Automatically Designing 3D Printers From E-Waste

No one wants to design consumer electronics that last longer than a few years. This trend is an ecological disaster, with millions of tons of computers, printers, fax machines and cell phones ending up in landfills. In these landfills, all the lead and chemicals used to extract minuscule amounts of gold plating leech into the environment. Turning it all around is monumental, but reusing some of this waste can help make a difference.

[Masterperson] and [Maaphoo] have been working on a way to turn those tons of e-waste into something useful. They’ve come up with a framework for turning e-waste into 3D printers. With a clever application of Python and FreeCAD Macros, this project can generate a model of a 3D printer using motors, shafts, and bearings taken from discarded 2D printers.

Right now a printer can be configured by adding the parts you have on hand to a configuration file, running a Python macro in FreeCAD, and waiting until the macro generates the parts to build a cartesian bot. This macro also spits out the files for the parts that need to be printed, and can interface with Plater to optimize the placement of these printed parts on an existing printer.

It’s a very cool project, but it’s not done yet: the team is looking for help to refine the printer designs and possibly growing more designs than a simple cartesian bot. Anything that is explicitly designed to pick the meat off of 2D printers is a great idea, and turning those into real 3D printers is the cherry on top.

12 thoughts on “Automatically Designing 3D Printers From E-Waste

  1. Certainly neat and cool. But but I couldn’t help myself from thinking “How appropriate – a machine made out of e-waste that generates plastic waste.” That almost sounds like an art project ^_^

    1. The community is working on closing the loop, I am printing a knife mill I designed using 3d printed parts, hacksaw blades, 800w motor and fiberglass composite right now. It should be able to grind most plastics up to be recycled in a filament extruder.

      The real problem with this e-waste is having the software hacking skills to use DC motors and encoders. Most 2-d printers I have run across have no stepper motors inside.

      1. DC motors and encoders from printers are easy and fun to use, not a problem at all. They are not very much more complicated to use than steppers, and they are fast, quiet, and accurate. Comparison: A stepper requires a step/dir controller driving a power stage with 4 big transistors. A DC motor/encoder setup also requires a step/dir controller that reads the encoder, and drives a power stage with 4 big transistors in H-bridge configuration. I’ve done it both ways and prefer DC motors/encoders actually!!

        http://www.dansworkshop.com/2015/04/cnc-wood-router/

        I was inspired by an article here on hackaday:

        http://hackaday.com/2015/01/20/closed-loop-control-for-3d-printers/

          1. 1/4 revolution – not a step – but whatever – the vagaries of dc servo control with optical feedback are immense and complex compared to stepper control.

    2. I used to make art.. One of my projects was a printer that had it’s output fed directly into a shredder…

      I also made an alarm clock that smashed itself the first time it rang.

      One critic called me “from the ACME school of fine art”

  2. This is a really cool project. With a large enough library, and even a way to simply tell it what printers you have, not specific parts, this could even be useful.

    Easy to imagine further extensions where you can tell it what you have, and it tells you what printers you could use to finish, it can specify what to buy to finish quickly, and so on.

  3. Interesting. Heck, you don’t need perfect 90 degree angle axis. Another idea that came about when reading this is that you can theoretically just project xyz into some uvw axis, using some good-old vector math? You could just gobble parts together, figure out the axis conversion.. theoretically no need to make more printed plastic parts for the final result.

    1. Yes, although obviously you’d need to ensure that each “axis” had some component in three tangential directions as it would be a bit pointless if uvw were parallel to each other!

      I suspect you’d hit limitations with quantisation effects (think of a set of pixels and trying to draw a line that is not quite at the same angle – it gets to a point where the maths says “jump to the next row”) so you may impair resolution but I think that would depend upon what shapes/motions you’re trying to reproduce.

Leave a Reply to daveboltmanCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.