🇪🇪 Precious Plastic Estonia
Today we have built the shredder and did a test run. A couple of things still to be tweaked, but overall we managed to get our first batch of shredded plastic.
Thanks to @davehakkens and the team for making this possible.
We would be happy to help anyone who is planning to start building the machines soon with some before-assembly advice.
December 2017 update: Shredder, Injector and 5/7 of the Extruder ready.
January 2018: Moving part of the production to Tallinn.
March 2018: The last – Compression machine almost ready.
May 2018: Started making wallclocks.
Sep – Dec 2018: Not so active season, looking for sponsorhip deals.
Estonian team: @jegor-m; @maximmm; @igor-smog; @dannydadog
Instagram /// Facebook page /// BAZAR page
Here are some photos
All the machine building and testing along with plastic sorting and preparation is done at my summer house in the fresh air. This weekend I sorted through all the plastic collected by friends and the team members. There result is on the photo below.
Plan for the next week is to finish with shredder (some aesthetic tweaks) and to start building the injection machine.
While the compression unit parts are being collected, we already have a small oven, which we want to use for HDPE heat up and further pressing in a separate wooden frame to get basic HDPE plates.
Lots of planning ahead, but it is for good! See you next week.
This weekend we managed to finish our shredder and test it even more. We added a hopper and tried shredding PS and PET. PS(6) got shredded nicely. Shredding PET was way more tough.
– First problem is that blades (56rpm) tear the plastic and pull it in between the ribs and the plastic then stops and needs to be replaced.
– Second problem is as soon as it jams, the physical connection between the motor and the shredder axle breaks. The link is in a form of bolts, placed in a hole that was drilled perpendicular to the shredder axle. The bolts are shear cut every time the motor jams. This weakest link is to be solved at our next meeting.
Also we built about half the frame for the injector. As we have all the electronics, the final injector assembly and testing is going to happen soon.
While working on the things listed above, we also tested the mini oven we got. We placed an Aluminium baking form with some shredded plastic in it into the oven and waited for it to melt a bit. Then we placed another one of these baking form over the plastic and left it in the oven for some time. The result below shows the texture we got. Initially it was filled with black, red and transparent PP.
– The mold needs to be at least warm when it comes in contact with hot plastic
– Wait till you think it is ready, and wait more, only then take it out of th oven.
– Hollow aluminium is too soft (not quite the mould material, unless it is in a block form)
– Better to overfill the mould with plastic
– Don’t forget about the stickiness of the mould. Something needs to be applied to the surface (baking paper, non-stick spray, mirror finish, etc ..)
The last weekend was a bit chilled. We shredded 3.2kg of PP (5) coming from all sorts of canister lids, lawnmower casings, plastic sledges & used office supplies.
We also bought some cheap sheet metal bowls at a shop to do a bit of manual compression moulding. For that we heated up some PP in a small oven. Plastic was placed in between the bowls, which were cleaned beforehand. These bowls were strengthened a bit to prevent them from warping when load was applied. When plastic melted, we took it out and compressed the bowls between two plates until some plastic was squeezed out of the mould.
– Both parts of the mould were heated up in the mould, so no rapid surface plastic solidification occurred.
– Inspection upon ejection of the product from the mould showed that plastic was not fully molten, hence heating up time should be increased
– Thin steel bowls used for moulding tend to warp, despite the strengthening. Thicker moulds to be used.
– Overfilling the mould with plastic increases the melting time, this to be carefully considered.
– WD-40 lubricant was used. Some areas got stuck to the mould, some were easy to eject >> ultimate solution still to be found.
– Transparent Polypropylene had higher melting temperature and did not melt properly. Black and red PP mixed well. The result – layered bowl of very low strength.
We contacted local clothes shop and managed to source plastic hangers that are thrown away as soon as they are used once. We separated all the non-plastic bits and shredded a batch. All the hangers we got are made of PS (6), so it is quite easy to break them by hand and they shred nicely.
We also continued practicing plastic baking. This time we heated up some PP (5) in the oven and then placed it between two plates into the press (not heated press). Then this assembly was pressed till the plastic squeezed out. The photo of the result is shown below (banana for scale; banana is not made of plastic).
You might think it looks awesome, but unfortunatelly the process was not optimised well and pressing didn’t go as planned. As a result the plate is not quite flat and thickness is not uniform (approx. from 3-8mm)
– Baking paper was used – no problems with plastic sticking
– Middle solidified faster so pressing did not go symmetrically. Metal form / plate should be heated up with plastic.
