Home Polymer Recycling Unit To Produce Filament

The trick is to make it open source, but hard to make or need special tools. For example to make a Makerbot you need a laser cutter and a lathe, etc. Makerbeam is another good example. They publish drawings, but how many people are going to get their own extrusion dies made?

---

[www.hydraraptor.blogspot.com]

Ben,
Even the simple HDPE recyclers are going to be high in vitamin content, low on RP parts. To make something strong, long lasting and reliable will probably benefit from some custom designed and built metal parts. That means that for several generations of repraps to come, having even a soft plastic recycler will mean purchasing a kit of special parts, if not a full kit or prebuilt unit. So there should be a long term market in these three areas. Also, once you are successful in small scale at home recyclers, you should look into scaling up to business or community level machines. Restaurants and other businesses to produce lots of HDPE waste might want to recycle in house to advertise how green they are. Since pool chemicals, hydrogen peroxide, and other chemicals are also stored in thicker, stiffer HDPE tubs, buckets, and bottles, there are a lot of other businesses that will produce lots of HDPE waste that may need a bigger, stronger recycler with a bigger motor. All of them would be able to turn expensive waste into a sideline profit if they sold the reformed HDPE filaments to reprapers or distributors for reprap. Depending on how popular HDPE becomes for FDM extrusion, this would help build a small niche industry. Manufacturers of HDPE recyclers would also make great wholesalers and redistributors of the recycled HDPE.

Also, progress with the softer, easier to work HDPE should lead to stronger, tougher recyclers that handle ABS. I think that many reprapers would buy their own ABS recycler if the price was reasonable to recycle failed prints and the ubiquitous ABS scrap from construction. That gives us a cheaper, more plentiful supply of filament, and also the satisfaction of recycling waste, and being part of the solution, not being part of the problem.

So even though the design maybe open source/open hardware, you will have a head start in building and marketing parts, kits, and units. And since it will be a niche market for some time, not a whole lot of competition. Simply forecast the rate at which people will want HDPE and ABS units over the next 5 years to make the basis of your business model.


Testing:
I have both a confetti making paper shredder and a garbage/food disposal/grinder. As I collect enough empty HDPE bottles, I will try video taping shredding and grinding the plastic bottles to help get the research off the ground. After that, I have to go back to finishing my McWire and getting it running and calibrated.

Mike

Have you thought about utilising large office shredders? Faulty units that have either a broken gear or duff motor can be had on the evilbay place for little money. I have used a retrofitted shredder for all sorts including plastic milk bottles.

Thanks for the input again guys.

Will look into them all and get back to you with anything interesting!

Would it be worthwhile documenting our progress in a Blog or something?

I'd suggest this forum, or better, the reprap-dev mailing list, since you're developers.

Also, we do have a lovely wiki.

---

-Sebastien, RepRap.org library gnome.

---

Remember, you're all RepRap developers (once you've joined the super-secret developer mailing list), and the wiki, RepRap.org, [reprap.org] is for everyone and everything!

Greetings all,

I tried running just the thin part of a two liter soda bottle through a supposedly high-power shredder, one that claimed to be able to shred CDs. That shredder didn't do very well on the thin, flexible bottle material, and I had to manually unjam it multiple times. I was finally able to shred up 100 grams or so of PET. I then attempted to dry it out, and then melt it. Apparently, I didn't dry it for long enough, or I heated it too hot during the melt phase, because the polymer became opaque (an off white color.)

---

Larry Pfeffer,

My blog about building repstrap Cerberus:
[repstrap-cerberus.blogspot.com]

I just ran a plastic milk bottle, HDPE, through a home/small office paper shredder. I cut the milk bottle into a bottom piece and 4 sides. Cutting through the handle took a bit of work ! When I passed a side piece through the shredder, it came out in one large piece still, but with deep lines between the cutter blades, and little cliffs where each tooth tried to shred it to confetti. So I ran it through a second time perpendicular to the first. It was still mostly intact, but now a tiny checker board from the deep rows between each cutter wheel. I wanted to video the test, but I can't find my digital camera. Pictures and/or video to follow as soon as I can find it.



Next test is the food disposal, and the meat grinder.

Mike

Edited 1 time(s). Last edit at 11/22/2010 10:44AM by rocket_scientist.

Thanks for the testing guys! What are the actual shredding blades like on the machines? something like a serated metal disc? I'm hoping it's not all bad that they have failed to efficiently shred the HDPE.

