like.. let's say 3mm (just what came up to my mind *gg*) too ?!

So what's easier to control for us? tension or pressure?

Any ideas?

'sid

Wouldn't our effort be better spent on a granule/scrap extruder? That's basically what a filament production system would *be*, after all. It also lowers requirements, since we don't need to fit the product through a rigid barrel.

... main problem with a granule/scrap-extruder could be bubbles in the filament, which would disturb the flow-homogenity and so the fabbing quality.

AFAIK the long paths and high pressures in common extruders are mainly needed for driving any embedded air out of the filament ...

Viktor

As far as I understand, the process for creating rod goes like this:

Run a hopper feed into a melting pot, with a hole at the bottom. The size of the hole is the first-order determining factor on the size of the rod.

Next, take the ooze of plastic that comes out the hole, and extend it horizontally under a cooling fan to set the plastic.

Then put it through a couple of rollers, which will draw the plastic at some consistent speed. Setting this speed is the hard part, and probably requires an optical feedback loop. This is the second order determining factor on the size of the rod.

After the rollers comes a very sharp-edged hole in a piece of steel plate, which will shave off any excess.

Finally, you have a large spool on another motor which collects the result.

From what I know of drawing ductile substances (not that much) the most important thing is the roller speed. There are a few methods to do this optically, but there may be a way to do it cheaply with a good quality webcam and a coded background image -- just see how many black dots are blocked by the drawn substance, and you have a size measurement.

hi folks,

the Gingery books have probably been mentioned on the forums here before, but this book seems so relevant to the thread i figured i would post it here anyway. disclaimer: i'm not affiliated with the publisher (i.e. i'm not selling anything), just thought you guys might be interested...

Plastic Injection Molding Machine by Vince Gingery

From the description:

You get complete step-by-step instructions revealing the secrets of building a small inexpensive table-top injection molding machine capable of molding up to a half ounce of plastic. Although a half ounce may not appear to be much plastic, the truth is, that it is more than enough to produce many small useful items.

Best of all you'll be molding with plastic recycled from milk jugs, soda pop bottles, plastic oil cans, and more. Your raw materials are free and plentiful.

[www.lindsaybks.com]

I built one of those. The problem with the either of the designs that Gingery publications offers is that they are shot injectors that only have a capacity of a few cubic centimeters. You need a continuous injector design in order to make significant lengths of filament.

hmmmm. just thinking out loud here, but does a reprap need long filaments? i have had decent results with re-filling the extruder on-the-fly, by manually feeding in a filament as the old one is drawn into the extruder (mkII extrduer). there doesn't seem to be much of a hiccup in the transition from old to new filament (at least its a minor hiccup compared to all the other things that go wrong during a build...) i wonder if we could make some sort of hopper that would hold say 100 short (30cm) lengths of filament and gravity feed them one-by-one into an extruder...

pumpernickel2 Wrote:
-------------------------------------------------------
> but does a
> reprap need long filaments?
>
Definitely. My machine can eat as much as 2 meters/hour of 3 mm filament (~27 cm^3/hr). Nops can use twice as much per hour as mine does. Unless you want to sit around feeding in short lengths of filament 24x7, you've got to have very long pieces of filament.

Those Gingery machines are seductive but not very useful for our purposes. As well, the one you link to is a right bastard to make from scratch. It was obvious that the Gingery brothers built it from scrap that came easily to hand in their shop, so it didn't cost them much to built. If, however, you try to build that damned thing by buying materials and all the tools to put it together you quickly find that you've spent upwards of $500 to get it working. It's mostly because they use a mix of welds AND bolts rather than one or the other. As well, you need access to a heavy duty drill press to drill out that cylinder that the injection piston fits in. Mercifully, I didn't have to buy one of those. It took me three tries to get one of those drilled right, even after I plunked down money for a cross-slide vice for the drill press. Drilling deep holes in hard metals is something for a heavy duty lathe, not a drill press.

Finally, the brothers specified cold-rolled steel, which was just plain daft. That damned stuff varies dramatically in hardness. When you are drilling bolt holes in the "L" profiles that's no big deal. When you are drilling the injection cylinder, however, your drill bit will encounter a hard spot inside the steel and tend to deflect to the side, especially if you are an inch or two into the job.

Edited 5 time(s). Last edit at 11/28/2008 11:58AM by Forrest Higgs.

pumpernickel2 Wrote:
-------------------------------------------------------
> i have had decent
> results with re-filling the extruder on-the-fly,
> by manually feeding in a filament as the old one
> is drawn into the extruder (mkII extrduer). there
> doesn't seem to be much of a hiccup in the
> transition from old to new filament

I've done this too. Worst I've gotten is a bit of an air bubble.

Another idea I've been kicking around in the back of my head is some sort of filament splicer. I'm considering modifying a crimping iron for the job. I'm told they go upwards of 200C, which should be plenty.

Forrest - Just a quick note. The Gingery's are Father and Son. Dave Gingery died in 2004 but his son has put out some new books and video's based on his fathers stuff. The thing I like about Dave's stuff is it is a starting point on DIY big time. I have built and used his foundry. But when was the last time you saw a 5 gallon metal bucket sitting around. I had to buy mine to get some. But if you follow his general thought trend and use the same basic methodologies you can be successful with what you build. Just like it appears that you do every day.

Bob Teeter

---

Bob Teeter
"What Box?"