– Integrated compression machine is way better solution than separate oven + press.
Nice updates Jegor! I like the lesson learned list you make. Very informative.
It’s time for weekly update N6:
This week we tried making some thin plates of PP(5). The shredded plastic mixture used was the same as for the things above. We are still using the oven separately from the press.
If you want a full description of the process, message me. The short description goes like this:
A square piece of metal covered in baking paper was placed into a pre-heated oven (about 230-250 degC). Some shredded plastic pellets were placed on top of it and all of this was covered by another square piece of metal (again covered in layers of baking paper). This arrangement was left in the oven for about 20 mins. It was then taken out and placed between bigger metal plates and squeezed with a press. As the plastic was fully molten, it was squeezed tightly and left to cool down.
As the result, the plates you see below are about 1mm thin and have some cool patterns. The patterns depend on the variety of colours you throw in.
– Leaving plates in the oven, which is not leveled (tilted), can result in plate slipping from the melting plastic. As we did not have any guiding walls / plates, the top plate moved to the side a bit, but was corrected in time.
– Cooling inside the press is better and gives a flat plate. Pressing hot plastic and taking it out while it is moderately hot / not cold warps the plate.
– Using multiple layers of baking paper allows easier and more safe ejection of the ‘mould’ out of the oven
– Using multiple layers of baking paper increases the risk of folds appearing, that could affect the appearance of the resulting plate. A good paper securing arrangement could be devised.
Getting real good!
Weekly update N7:
This week we split our Sunday workday into three parts:
– Work with materials: Materials got sorted, shredded and stored.
– Injector building: We decided to build the hopper from a single part of sheet metal. The plate was bent and hammered into the right shape and welded onto the pipe. The overall structure of the injector is ready, but there was a problem with the pipe straightness. After welding the hopper onto the pipe, the welding seams bent the pipe slightly, so that the piston jams inside. Metal twisting and warping is a regular thing during welding that needs to be accounted for. A couple of solutions are brainstormed already and this problem is going to be corrected at the next session.
– Third activity done – Flat sheet pressing. This time we made two thin PP(5) plates of about 1mm thick. One of the plates coloured (picture below), the second one made from transparent PP. While it took more time to melt it, both of plates had a good quality and nice surface finish. In both cases we used flat metal plates and layers of baking paper.
– Water jet pressure gun can be used for peeling labels off the plastic. This is a fast and efficient enough method. The drawback is that it has to be done outdoors and during cleaning it is very easy to get soaking wet, so waterproof clothing is to be worn.
– Welding seams ‘pull’ the metal after welding, which results in bending and twisting. This is easily seen on thin structures. In case of a pipe welding, it should be monitored and checked all the time. The advice is to avoid welding too much, expecially where it is not needed.
– Electronics can be prepared, assembled and tested alongside the machine building. All the necessary instructions and guides need to be accessible during the building stage. It is very annoying to finish the construction only to find out that some of the components need calibration and the instructions for those are in Chinese.
– For old ovens used separately for heating up plastic – good to have a temperature reading for more accurate process optimization.
– If plates are made of transparent PP plastic with text printed over it, which cannot be cleaned off, then the result will have this printed ink inside the part warped, which will look dirty and generally unpleasant.
– Any dirt on the transparent plastic is easily seen in the resulting plate. Plastic is to be cleaned well.
Totally love your work!
Keep it up 🙂
Lessons learned are gold. Keep it up man!!!
Weekly Update N7.1
This mid week update is about a problem we had during shredder assembly.
To cut down on the price for the lasercut parts we decided to order only the essential parts, which excluded most of the walls and the net holder plates. As I was told, a friend of a friend of a friend had a laser cutter and could cut the parts for us cheaply. So they did.
After the blades were polished we found out that instead of 6mm we had blades of approx. 6.2mm thick. When all the blades were stacked, this extra thickness took the space of the last blade and overall the structure got a bit loose.
Test footage of PS shredding below can clearly show how the back plates move.
(awesome to look at PS flying around at 0.25 of playing speed)
At the time we dismissed this as a minor issue.
Now, when we have a list of issues like that, we are planning to rebuild the shredder into our own v2.1
– Check what you get from laser cutting company and adjust the design respectively.
Hello everyone 🌎
Time for Weekly Update N8
During our last session we almost managed to finish up with the injector. We wired it up and tested and it did not work. 😐 The problem was with the the PID controllers. The controllers are not set up properly, while all other components work well (we tested heating elements separately, they worked).