I've got hold of a 600W AC motor on a grass strimmer from a car boot sale. next stop the workshop and see if I can get some sort of blade on it! Going to try and put a whole in the bottom of a plastic bin and try and make a blender out of it... can't hurt... well it can but not if I'm careful!

Most shredders will happily go through CD's as the material shatters when put under enough force. PET and HDPE will simply bend and elongate under those sorts of forces, rather than shatter.

I would consider running vertical blades, possibly counter-rotating, with a horizontal blade (like a blender blade) underneath. The horizontal blade is there to kick up plastic that gets through back into the upper blades. There will still need to be some sort of system to divert larger particles back to the start of the process, so that the particles get down to size.

Note: I would spend a bit of time making sure the system is enclosed and well sealed. The goal is to recycle plastic, not to create large amounts of tiny pieces that then somehow get back into the environment.

Disappeared for a bit there sorry guys. End of term always gets real hectic with hand ins and all.

So we've made some progress. Using an industrial grade granulator we've gone through 40 odd milk bottles and hence got ourselves a nice supply of HDPE flakes. Have also tried a small blender, that got through a bit of bottle quite well but it was so small that it overheated and made a horrid smell.

So moving forward with the project we are developing a blender off of a 600W AC motor and some sheet steel cut,bent and sharpened into a blade.

Does anyone know of a low cost source for such motors?

we've salvaged ours and would need to quote an actual supplier for our report.

So we're in the design development stage of hopfuly producing a machine that:

-Shreds bottles or malformed RepRap parts
-Heats and extrudes the chips in a screw type extruder
-Hauls off the melt through a water bath or cooled nozzle
-Spools the material into 1kg spools

so stay tuned for progress report and prepare yourselves for a barrage of design related questions!

Many thanks.

Edited 1 time(s). Last edit at 12/15/2010 05:01PM by BenBrazier.

Try surpluscenter.com - I've gotten a number of motors from them, and they have a 1HP for $60:
[www.surpluscenter.com]

Of course, shipping would probably run half that for such a beast...

---

[haveblue.org]

wow! its becoming more and more unfair/annoying that no one is bringing this stuff at this price into the UK!

Oh, sorry, didn't check back far enough in the thread to see that you're in the UK. Hopefully you can still use it as a price point for the purposes of your report.

---

[haveblue.org]

Ben,
When making your own custom, giant blender, make sure you use the correct bearings to support the drive shaft. The standard bearings used in repraps are ball bearings. They can handle large radial loads, but have very weak side load ability. Tapered roller bearings can handle both radial and axial loads, and should work much better, and be much safer for holding your whirling blades of milk bottle death. Make a box of wood or metal, drill through for the drive shaft, counter sink each exit hole to fit the outer bearing race, then slide the drive shaft in and use nuts to clamp the bearings between the two nuts. You can probably use a single piece of studding for the whole drive shaft, and that way you can bolt on the blades in the blender section for easier experimenting and replacement.

Mike

Ben,

Regarding your polymer recycling effort:
Great stuff, but *please* be careful, particularly when dealing with such a high-power cutting/grinding machine. Any such that can cut up milk bottles could -- if things go wrong -- cut up the operator or bystanders. Please consider the following design aspects carefully:
Both static and dynamic balance of the cutting blades.
Blade/shaft attachment: extra secure -- vibration resistant
A secondary containment strong enough to stop a wayward blade (I suggest polycarbonate.)

---

Larry Pfeffer,

My blog about building repstrap Cerberus:
[repstrap-cerberus.blogspot.com]

I like this idea a lot in general , so here are my two-bits (or pence or w/e).

I would first of all aim for a box-in-the-garage type design (in lieu of beside-me-and-my-computer) for two reasons. First off, you'll be grinding plastic down to itty-bits and it will without a doubt increase the particulate plastic matter in the air you're breathing. Secondly, A spinning blade a do-it-yourselfer has mounted to a fairly fast little motor is not something that should be in the same room as people... unless they're your in-laws... but even still... it would make a bit of a mess.

For the motor , the 60$ 1.5 hp would be better https://www.surpluscenter.com/item.asp?item=10-2552&catname=electric.

Other than that, I made a 10 minute pitstop at the offices autocad and pieced together a containment unit that is made of wood and still safe. 4`x4`x4' in dimensions (my local lumber mart would stock 2x4's in 8' so this minimizes waste).



