Moving in the Z axis

The recent posting on the builders blog, on repstrapping with an old Sherline desktop mill has me wondering why isn

Um... Tommelise is built in the way you suggest. It works fine.

[www.3DReplicators.com]

The Sherline/Bridgeport mill design, with the milling head going up and down (z axis), and the piece attatched to the x-y table, has good rigidity and low vibration, which is important in a mill you might be cutting steel with, but the x-y table takes up 4 times the space as having a head that moves in x-y. If we did that with Darwin, we'd have a (60cm+)by(60cm+) / (2ft+)by(2ft+) foot print, compared to the current (30cm+)by(30cm+) / (1ft+)by(1ft+) footprint.

Darwin's design is basically a gantry type mill, only with a dropping table. A dropping table is better if you can use it, since you're decoupling the z axis mechanism from the x-y axis mechanisms. Since we only move the table down in one print job, as opposed to bobbing it up and down as we would be in a cnc router, it's doesn't seem too hard to keep it flat. (I don't have my hands on a set of Darwin parts yet, but I think this is the case.)

"Since we only move the table down in one print job, as opposed to bobbing it up and down as we would be in a cnc router, it's doesn't seem too hard to keep it flat."

In practice you may find that that is not quite true. If you print a track of extrusion and then need to move to another part of the print you need to raise the print head (or drop the table in Darwin's case) for a distance to avoid dragging a hot extruder head across plastic that you've already printed and then drop it down to the current print level again.

This isn't a big deal, mind, but the z-axis gets used more in practice than you might imagine in theory.

This is all really interesting... hopefully there will be more movies uploaded as devices are built, so I can see just how much movement there is on the Z axis.

The reason I'm asking is that I have the impression the CNC crowd would recoil in horror if presented with the Darwin design. I'm not so much interested in a self replicating machine, as interesting as that is, so much as I am a multi-purpose fabricator. So I'd rather just build one thing, rather than a series of ever improving things. (or hopefully a short series of ever improving things with big productive gaps in between)

The big difference between CNC mills and FDM is that a mill needs to be extremely rigid to handle the lateral force generated when the tool meets the workpiece. I learnt this the hard way. Darwin can get by with a cheap skeleton of rods because there is no force through the extruder barrel.

You can turn a CNC mill into an FDM machine by changing the head, but not vica versa.

Edited 2 time(s). Last edit at 05/29/2007 03:56PM by nophead.

---

[www.hydraraptor.blogspot.com]

Right, but we should be able to design a bulky design made from plastic that looks like a generic desktop cnc routers but is made with lots of RepRapped parts. Such a machine should be able to do light aluminum milling.

However, right now, without a cheap 3D printer/cheap 3D printer feedstock, it would be expensive to develop, because it would take a lot of plastic to make. Something to worry about after the 1.0 release, maybe.

I daydream about the larger CNC routers that have a 4' x 8' (1.2 m x 2.4 m) workspace and can handle full-size plywood sheets. I wonder if it is possible to make on of those using a RepRap and commercially-sourced parts.