3D Ultimaker Gantry

221 views
Skip to first unread message

Ryan Carlyle

unread,
Feb 14, 2016, 10:52:54 PM2/14/16
to 3DP Ideas
This is a terrible idea, but it's interesting.

I like thinking about what happens if you take a 2-axis gantry concept traditionally used with a moving bed, and extend it into 3-axis gantry for a stationary bed printer. You've seen my CoreXYZ design -- that's the extension of CoreXY into 3 axes.

Here's what a 3D Ultimaker gantry would look like:


It needs four Z stages, four Y stages, and four X stages to drive the central cross gantry. The parts count would be nuts. But it is a three-axis parallel cartesian mechanism with an acceptable dexterity and probably acceptable rigidity for 3D printing, which is a pretty rare creature. 

whosawhatsis

unread,
Feb 14, 2016, 11:50:50 PM2/14/16
to Ryan Carlyle, 3DP Ideas
Would it have "acceptable dexterity"? All those torsion rods are going to add up to a lot of angular momentum, and lots of linear inertia on each axis as well, not to mention the compounding linear slide friction. My guess is it would need significantly lower acceleration than a comparable decently-built CoreXY.

An unfortunate feature of this build is that the Z axis would move most freely.
--
You received this message because you are subscribed to the Google Groups "3DP Ideas" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 3dp-ideas+...@googlegroups.com.
To post to this group, send email to 3dp-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/3dp-ideas/a90aa3a0-7ca7-4dc0-91cb-877e8af87e55%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Ryan Carlyle

unread,
Feb 15, 2016, 12:09:53 AM2/15/16
to 3DP Ideas
Dexterity here in the robotics sense of places reachable by the end-effector without crashing into anything. Many parallel mechanisms move linkages through the working volume below the tool so they're only useful for things like pick-and-place, not 3D printing or CNC.

I don't think I would use UM-style linear + torsion rods for this. That's definitely a lot of angular momentum. Probably go with a lightweight cable sync system like synchromesh, or smaller (5mm?) torsion rods like you see in Replicator 1 style Y stages. The benefit to synchromesh is that the sync and drive cables for each direction could be one continuous run, if you don't mind a lot of idler pulleys.

You could put the four Z stages on screws instead of belts, if you want.

Reply all
Reply to author
Forward
0 new messages