Keep it simple

I am a big fan of simple, I have always striven for simplicity in my designs.  Recently having been inspired by the clean and elegant design of the Deltamaker printer, I challenged myself to come up with the simplest printer design I could.  After several iterations, all non-cartesian, I came up with the Polar based configuration, shown here.  Einbot ( a nod to A.E.) has only 4 moving components.  Some of the features as can be seen in the brief (53 sec.) Proof of Concept Demonstration Model video below, are as follows:

    -  A directly driven turntable style build platform (equiv. to "Y" axis).
    -  A single directly driven swing arm (with integrated print head). (equiv. to "X" axis).
    -  A directly driven lead screw actuator that translates the build platform in the Z axis, descending ( I am also a big fan of Gravity).
    -  A linear filament feeder ( shown, but not in detail).

[www.youtube.com]

Although the hardware and math are straightforward - I have also included a proof that shows that 'A' point and therefore 'ANY' point in the Cartesian grid can be located by setting two angles of rotation - I expect the coding, particularly the tool path G-code is a bit trickier.  I think it might be as easy as an on the fly conversion or a look-up table, but this is way outside my playpen.  Any helpful thoughts on this subject will be appreciated.

Because this swing arm design will benefit from low mass at the tip, I am also designing a new style light weight filament feeder.  Should that not perform as expected, I will probably go with a more conventional gear drive/hobbed bolt system but chassis mounted the stepper and use a drive shaft or flex drive.  This adds a little complication but will have a consistent feed rate.

Next step:  I may build an operational prototype of the printer, but first I want to further develop and test the filament feeder.  More on that later.

Your video is marked as private?

First, I like your direction!

A few thoughts...
1) There is a singularity in the middle that will cause you to reduce your speed drastically when you are anywhere close to the middle. This can be solved by not placing parts over the center but occasionally you will want to.
2) Your machine will have to have perfect alignment to actually print in the middle. If you are out of alignment by 1mm you will have a 2mm unprintable region. Rigidity also comes into play here. Say you want 50 micron resolution. This would mean that the hot end would have to arc within 25 microns of the center of the turn table. Trying to reduce the flex of your parts down to that level could be hard.
3) I would move the light arm up instead of lowering the heavy turntable.
4) Have you considered a bowden tube setup. It gets more bad press than it deserves. A well tuned bowden will be better than you need.
5) If you build it someone will help you with the math. I actually have a coordinate transformation program that takes in normal gcode and outputs gcode that you can run on generic cartesian firmware. It would be quick to put in the kinematics for this machine.
6) I would look into ways to remove the linear rails on the z. This is easy if you want to sacrifice the middle of your platform.
7) If you could come up with a way for you to get an platform that doesn't continually rotate about a fixed point you will be able to get rid of the dead space. Polar3D

Good luck! If you haven't already, read through the other posts in this forum. There are a quite a few existing designs and ideas that should accelerate you. I think there is a need to build a machine in this style. I keep on going back to it but I am never happy with the results when I put pen to paper. Maybe you will come up with something that will help this design gel.

Thanks, I'll fix that.

Nice concept!

I do have one idea similar to this concept (but its the arm that goes up on the Z) and from that i come up with one similar to the PiMaker.

I do hope that the software for this type of printers gets some development because thats the reason my ideas are not getting more developments. I do not have the skills to do software development

Keep us posted!

Cheers,
Paulo

For what it's worth - my just previous iteration for this design moved the arm not the platform, but I realized that it would be easier to stabilize the swinging mass by mounting it on the stationary frame. Plus, the rotating platform is inherently more balanced and therefore requires less constraint. In addition it just seems easier and more accurate to work with gravity than against it - although I'll bet that can't be proven mathematically!

Thanks for your comments,

Bob