My CNC Mill
- March 27th, 2011
- Posted in CNC
- By punish3r
- Write comment
0
Stay Tuned!
It isn't Done until it's Documented!
Welcome to HackHut
HackHut is a free hosting service for hackers, DIYers, etc. to document their work. Each site in the network is independently run by its owner. Sign-up for your own site here on HackHut, and start documenting your hacks today completely free.
-Return to the HackHut main page-
Recent Posts
Pages
Archives
- March 2011 (2)
Control Panel
LINKS
- Copyright © 2011 It isn't Done until it's Documented!
- Theme designed by mono-lab
- Powered by WordPress
Yes, using EMC2, a digital caliper, a dial indicator and a micrometer the worst-case tolerances are in the neighborhood of 0.003 after a rough tramming, NOT accounting for table or gantry deflection (it’s MDF and acrylic after all…) and offsetting any backlash by making like-handed axis approaches on every move. Repeatability is 0.002 at worst so far, but I haven’t made enough observations of either to reliably compute Cp and Cpk values for where I want it to be.
I’m hoping I can pull those numbers down a bit (obviously).
It IS an open-loop stepper system, so my assumption as of right now is that I can’t get much better than where it’s at right now by tweaking the software, regardless of what the statistical distribution currently looks like.
As an unrelated side note to anyone who has bothered to read this far: SAVE YOUR HEARTACHE AND USE EMC2!!!!! I ran this rig with Mach3 for testing and it sounded like a lawnmower chewing on gravel. On EMC2 it literally SINGS and my rapids are roughly twice what they were on Mach3 (14 IPM on Mach3 to 25+ IPM on EMC2).
So if Time = Money and
Velocity = Distance / Time,
then Velocity = Distance / Money and assuming a uniform Distance:
Velocity = Money^-1
So when Velocity goes up, Money goes down.
and I don’t know anyone who wants to get LESS performance per dollar than they get currently!
Awexome! Please do finish documenting everything, as I’d love to build one of these… or something similar. Your build looks very impressive! Have you cut anything with it yet?
I have done isolation routing on PCBs, but I don’t know if that counts as “really” cutting. I have cut some vinyl stock and “sheet goods” with a modified X-Acto X3241, but again even cutting isn’t “cutting” in this case.
Since the design is entirely dependent on the weight of the gantry itself to stabilize the cutter, this particular design has some severe limitations on material capacity and feedrates.
If you’re interested, I actually purchased the gantry as a kit which I think is still available. **I just checked on availability; apparently you now have to buy everything (motors, drives, gantry, drivescrews, breakout board and Z-Axis holder) at once and it’s $425 US. I did NOT spend that much…
I would say PCB routing is definitely cutting, but obviously cutting something out of a stock piece of material is kind of a good mark of “yes it cuts”. Have you done any testing to see what kind of tolerance your setup has and how repeatable is it? I would love to build a small and accurate mill before I go large, but I also wish I had a laser cutter and a whole slew of other tools. The problem is finding the time to use them all effectively.
That’s what I’ve been finding in the DIY CNC market… some people will start to put things together and claim it’s open source and all… the price is cheap, and they are not making a huge profit or anything. And then the next thing you know they are selling KITS and marking the price way up, because there is a booming market for this stuff and they just can’t product it fast enough.
When demand is higher than supply, the price usually goes up until the demand stabilizes with what can be supplied. I guess I don’t blame them for doing what seems natural though, having fun designing something and then getting paid for it.