[sp1nm0nkey]

I'm trying to make a robot similar to [ame]http://www.youtube.com/watch?v=8MVV0snHw_g[/ame] . I'm just not too sure how the Z axis works. I can see helical grooves cut into the Z axis rod... thing... so I'm thinking that might have something to do with how it works. Somehow I'm guessing it spirals up and down and the rotational axis compensates. Anyway... any ideas for doing something similar? I was thinking maybe something like a quill feed from a drill press but I like how on their robot there's no rack showing. Capstan maybe? I just really don't want any chance of it slipping.

[Zumba]

The ram is probably ground like a high lead ballscrew, and a belt-driven rotating ballnut is held between ball bearings in the head.

[sp1nm0nkey]

What keeps the ball screw tracking? If the ballnut is spinning then the shaft has the option of moving and down or spinning. I don't see anywhere to grip the screw... if it was made like a combination spline and screw that'd make more sense to me.

[Zumba]

Not sure. Here's my guess...

First note that around 0:25 and 1:13 in the video, the end effector spins around, and the rest of the screw/ram spins with it. I'm guessing the ballnut assembly has a brake. When it's actuated, the Z-axis turns into a C-axis. One motor controls both. That would explain how the indexing works.

So all that's left is ball screw tracking. Note that the ram travel is at least equal to, if not greater than the head height, so whatever's gripping the screw can't possibly be moving up and down. It must be stationary. It must allow the ram to move up and down freely, and also be able to disengage during indexing.

I'm guessing there's a round linear motion bearing around the ram that has zero angular play, and is itself held between two spindle type bearings just like the ballnut. When the Z-axis is in motion, a disc-brake sort of device is "on", holding the linear bearing in place such that the ram can only move vertically, not angularly. When the C-axis is in motion, the disc brake is "off", allowing the ram to rotate. Meanwhile, the locked ballnut prevents unwanted Z-axis travel.

Like I said, just a guess. It would be quite easy to implement though, as long as you can find the required components (ballnut lock, linear bearing).

[sp1nm0nkey]

Hmm... it looks like THK makes ball screw splines with tracks for both ball screw and ball spline nuts. If you put a ball nut and a ball spline next to each other and spun them both the right way, you'd be able to create that motion. So either way... one needs two motors and one needs one motor and a brake.

[Zumba]

I didn't even know they made them... very cool. http://www.thk.es/EN/Products.asp?CO...llscrew+Spline

The ram in the video doesn't appear to have any splines, but that doesn't mean yours can't. I definitely like the splines.