Top down just throw stuff in like a blender (would need a lid too)... fill in side cavities with rock/soil (pretty cheap last time I checked) which should make short work of the blade in case it decides to change vocations. Make sure you get the whole washer/bearing/shaft assembly figured out well and good luck with a blade (maybe ask blendtec if they have any 4"-8" ones? ). Have a mesh for the chunk size of plastic you're willing to accept and then just flex duct it down to a container. Silicon seal all the plywood/osb/mdf walls on the inside if you're going for plastic dust instead of chunks. Aside from the blade assembly I can`t imagine anything else that would give you trouble... except for buying the parts, building it, messing up, fixing it, running it, messing up, and then giving up because it plain doesn't work.

Aside from that it should be easy!

haha yh, hope that doesn't happen.

Real nice for you all to show your concern about me.

The university project must be up to scratch otherwise they simply won't let me build it.

Have checked up on several european standards regarding:

Degrees of machine protection
Safety of rubber / plastic size reduction machines
Safety for polymer extrusions
Safety for polymer extrusion haul off

If we keep to that then we should be great!

Regards to bearings for the shaft. I hadn't thought about actual selection yet. I'm sure one of my tutors would have hit me hard enough for tapered to sink into my head, after all the Critical Design Review has to go through the ex chief design engineer at Rolls Royce.

We are currently moving through the embodiment stage, and your ideas are helping us a lot. Will post Pro|Engineer designs as they emerge after the January exam period.

I was currently thinking of a polymer sheet blender body, laser cut and screwed together. I presumed that choosing high impact grade polymer would help, polycarbonate does seem a winner there.

Getting hold of the AC motor in the UK is a continued headache! How do they afford to put them in things like lawnmowers and grass strimmers if they cost a fortune?

haha, all these problems and we havn't even gotten onto the extruder yet! ...

Ben,
Would it help if one of us lucky ones in the States bought one for you and then shipped it out? Things can be sent as 'samples' through customs without major charges or problems from what I have heard. To help advance the reprap world state of the art, I would be willing to act as go between myself.

I have been looking at push lawn mower ideas. They have the best potential for taking whole bottles and chopping them up, I think. But they will not make for very small or very uniform. To improve that I would put the rotating blade spool inside a much larger rotating drum, with deep paddles that would catch the oversized pieces, raise them up and run them through again and again until they are small enough to fall through the open mesh.


Basic side view of push mower (rotary cutter with 'blade spool')


The milk bottles are much larger than the blade separation in all the used push mowers that I have found so far. This means that something will have to be done to either make whole bottles into small pieces (a wasted operation), or force the big bottles to get chewed up by the smaller blade spacing anyway.


Adding an outer drum increases the complexity of the build, but no provides a way to cram the large whole bottles into the blades to get the initial cuts made so the remaining pieces will be able to fall in and be chopped and chopped again until small enough for stage two.


Here I have added a funnel to catch the small pieces that fall through the mesh of the outer drum. I would suggest somewhere between 1 and 5 cm sized openings. Then the small pieces fall into a coarse grinder. The grinder has a cutter blade on the exit orifices to chop the plastic into moderately small, uniform pieces. These then fall into a hopper to buffer out the fast and slow rates of plastic granule production. The process could stop here for a granule fed FDM extruder, or feed into a second, smaller, more carefully rate and feed controlled grinder that would spit out finely chopped plastic into a tapered melt section. Using a long, slow narrowing melt chamber should make the polymers in the final filament more parallel and strong rather then squirting though an undersized orifice and having the plastic partially spring back. Then I have suggested a thermal brake and a liquid cooled section, all at the same diameter, to quickly cool the filament down to make it strong enough to pull and wrap onto a take up spool (not pictured).


It would be nice to use to rotating spool cutters that rotate in opposite directions and would help grab stuff and push it into the cutting blades. However, when the blades rotate towards each other, they provide no shearing action when they meet, no matter which slant the blades have, and can only cut by squashing the plastic. Not very efficient. The only alternative would be to have a double sided stationary cutting blade in the center that each set of spool blades cut against. This might be more complicated for an engineering test than necessary, but might be a goal for production design. And even then the results will still have to be recirculated somehow until the pieces are small enough for second stage grinding.


