Subscribe to Make Magazine Today!

I have written before about Thingiverse user fdavies’ ongoing project to produce a 3D-printer that requires no precision-ground shafting or bearings using printable hinged actuators based on the Sarrus linkage (Wikipedia). Why would you want to do that? Well, because precision shafting and bearings are currently beyond the abilities of most 3D printers, and if you can build the printer itself using printable substitutes for them, then you’re that much closer to a truly self-replicating home fab system. His latest effort has produced a working two-axis system that he hopes to outfit with an extruder in the near future. Keep it up, man!

Sean Michael Ragan

Sean Michael Ragan

I am descended from 5,000 generations of tool-using primates. Also, I went to college and stuff. I am a long-time contributor to MAKE magazine and My work has also appeared in ReadyMade, c’t – Magazin für Computertechnik, and The Wall Street Journal.


    It could really stand to be tagged “game-changer”.

    Also, some of the nicer features of this gantry should be noted. The righthand-most feature in the center of the image above is a motion and torsion-resisting printed spring (dubbed Spider-Spring by me, even though it has twelve elements)used to keep the belt under tension. It is pretty stiff, but does still give slightly.

    The motors in use are both basic DC motors – Frank’s design is based on getting positional accuracy by combining cheap DC motors and linear encoders (usually both salvaged from cheap printers). The motor is controlled by an arduino that takes the incoming pulses meant for stepper motors and uses that information to figure out where its DC motors need to move the linear encoder (and therefore the stage). fdavies essentially uses the Arduino as a hardware-based hardware abstraction layer to avoid having to do more time-intensive code changes to other software and firmware.

    Finally, assembly of these items is pretty easy and straightforward. They are printed with all of the holes for screws and hinging pieces (either brass rod or copper wire) already in place and just needing a little clearing and sizing. I assembled most of a single axis on my own without supervision or instructions while fdavies was assembling the first version of the two-axis table above. Parts make sense in the context of the larger device and it’s exceedingly difficult to assemble something out-of-alignment.

    The only “vitamins” (RepRap term for unprintable but necessary parts) needed are about 12 inches of brass rod or 10-gauge copper wire and a large handful of bolts and nuts. In that regard, any eventual printer based on this design will be _much_ more readily printed on existing machines. It is worth noting that the individual components used are all small enough that they can be printed on a Makerbot!

  • Robert

    If I could comment on thingiverse I would suggest using the optical encoders from old ball-mice rather than printers. Cheaper and easier to obtain.


      Very true – however, you’re already tearing a printer apart for the long encoder strip. You might as well go ahead and remove the encoder and motor while you’re at it.

  • stunmonkey

    I like the ingenuity shown here, but can’t help that think it is pretty misplaced.

    The idea of sustainability and reproducibility is important, but I think the idea of self-replicating is a silly party trick.
    Even if you ignore for a minute that self-replicating is a lie anyway as it can only ever replicate some parts of some of the machine systems, the difficult parts still have to come from elsewhere, it still makes no sense.

    There is more than one way to ensure its easy reproducibility, and all this effort driven into “self-replicating” natures is taking these projects off the rails and away from many of the the initial intents of this kind of project.

    A lathe is pretty rudimentary equipment, in fact one that can be generated from virtually nothing and is rather self-replicating. If you can’t make and maintain a lathe, you can’t possibly expect to pull of a 3D printer.
    Given that, wouldn’t it be nice to just accept the use of a lathe as a precursor and just make simple, cheap, durable linear rails?
    You are allowed to use more than one tool.

    What do people really want? The most simple and effective machine that can easily and cheaply be made anywhere, regardless of whether or not you have to “cheat” and use more than one tool, or a complicated and fragile machine that you can brag is “self-replicating” (even forgetting the fact it still isn’t, mind you)?

    This is spending more time trying to go Uber Meta than it is trying to be easy and practical to make or own. Please people, drop the “self-replicating” and pick up a second tool, or even a third.

  • Chris

    I agree with stunmonkey – the design is very cool and notable for ingenuity, but I think the efforts are going into the wrong place. “Self-replicating” is a good idea for parts that have to be machined into intricate shapes, but precision rods can be purchased easily from sites like McMaster, and will give _much_ higher tolerances than the linkages here. Unless that last part is wrong? If you can use these linkages to give equally high print tolerances as precision linear bearings, then I’d be willing to accept this design. My intuition though is that the designs would be better off accepting a few off-the-shelf parts.

  • Bob

    I think you guys are missing a key point here. A big part of the reprap movement was to bring 3D printing down to a level where anyone who wanted one could afford it. Before they got in the game, it was really only viable for larger corporations. The big advantage of this design is that by making the XY axes completely by printed plastic parts, the cost is almost nothing. Anyone who has one of these machines knows how cheap it is to print parts on them, especially if you buy filament in bulk. While making the XY axes in this method may be less accurate, it will be must less expensive than using linear guide rails and ACME screws or some similar “precision” method. For that reason, this design is a big step in the right because it is lowering the cost of the overall machine. Obviously those that want better accuracy and who can afford it can pursue a different route.

  • old maker

    Is it only me or when I look at this, I see is the Eagle Transporter from Space 1999.

  • air max pas cher

    examined present just about all invisible files within storage space nonetheless zero htaccess document. my partner and i dont get it. it is insane, considerably help required in this article. almost all i want to perform is usually be capable of upload plug ins with hubpages hi there jamesi enhanced by starhub cable connection high speed broadband 50 mbps in order to fibre braodband hundred mbps i find fibre is actually sluggish when compared with cable and becos on the nonsense offered around food fibre created typically the move and found that will newly pvp bot router additionally are not able to use not compatiblebetter in order to stick to cable tv broadbandbest rgdshozefa
    [url=]air max pas cher[/url]

Related Supplies at Maker Shed