[chrismp]

Hi there!

I'm planning on building a steel base cnc mill with Igus Drylin W linear rails and a roller chain or belt drive (leadscrew for the z-axis).
I can get the linear rails for free from my university. They are the 16mm diameter version with single sided adjustable play.
The aluminum extrusions are 8020 with 45x90mm. I'm getting these for free from university as well.
The base frame is planned to be made of steel I-beams with 80x160mm 5mm thickness and some steel tubing with 40x60mm 2mm wall thickness (i have those lying around at home).
All the other parts (z-axis carriage, motor mounts) will be made of 4mm hot rolled steel.

The whole z-axis including the y-axis stepper motor will weigh around 8kg and the y and z axis together should weigh about 16-17kg.

I'm planning on driving the whole thing with four nema23 425oz stepper motors running in bipolar series mode (2.1 amps/phase) on a TB6560 based driver board (but definitely not one of those badly designed ebay ones).

It's inteded use will be routing woods (mostly really soft stuff, but also some beech) with small sections of polyurethane cast into the wood (think ski/snowboard woodcores+sidewalls).

Attached is a couple of (bad) renders of a quick sketch up I made.

What do you think of this design? Do you think it could work?

PS: the thing I'm most worried about is if the 8020 y-axis beam will be strong enough or if I should swap it for some square steel tube (weight? maybe need bigger motors?).

[PaulRowntree]

I have used similar drive configuration on my machine, and I am VERY impressed. I can run the X and Y speeds and accelerations up to the limits of the Mach3 demo limits and it is scary fast. 1400 ipm on both chain-drive axes, and the thing looks like it is trying to lunge out the garage door. My table is now the weakest link.
I have tested the reproducibility on Y after +/-17" excursions, and it comes to within 0.001", with 0.002" accuracy, even at these speeds. I am more worried about X (chains on both sides, common motor drive on the gantry), but have not tested yet. My rails are built up from angles & roller bearings.
Cheers!

[chrismp]

thanks for the reply!
what does your drive config look like (eg what motors, drivers, linear guides)?
your comment makes me pretty confident that i'll be able to get at least some decent speed out of this one even with the Igus rails.

btw, i finally decided against the 8020. with a steel tubing being only slightly heavier, but a lot stiffer it was a no brainer. and it's cheaper too.
only downer is that it takes a little more work to get the linear guides attached.

[PaulRowntree]

My machine is just coming 'on-line' now, but so far, it is better than I expected. I am using bipolar NEMA23's from Kysan Electronics, 3.5 Amps, 2 mH coils, driven by a Gecko 540 interface. I am using an Antek power supply (nominal 45 V, 15 Amps (WAY overkill on current) with 12V and 5V supplies built in. I got the inspiration from Patrick Hood (www.BUildYourCNC.com) to use chain drives. I am using 9 tooth drive sprockets on X and Y, with #25 chains. My X has a chain on either side, while Y has a loop running inside a 6" high Al channel that forms the Y gantry. Both axes advance by 2.25" per stepper rotation; with 10 microsteps there is the theoretical resolution of 0.001"/microstep. Z is a salvaged 10 tpi 1-start acme rod, with a delrin (I think) nut. Total travel is ~24"x48"x7". Patrick now sells a lot of components for these chains, including idlers with bearings (I cheated and used roller-bearings as pullys on both X and Y. A bit weird, but way cheaper, and they seem to work)
The mechanics are a sight : lots of salvaged Al, so it is a one-of-a-kind beast! The X rails are supported off the cutting table surface by ~6" so I can slide large work pieces beneath them . Each X axis is a 2.5"x3" Al I beam, with 1.25" Al angles top&bottom to run the roller-skate bearings on. If I was to do this again I would look harder at the V-groove bearings for X and Y (I use them on Z).
I will get a build note up soon with first-cut photos, and will PM you the link.
Cheers!
Cheers!