First HDPE melting test performed – resulting material is quite strong and uniform and melts nicely. The problem we stumbled upon was with the baking paper. As we used our first pack already and forgot to buy a new one, we had to use the one from the kitchen. Unfortunately, the paper from the kitchen had no non-stick properties 😕.
– Reading prevous lessons and learning from them. (Obvious)
– Old baking paper might not be suitable due to lack of non-stick surface properties.
Our injector below doesn’t look nice and shiny, but we are planning to paint it at some point for publicity purposes. ♻️
♻️ Weekly Update N.9
This week we tried to start the injector. We got the instructions for the PID controller and later found out that the instuctions are not suitable for our model. The injector was left un-calibrated. Photo below shows Max trying to calibrate this crappy PID controller. To be continued..
Along with that, we sorted some bits in our workshop, now we have got shelves for materials. One shelf is on the photo.
Left to right: 2kg of transparent PP(5); some pink/red PP(5) and 3.2kg of PS(6)..
– For more control over the colour patterns, plastic should be sorted by colours before shredding and should not be mixed during shredding. Shredding all the plastic of the same type gives you almost no control over the beauty of your products.
– Usual industry injection moulding process includes material drying before injection. This lowers the moisture content and improves the melting/connection between different colours.
🌎Almost weekly update N.10
Why were there no news in the previous weeks?
-Well, project participants attempted to get some rest before the summer ended, some of us had holidays / pre-planned events and so on.
👨🔧 We got a little bit of planning done. Prepared a huge list of things to be covered, from machine and mould building, to registering and plastic collection organisation. A lot of things require our attention and as you can imagine/ may be experienced – it is going slowly.
Meanwhile, we got a full-sized working oven, made more plates and tried some wooden moulds. Also we started painting the machines, this to be continued..
– Wooden moulds tend to dry up in the oven and distort with heat, so it is better to use a thicker layer.
– Different colours of the same type of plastic tend to melt at a slightly different temperature. Material melting and mixing compatibility tests should be performed before any batch production.
– Plastic still sticks to a polished plate of stainless steel. More fine sandpaper is required. (I used 150 grit size)
– ! Sadly our bag of coloured PP(5) got contaminated with some amount of tiny metal chips, that are visible in the final plate. These chips don’t stick out, but still a plate of 220 x 220 x 2mm had three pieces. The material chosen randomly from the bag. Good material sorting and storage is essential!
🌎Weekly update N.11 – PART 1 – ‘The cool stuff’
This weekend we did some shredder upgrading. The photos and some ideas behind the upgrade will be posted next time (when the paint is dry).
Today I’ll tell you about our recent plastic melting experiments.
1) The plate that we showed in our previous post (weekly update n.10) was cut into roughly equal squares. Then placed in a chess arrangement onto a squared metal plate, covered in baking paper. We used this metal plate for all of sheet press moulding tests. As the plastic got molten in the oven, I placed another paper covered metal plate onto it and waited for some time. Later these plates were placed into our sheet press, which at this moment is just a metal frame with a car jack. This plate moulding arrangement was then left there to cool down.
We say it was left there, it actually means we switched to another task and forgot about it 🙂.
As the result, we got this cool patterned sheet of PP(5). Photo1 is the pattern, Photo2 is the same pattern with a light source behind. I had this idea for a while and wanted to make a chess set.
How did it happen anyway?
In the previous post we made a plate. One side is more red and the other mostly black. It happened that red PP plastic we used had a lower melting temperature so it melted before any other plastic. As all the plastic was lying on a plate, red molten plastic filled the lower layer. The black plastic took longer to melt so it mostly stayed on top. It takes time to get to know the material you are working with.
2) The other test we did was filling an aluminium beer can with PP(5) and leaving it in the oven for some time. From the very start I decided to do it gradually by adding plastic in layers (approx.2cm each load up).
For that I took the top of the can off and prepared the ‘batches’ / layers of plastic pellets. The result below is a partially filled can / cylindrical block of PP(5). We don’t know yet what to use it for, but will come up with some ideas.
Photo3 is the ‘mould’ being taken off. Photos4-6 – cylinder / pattern / texture result.
As there is a lot of things involved and this is merely an overview, feel free to post questions here or message me directly.
🌎Weekly update N.11 – PART 2 – ‘Not that cool after all’
This part is going to be mostly about recent failures.
