3D-Printable 3D-Printers

Pythagoras was a genius. To conceive the mathematical rules of a triangle and recognise a consistent formula for the dimensions and area of any triangle, without so much as a trigonometric lookup table... remarkable. Literally changed the world. And is still doing so.

Now, before I proceed, 3D printing a 3D printer is not an original idea. Josef Prusa makes a living out of printing printers that can print printers. The whole RepRap movement is founded on this concept, but today, I want to share the basic concepts of how it can be done, and look at maximising the printing, while minimising the purchases. Essentially, using Pythagoras to conceptually "print" a printer frame. For example, like this...

A Pythagorean printable printer frame, showing bed alignment of a single part as it might be printed on an earlier generation of itself.

How's this work? Well, Pythagoras worked out that the diagonal of a square is equal to the side of a square time the square root of 2. For instance, the 250mm build area in the above concept sketch has a 353.5mm diagonal. Well more than enough to enclose a rectangle of 300mmx30mm on a diagonal across a 250mm build plate. Then you only need a 310mm build height to complete it... or even a 160mm build height and do to top and bottom frames in 2 halves, each. (This eliminates the need for most support structures in any printer.)

As much of the printer that can be printed should be printed. Designing a printer, that can be printed in 4 quadrants of the frame, also allows for construction of printers that only need mechanical parts for the mech to move (lead screws, linear rails, bearings, wheels, electronics and motors), in materials that don't warp and have high strength (PETG-CF or PC-CF), can allow for machines that can be built, adjusted and calibrated to very high precision, potentially eliminating linear rails from the parts lists. The quadrant schematic looks something like this...

A quadrant printable printer, using inbuilt dowel joints. Concept sketch only.

Now, while there is less potential for failure in the smaller print inherent in the quadrant model, there's a tad more potential for misalignment of the centre joins, while the pythagorean model, being of a larger top and bottom frame, might have more chance of print failure, but, provided stiff hardware is used to vertically link the top and bottom frames. Both models are best suited to a core-XY mechanism, as this provides a more reliable, and more accurate movement, something that needs to be considered. Provided a printer build is well tuned and calibated before printing a copy of itself, being a digital syste, the accuracy of the print shouldn't be prone to generational errors. Obviously, though, a printer printing a printer, the stiffness and accuracy has to be spot-on.

So, why would we want to print printers from printers?

• Rapid and more affordable printfarm rollout for small scale manufacturers,
• Start your own 3D printer brand from an open-source model where the licence allows this,
• Community groups involved in youth development/aid, right-to-repair, community open-fab workshops,
• Charities working to develop technical skills in developing nations or after disater relief.

Naturally, there are many people already working on making printers more accessible, but the dominant efforts at the moment are more focused on commercial/retail printers, like Bambu, Creality, Sovol or Anycubic, to name a few. Many of these commercial operations are using open source projects as their root sources, but many of these companies have started closing the door on repair, 3rd party firmware and user mods, too, and this locks users out of printing replacement parts where it should be a simple, legit thing to do without warranty implications.

However, maintaining development of truly open source, open manufacturing printer design is certainly worth embarking on. I strongly believe that competition is good for human development, but that capitalisation and proprietorialisation of that competition tends to lock away access to human development and deny us rights-to-repair. Yes, I'm a "lefty." Yes, the rep-rap movement presents, as I see it, a way to make a more humanitarian, accessible economic environment.

So, lets compete, lets make machines that can make machines. Lets use open source software and hardware, and keep our derivatives of that as open source. Charge only for the stuff people can't manufacture in their homes, the stuff we have to buy at current technological development levels. Never charge for the designs or the use of those designs.

Lets start a new wave of open Rep Rap Manufacture.