It's beginning to look like recycling plastic into new filament isn't going to work so well.

[forums.reprap.org]

Given an inability to use scrap plastic except in small percentages in filament production, I am beginning to think that local production of filament isn't ever going to be worthwhile.

bobt Wrote:
-------------------------------------------------------
>
> But if you follow his general thought trend
> and use the same basic methodologies you can be
> successful with what you build. Just like it
> appears that you do every day.
>
Make no mistake. I adore the Gingery philosophy. I've bought a bunch of their books. What I was saying is that the old Gingery injector is a bastard to make and can't be done by somebody without a well-equipped machine shop for anything like their cost estimates.

THAT is the weakness of the Gingery approach, IMO. To do their projects it assumes that you have space for and time to build their line of machine shop tools. Once you've done that you are pretty much home free. Keep in mind, though, that it going to take you quite a while to make their stuff and you still need shop-quality floor space, and a substantial amount at that, to build their equipment.

My worry is that most people don't have that kind of floor space any more, especially the 12-18 year old kids who would benefit the most from this kind of tuition. Most of the poor bastards are living in apartments with their single mothers and, if they're lucky, have their own bedroom ... and precious little else.

Given that take on things, I've been quite taken with Reprap. A Darwin or a Tommelise is something that such a kid could hope to build in their bedroom and operate without the neighbors complaining. When we start talking about slipping in things like Gingery foundries and tools, we are disappearing right over those kids' practical event horizon. IMO, we've got to focus on stuff that a kid in that age range could build without an adult having to get involved. Most kids like that have a marginally functional mom who most likely doesn't know which end of a screwdriver to hold. Male role models tend to be hapless and transient. Either the kid figures out how to make it happen himself or it isn't going to happen.

Edited 2 time(s). Last edit at 11/28/2008 03:47PM by Forrest Higgs.

Very sad picture you paint there Forrest.

unfortunately enough that you're right with that,
that does mean we have to enhance almost everything for safety reasons.
I'm quite unhappy to know that untrained 12yo snip of cables of PC powersupplys and shorting them.

I know that there are enough 10yo that can do that without advice, but I do also know that there are 20yo that can't!

So... powersupply adaptor boards anyone?
(and that's only one small step of various steps to make it kidsafe)

'sid

Forrest - I agree with your statements. So how do we solve the problem? Can most of the parts for the Darwin be built out of wood? If it can then all it takes is a portable drill/electric drill and some sanding paper wrapped around a block of wood to finish the parts. This same methodology should work for most 3rd world countries. Its real nice to build a darwin but I started with a repstrap because it was not realistic to start with the finished product and I do have a complete workshop available.

So back to solving the problem. What materials do most people have available to start. I think in most cases wood is available. Then what and how much needs to be built to be able to fabricate parts to convert the unit into a finished Darwin.

I would think with all the brain power on this forum we should be able to solve this problem.

---

Bob Teeter
"What Box?"

Bob, et al,

IMHO, there are two slightly different issues here:

1. How to lower to cost and/or increase the availability of feedstock to extrude. (And recycle plastic items into useful things, if at all possible.)

2. How to make it easier for people (esp. without lots of tools or money) to build repstrap machines?

Regarding issue 2:
Except for the rods, Bruce and Nick Wattendorf built much of their (first) machine out of wood. Some photos are at [picasaweb.google.com]#

Wood is pretty hobbiest-friendly, and is usually cheap/available. I think wood is particularly useful for the rectalinear parts, such as the corner blocks, especially with some threaded inserts (helicoil, allthread, or T-nuts) for the screw threads (and to help prevent splitting.)

Regarding issue #1:
Extruding from pellets or ground scrap is still an unsolved problem (at least on our typical budget), but I'm not convinced that it's insoluable. Maybe the problem could be solved better in two steps:

A. Convert the little particles into a prismaatic, bubble free form. (e.g. via vacuum casting? Probably a batch process, I think.)

B. Stretch/roll the void-free prism into progressively longer, thinner diameter rod, using feedback to arrive at a consistent diameter for reprapping.

Related side question:
Has anybody tried thicker welding rod? If so, how did that work/what were the limitations of using thicker rod?

Food for thought,

-- Larry

>Has anybody tried thicker welding rod? If so, how did that work/what were the limitations of using thicker rod?

Not intentionally but I got some ABS that is about 3.5mm oval and I have noticed it is a lot harder to bend in the stronger direction. I think it is a 4th power on diameter so any bigger than 3mm will get a lot harder to spool and feed.

---

[www.hydraraptor.blogspot.com]

Depending on the plastic, 3 mm can be a problem, too.

For beams the bending moment capacity (and hence stiffness) is proportionate to the depth cubed multiplied by the breadth - so for a cylinder dimaeter^4 makes sense.

I've no idea if structual mechanics can be scaled down this small reliably, but the principle seems good.

madd0ct0r Wrote:
-------------------------------------------------------
>
> I've no idea if structual mechanics can be scaled
> down this small reliably...
>
Yeah, it can.

Forrest
Why is diametric accuracy critical? I don't have one of these units working yet but would like to know.

Because we need to feed the filament at the correct rate. If we know the diameter it is simple maths to work out the motor RPM required but if the filament diameter varies we would need to measure and adjust the motor RPM to suit.

---

[www.hydraraptor.blogspot.com]

Thanks, Nop!