Mike

Many years ago, "the first oil crisis" the company I worked for could not get polymers, so installed a recycleing plant,
lorries arrived with various plastics, this went into a cracker, which was much like the above sketches by rocket scientist,
it had blades over a metre long, "which I had to sharpen" the machine had a ladder up to a platform and a shute where the
bandsawed parts were fed in, " the linings of fridges etc", the blades rotated at high speed and a bin at the bottom collected
the cornflakes produced, the flakes then went into a "banbury" with colouring, which kneeded the molten polymer, it then dropped onto a conveyor and then into the extruders.
the cracker had a very large motor, it did not stutter when over fed, but a pair of rollers like a tree chipper uses, could feed
material at a defined rate perhaps, I would seriously consider making the case from sheet metal to be on the safe side.

Industrial machines tend to use brute force, but for home use it may be better to nibble away rather than take big bites.

---

Random Precision

@Mike

May well have to look into getting a motor shipped over from the States, thank you very much for your kind offer. We shall see how we progress with procurement over here, but I may well have to take you up on it. Will let you know when I do!

I love the idea of your tumble-shredder and can picture it with several of the smaller internal pick up assemblies staggered around their PCD. It would be driven by something fast approaching a planetary gear box and would perhaps resemble the machinery used to excavate tunnels?

It has all the principles embedded within it, and the part count to go with it. By the time we're done buying all the parts that we would need for the extruding, cooling and spooling it feels like we may be approaching the £500 budget already.

But as a production design, it is the business.

Just a concern that the small domestic target that this would be primarily aimed at isn't ready for it.

The primary objective is to produce a filament that can be used in a RepRap or RapMan style machine, for this reason we have been asked by our tutors to focus on the extrusion and spooling. The official line was "Better to get 50% done well than 100% of the parts started with 50% of them all finished."

I can assure you that all designs will be recorded and archived for future iteration, (I have an individual project to do next year too).

@Random Precision

Yes exactly, with the amount of budget we have left after making a very nice extruder (which I'm sure will require just as much input from you guys) we would be looking at a much smaller, one-bottle-at-a-time-shred-as-you-go principle that would take its time to eat through enough bottles that were slowly accumulated, storing the granules ready for a bulk process at some point.

@Everyone

Thanks very much for you persisted interest and input in this project. Designs are being drawn up at this stage and we aim to get them up on the web for all to evaluate!

Ben,
I think the internal shredder, cutter, shear, whatever inside must remain stationary with the outside world, while inside a rotating drum (similar to the squire cage fans of evaporative coolers for those who have seen them). This means that a structural member that holds the lawn mower cutter must go through a hollow axle and bearings of the outer drum. Worse, to power the spool of blades, which will need the highest torque of any part of the machine, it may be necessary to make the support member hollow too, and send a drive shaft (or motor wires) down the middle of the structural member. This means that the lawn mower cutter is built onto the inside of the two hollow shaft pieces, and the whole frame work connects to the outside. To be able to assemble and dis-assemble the system for repairs or improvements, the outer connection at least must be temporary (bolted or clamped on, not welded) so that the spinning drum bearings can come on and off. This is viable for commercial construction, but a bit artsy for the garage/shed tinkerers that we are aiming at.

I therefore contradict myself and make some new recommendations. The first is that the greatest problem with turning the first design from paper to reality is the constraints of working inside a rotating drum. But we need some way to catch the big pieces and feed them back in again, yet allow the pieces that are now small enough fall through. So instead of closing off everything, what about a conveyor belt underneath the cutters that only goes up one side to drop the big pieces back in? If the belt is made of side-to-side rods and links, it makes the mesh automatically to let the small stuff fall through. With guide rollers on each side, pushing the top and bottom sides of the belt downwards, it should be possible to make the blet follow a half circle path, leaving the other side open for whatever cutter system you want to use. This means that the main structure can now be two vertical sheets of metal or plywood, one on each side, and the conveyor belt and paper shredder, lawn mower cutter, etc can be placed in the center of the half circle and the whole thing assembled just like the cross sectional view.


Here is another drawing replacing the single acting lawn mower blade spool with either dual blade spools (and middle cutter blade) or oversized paper shredder drums. Along the bottom and right hand side is the conveyor belt that lets the little pieces fall through and returns the big ones to be cut again. Since a tight belt tries to make a convex shape, to get the concave shape needed here extra rollers or spur gears are added the push the top of the belt downward to the desired shape. You can see the paddles that catch bigger pieces and carry them back. AT the very top, the belt bends one more time to go upside down and drop all the recycled pieces. This may need fewer, deeper paddles depending on the number and size of big pieces.


