Polar CNC design. Pretty cool!

[Jimustanguitar]

http://www.youtube.com/watch?feature=pl ... gIVZzDyhsA

This will be interesting to follow. Anybody else seen this yet? It certainly seems simple enough to be foolproof and accurate.

[Flateric]

My initial thoughts are..

Very large moving bed that has to support the weight of alot of moving equipment and yet be stable rigid. This would lead to the use of more expensive parts to remain accurate and constant.

Z-axis build height will affect the accuracy more and more and the height of the machine is scaled up.

Not sure how gcode would be generated for such a machine or if this would be handled through the firmware translating it to be accurartely used by the machine?

Those motors are going to have to be fairly decently beefy and require a similarly beefy power supply and supporting drivers to be able to perform the routing well without skipping from the torque of the cutting process. This is not going to be an issue with printing really.

I once many many many years ago made a polar digitizer and the biggest problem I had with it was that the resolution varied depending upon your distance from the center (highest res) to the outside (lowest res) positions. This is something I was not able to ever overcome to a level I found acceptable. Which is not to say smarter minds couldn't.

I personally am against trying to combine a cutting and a printing machine into one unit. I think there are far too many issues in doing so. Dust, vibration and accuracy and are put to different stresses between the two and seperate machine are the way to go regardless of the machine configuration (delta etc)

[Jimustanguitar]

That blog has an algorithm program that converts normal G-code to work in the machine. I think the drawback, especially if you were milling, would be orienting the work in a square and repeatable manner.

For the Z axis, I think a moving tool like on a lot of gantry style wood mills would be best. Trying to move the rotating platform up and down wouldn't work so easily.

To make things rotate smoothly and evenly, I'm thinking about using automotive wheel hubs from a junkyard. That would be much more attainable and true than the mash up of home brewed linear bearings that exist. I think they'd do a decent job, and be inexpensive too.

For the steppers, just use worm gears to hold the torque. The problem would be using gears without lash.


Anyway, I think it's pretty doable. I've got some steppers and drivers sitting around (from scrapped plotters, donated to the local Makerspace). I think I'll give it a try. Not sure how soon though

[626Pilot]

I was expecting this to be one of those cute egg plotters, but it is a cool design. I don't see any significant advantage other than for multi-head systems. If you wanted to have three or four extruders, this would be a decent way to go about it.