CoreXYZ?

Also posted in the CoreXY section:

While waiting for the parts to my Mini Kossel to arrive, I started thinking about other printer designs that fulfill a similar function. As everyone knows, the purpose of a delta printer is to print fast and look cool (yes, people also say that they are mechanically simpler, more easily scalable, etc. when they don't want to use "looking cool" as a justification for purchasing one). Anyway, I became aware of the corexy system a couple of days ago. And while it appears simple, fast, and cool-looking, it is constrained to two dimensions with the only partial exception that I have seen being that corexz [www.youtube.com]. Seeing a gossamer, filament-powered system on top of a heavy-looking print bed cantilevered off of a lead screw seems somehow wrong to me, so I endeavored to design a system to add an extra dimension to the corexy in order to maneuver the hot end in all three dimensions while the motors and print bed stay stationary. Below are the two systems I came up with over lunch.

The Corexyz (mentally superimpose the three systems in the column on the left of the attached photo of my sketchpad):
Take your standard issue corexy. Fold it upward 90 degrees one third of the way up. Fold it back 90 degrees at the two thirds point. Add the y-axis system on the bottom. Voila. In the xz plane it works just like a corexy. All three motors, however, must power movement in the y axis. The printer would look like a scaled-down cnc router with a moving gantry. It uses a total of 22 pulleys, and if we are to assume a cube-shaped print area of side s, each cable would be about 6s long. I would anticipate six linear motion shafts and 12 bearings. Pulleys are small circles in the drawing, tensioning and/or connection points in the line are dots, and spools are the small cylinders.

The Corex+y+z (mentally superimpose the three systems in the column to the right. Trying to draw it all together gave me a nasty headache that I would hate to inflict on unsuspecting forum members):
This one accomplishes the same in a different manner. I envision a very attractive printer built between two acrylic walls powered by a column of dual-shaft steppers. Anyway, the y axis moves two linear motion shafts back and forth by pulling on their tops and bottoms. The z axis yanks line from the top half of the printer to the bottom half and vice versa to lift and lower the x axis. The x and z lines must constantly recirculate as the effector moves in a different direction. This one has a total of 34 pulleys. The Y-strings are about 4.8s. The Zs are 5s. And the X-line is 7s. Anticipate eight >s-length linear motion shafts and 16 linear bearings.

Is anyone working on anything similar?
Does anyone have any thoughts/critiques for either or both systems?
Am I missing something crucial/barking up the wrong tree?
What is the answer to the great question of life, the universe, and everything?

Some issues I would anticipate finding if I were to build one of these systems:
Friction - This many pulleys can't be good. If a gantry pulley bound, one would quickly end up with PLA spaghetti.
Mechanical advantage - In the Corex+y+z, the x and z lines move with twice the power/half the speed of the y-axis (to be corrected by either the spool or the programming).
Stretching and sagging - The cables in the corexyz are 50% longer than in an s by s corexy system.
Abrasion - The whole crisscrossing thing on the y axis of the Corex+y+z.

I am most definitely interested in a CoreXYZ, I'm also not too far into my build, if someone could lay down a firm, basic design I could adapt it to what I have and build it to test. That wouldn't be any time soon, though, as progress is slow.
Either way, I've been waiting to see a CoreXYZ for a long time. maybe try contacting the same people who did CoreXY and getting their thoughts?

---

Realizer- One who realizes dreams by making them a reality either by possibility or by completion. Also creating or renewing hopes of dreams.
"keep in mind, even the best printer can not print with the best filament if the user is the problem." -Ohmarinus

There are a few things I would change (I mentioned them in the other thread you made for this) and there are a few places where you can simplify each stringing path, but mainly you have to answer the ace old question of "why?".

And then you ask "Why not Core XYZE?" where there's a fourth belt to remotely drive an extruder at the head.

Necro-posting a bit here, but I just wanted to mention that I have a working* CoreXYZ design. I consider this the "true" CoreXYZ implementation. It's a four-line 3D gantry that extends the CoreXY concept into three dimensions. All four belt paths are symmetrical, and all drive forces are balanced (no racking).
Diagrams, kinematics, and build log here: [groups.google.com]

* Ok, it's not printing. Why? Because it needs four-axis motion control to operate right. It's redundantly actuated (over-constrained) and none of the major platforms support that. I have a three-motor (w/ idler fourth line) version built using a special Sailfish build to do the kinematics, and it does indeed produce the desired directions of motion in jog mode. But a lot of backlash is introduced by not driving the fourth axis. I tabled the project a while back to look for a 3DP controller that can handle four synchronized motion axes... But that's been slow going. Particularly since I'm an engineer, not a programmer :-)

Just wanted to post so people looking at the concept know that it's a "solved" design problem in need of some more implementation horsepower. If anyone's interested in working on a firmware/controller that can run it, let me know. The kinematics are dead simple, it's really just getting the fourth synchronized motion channel running in a 3D printer controller that is causing issues.

I might be interested in implementing this in RepRapFirmware for Duet electronics. You would need the Duet Shield add-on to support more than 4 drives though, so that you can extrude as well. The google groups thread that you linked to is quite long, so can you point me to where you define the mapping from Cartesian coordinates to motor coordinates? I would also need to know how the printer should be homed.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I've never seen it done on a 3D printer before, but I have once seen someone simply wire two steppers in parallel to one controller and it appeared to work as expected, with both steppers moving synchronized with each other. It was a very temporary kind of build though and I don't know if there are negative consequences to such a shortcut.