This takes a closer look at one easy way to make the conveyor belt. The horizontal pieces are thin rods, initially metal, but possibly thickened up and made of RP plastic. The 'chain' on each side is made by using short links and rods, just like a bicycle chain. This is where the drive spur gear will connect, and where all the guide rollers will contact the belt. The vertical pieces are 'V' shaped links that make up the bulk of the belt, and also become the paddles to catch the big pieces. The links are spaced apart with short pieces of tubing, which can either be cut from metal tubing or made by repraps from plastic. The gap between the horizontal rods and the vertical links sets the mesh size for what falls through and what goes through again. It will require making new belts to make changes, but eventually the whole conveyor belt can be made of RP parts, greatly improving the ability to put one in every repraper's garage. There is very little stress on the conveyor belt, so using plastic should not compromise its functionality.


Here is a side view show the 'bicycle' chain along the edge, and the 'v' shaped links bending backward to catch the big pieces.


Construction is much simpler with this design. The conveyor belt must be at least as wide as the paper shredder or lawn mower blade spool is. The right side plate is drilled with mounting holes for the shredder/cutter and the baffle to divert pieces that fall of the top of the belt over to the best cutting spot. One of the two end rollers of the conveyor belt is replaced with a drive spur gear that mounts through the side plate and connects to the drive pulley/belt off the motor, which can be mounted on the same side piece, or between pieces to the left of the cutter. One or both of the cutter spools, paper shredders are also connected to drive pulleys and belts. Then the left side plate is drilled to match. The right plate is temporarily laid on its side and all the pieces fitted in, the conveyor belt threaded between rollers. Then the rollers are mounted on the left side, the left side placed on top and everything passing through bolted down. A few bracing cross members and the whole this is ready to run.

I think this is a better design for small scale, but a well built version of the first design may be best for commercial production if that ever becomes important.

Mike

Edited 1 time(s). Last edit at 12/22/2010 06:12PM by rocket_scientist.

I just tried using a hole-punch to nibble bits off a milk bottle and it worked quite well. If you constrain the requirements to processing milk bottles I'm sure something far less elaborate would work fine.

Not recycling, and probably suggested before, but Its just occurred to me that the binding rope used by farmers, " not sure what polymer" could be put through a die to extrude it into a filament., other "Ropes" could be considered perhaps.

---

Random Precision

Hello everyone.

We are still here and are still working hard, headaches and all...

Currently in the embodiment stage of the project and an interesting idea has come up.

There is no need to make such a big shredder... Working on a 'Shred-as-you-go' principle where you destroy bottles as they become available means there wouldn't be scores of bottles piled into the machine at once. It is a not-so-harsh requirement to ask that bottles are sliced up by hand with a craft knife or scissors.

On this basis we are looking at a 100mmx100mmx150mm shredder volume operating on a Blender principle with a 3 or 5mm diameter screen. Granules are collected into a post shred hopper above the extruder. When enough granules are present to extrude X kg of HDPE then the extrusion process can begin.

We've chosen Arduino to control the thing at the moment and i'm loving getting to grips with it. I have a few questions about it though which are causing a slight headache. I will post on Arduino boards for specific help with those when I have enough information available.

Is there a full electronic schematic for the RepRap electronics that use Arduino? Would be nice to see how a similar machine goes together.

Schematic of the v1.2 Motherboard.

---

Bob Morrison
Wörth am Rhein, Germany
"Luke, use the source!"
BLOG - PHOTOS - Thingiverse

haha, my facial expression just then was most definitely the image of the phrase "be careful what you wish for!"

Thanks Bob for such a prompt reply!

I guess I may have to take a walk up the tower block to the electrical engineers... haha!

ok. someone with an electronic brain...

is there a transistor or other component, that is activated by the small 5v TTL signal of the Arduino, and when activated connects another power supply, like 12V DC at 0.08Amps. (the sort that would power a small fan).

This is just one of the things bugging me right now...

thanks

Not on the Gen3 motherboard, but the extruder controller uses transistors to control the heaters, heated bed and so on. The schematic for that is here:
[objects.reprap.org]

sorry Andrew, I'm not too good with these schematics. Where are these transistors or what are they called? Would I have to develop a small board that had a few of these transistors on for the specific task of switching on and off several different 12V appliances with TTL from the Arduino?

Thanks very much for your help, and patience

Look at the PWM Driver section. There are three mosfets there intended to drive up to three loads, though most people only use one to drive the extruder heater.