– We still use the same mixed shredded PP(5) that includes lots of various sources of plastic like lawn mower covers, a sledge, food containers, washing powder caps, office supplies. All of these materials come from different places / originally manufactured in various countries. The material slightly differs in melting temperature and this impacts the composition / structural integrity of a part.
The first ever plastic item we shredded was a pack of transparent PP cups. Ironically this is what caused most of the problems for us. Initially we did not consider sorting plastic by colour, as we wanted to start experimenting with the material asap. This lead to a bag of PP mixed in colour. Transparent PP cups, mentioned above, have a slightly higher melting temperature and melt only after all the coloured PP chips are in liquid state.
– When filling an aluminium can with plastic pellets and melting it, keep in mind that it goes faster if done in layers. As I was doing multiple things the same time, the top surface of some layers gt a bit burned. If you put plastic pellets over that burned plastic surface, it will still blend, but might be weak.
– How do you expect a block of plastic to cool down?
Imagine: you have an aluminium can filled with molten plastic. You take it out of the oven and place it onto a table outside. How would it cool down? Some people think, that at the end of the day you will have a cylinder of plastic. Unfortunately, the center of the can stays hot, while the outer side of plastic touching the Al walls cools down. This brings the effect of the center dropping down (like a volcano crater). So in the photos above, roughly a half of the height of that ‘Plastic Can’ is empty in the middle. All we can use it for now is one side of a Yo-Yo. 😎
– When shredding, sort the material not only by type, but by colour as well.
– When melting plastic in an aluminium can (or any other container), keep an eye on time to prevent it from burning.
– Plastic doesn’t stick to burned plastic well. OK, but not great.
– Molten plastic needs to be cooled down gradually, ideally in the oven.
🌎Weekly update N.12
This week we were still tweaking the shredder. As a test we got some office highlighters shredded.
Yes, yes, I remember, I said that you shouldn’t really shred different colours together and should sort it, but this is not the case as we are 100% sure the manufacturer is the same for all the colours (same markers), so this gives us a high chance that this plastic would melt nicely together when the time comes.
We also got a filter / netting installed (in our own way of course)..
Apart from that, we got the auger wooden drill bit, so extrusion machine building is going to start soon-ish..
In september we tweaked our shredder and now our badboy shredder looks like this (first photo)..
Also, you might remember the story about us filling the aluminium beer can with plastic. The second photo shows what a cross section looks like. As you see it is quite porous, mainly due to the fact that no compression was performed and it was just melting in a can. The air stayed trapped.
We tested our shredder and got this nice enough grey HDPE. We will probably use it for the skateboard we’re gonna make at some point.
I think it’s very good idea
We have finished the working model of the injection machine and tested it out. First I have to say that I took quite a while to set up all electronics to work predictably. Then the end bit for the barrel and the injection point link is still being optimised. We tried using ball gate valve, regular pipe reducer and just a carjack press holding the mould. The first tests are seen below – coasters.
Red and blue mixture coaster:
PP in an PMMA / Plexigalss / Acrylic mould / Injection point in the middle
Process temperature set to 290 and 300degC (yes, yes, I know)
Mould cooling in a barrel of water (10mins)
First ever injection test. Badly overheated plastic. Resulted in visible bubbling and additional sinking (surface not flat, bent inwards). The mould was attached directly before injection, so that plaxiglass won’t melt. As the mold was quite precise, the only post-processing was to get rid of some small side flashing.
HDPE in a metal mould / Injection point off center
Process temperature set to 240-250degC.
Slow mould cooling (1h)
Third injection test. Plastic not visibly overheated. Bad plastic smell. Surface sunk a bit, but to an acceptable level. The first plastic bit to touch the mould was a bit cool – resulted in a surface irregularity on the outer side of the injection point. The mould made of metal had some wide gaps – plastic flashing > needed additional finishing up.
Mould arrangemets are shown in the photos below.
Second injection test, which I intentionally skipped, was a fail. Reasons listed below.
Lesson learned: Injection
– Before plastic moulding, you need to perform a heating test. Take a finished injection machine and measure how log it would take to heat up the inside of the barrel. This way you can predict when plastic is starting to melt. This test is yet to be done.
– Do not blindly increase the process temperature (like we did 300degC) hoping to heat up plastic faster as this results in burnt plastic, among the other things like excessive sinking and pressure buildup.
– Optimise the injection arrangement. This is done by trial and error.
– Clean the inside after each use. Sometimes plastic gets stuck to the inner walls and can contaminate your product.
You must be logged in to reply to this topic.