v4 Shredder Development
I’m Friedrich from Germany – a Industrial Designer and Engineer. I will work on the next revision of the Precious Plastic shredder in Eindhoven until the beginning of the next year.
As most of you know there are some things about the shredder which can be improved to build a more reliable/productive and safer machine. The recent weeks I ran some tests with the current shredder, a fast-spinning shredder and a double axis shredder to compare them against each other.
Some points which we are sure we want to improve:
Right now we have a max output of ~10kg/h – to run a workshop or small business with our shredder we need at least 40-50kg/h. The goal should be to shred in half a day (or less) the complete plastic for a week.
The next version should not jam that often as the current shredder does. This is not just a problem of the motor power only but also concept related. Furthermore we want to implement automatic reverse (Torque detection) and back to normal operation.
This is very related to the topic of reliability, due it is not possible to build a save shredder if it is not processing the input material in a reliable manner. The shredder should run without any correction of the user during the shredding process. The shredding blades must not be accessible during the operation, this is ensured by mechanic constraints or sensors.
Some of you report a fast wear down of the shredding blades. The new version should have a way to reduce the necessary time for maintenance and changing of blades. I will try to increase the lifetime of the blades by optimizing tolerances, cutting angles and material selection.
The revision of machines will come with a guideline for a easier part-selection process depending on personal needs.
So how will we start?
Our biggest problem right now is the huuge variety of plastic we recycle – small/big, thin/thick, flat/geometric and different materials. And most of the shredders are specified on one particular task. E.g. slow running Granulators to recycle runners of the injection process.
I found a different concept from our existing one which i want to adapt to a smaller size shredder. I think it can solve the problem about the big variety in the best way. The biggest benefit of this shredder type is that, the process is more reliable and you can throw in even bigger rests of plastic like left-overs from the extrusion machine or half injected parts.
Its called single shaft shredder which looks like those ones here:
Of course our shredder will be a bit smaller 😉
The build will be similar to the current one (stacked blades on a hex-shaft) – just the blades will be between 10-20mm thick, to assemble the teeths on them. The width will be the same for all, but the length will be adjustable to your needs (education vs. production)
What do you think about this concept?
I you look at the Weima video 0:06 it appears to be using carbide inserts (such as used for lathe tooling) for the teeth. This would allow using a cheaper softer material for the bulk of the cutter and you can just replace (or rotate) the inserts when necessary.
I would recommend a dual shaft approach though, they seem to grab material and feed it through much faster. Also as you can make the cutters half the diameter to cover the same shredding area, you need less torque = less power. My dual shaft shredder is only 1/2Hp (375watts) and it rarely jams unless I really push it.
I also implemented an auto jam detect/reverse/restart feature, quite easily done with a microcontroller and an encoder on the shaft. What you can’t see on the video is that it will try several times to clear a jam (almost always works first time) but if it can’t it will turn off the machine.
(For some reason the video doesn’t start in the right place, skip forward to 3:15)
Looks like a very nice shredder!
I would be interested to hear why you decided for an encoder based solution to detect jams?
I tested also a 1.1kW dual shaft shredder quite extensively, they work perfect on thin materials such as bottles, single use plastic products and especially flat items.
My big problem with them was, as soon as I inserted harder and bigger pieces it tended to jam due as soon as they cut deeper into the material, the direction has to be reversed, the part removed and repositioned by hand. This is always a dangerous process, due the machine is running while you have to hold the part in the optimal position to be feeded inside.
Where in other cases the parts just are not feeded and jam the inlet of more material (see picture below)
You would need enough pressure from a pusher which breaks those parts. You cannot just fill the hopper with material and walk away -> so in a workshop there needs to be always one person standing next to the shredder and feed the shredding material in the best direction – piece by piece.
For bottles and parts that are under a certain ratio to the blade thickness/size a double axe shredder is propably the best choice, but for random plastic I don’t belive its the best choice. Unless you build a really big one with a blade-diameter of >20-30cm.
Nice shredder 🙂
I saw a shredder that used saw blades a couple years ago. Used in Central America.
Cant find the video anymore.
I did see the video by Jeremy Fielding. It looks promising but needs to feed from the top thru a long hopper to keep fingers out and flying chips from being flung out backwards
Hi, yes I have seen the saw-blade based “shredder”.
Dave made experiments before the V1 of the shredder already. It is a very inexpensive and tempting solution to go for!
The problem about the setup is that the user has to feed in the material, take it off again and feed it in in another direction. And as soon as you want to divide small granulate into smaller granulate you have a real safety issue.
It might be a perfect solution for flat sheets or to divide thicker pieces for shredding. But overall this solution is not universal enough for our huge variety of plastic.
Have a nice day
I used an encoder because it’s a simple, cheap and reliable way of providing speed feedback to the controller. The one I used is really simple (made it myself) only 4 pulses per revolution, no quadrature, I just time the interval between pulses, longer than a set time and the motor is bogging down or has stalled. I also put the encoder on the far end of the driven shaft, furthest from the motor, so any problem anywhere along the drivetrain will show up.
Thank you for the information!
I was already planning to measure the current of the motor consumes, to create a torque-amp relation. So there you would see the torque curve raising up if the motor stalls.
I will look into the electronics more after I finished my current mechanical build. Propably it’s worth to check which option is cheaper and easier to build. I assume the output will be usable in both ways for our purpose..
You must be logged in to reply to this topic.