Power Route complete!

[ttabbal]

Got it built, jogged around to verify that everything moves properly and adjusted the homing switches. Home works properly and everything moves around as one would expect.

Since the instructions don't cover it, the next steps I have in mind..

Chuck up an end mill and put a test board on the bed. Connect the probe clip to the end mill and the puck on the wood, run a probe cycle to set Z height. Set up the work origin. Load a file and go.

I also need to set up the laser. It's mounted and wires run, but I haven't set up the control board yet.

[Bruce]

Congrats on your Power Route build!

[ttabbal]

Thanks. Z probe fails here. Either I get an error "Probe is not in the expected initial state before starting probe cycle when G38.2 and G38.3 is not triggered and G38.4 and G38.5 is triggered." with 3 dialog boxes to clear, annoying. Or it seems to start, then nearly immediately stops and appears to think it's done without touching the probe. I start with the bit about 20mm above the plate, so it is clear of the Z switch. I'm not sure what to make of it.

[Derek the Admin]

Congrats on the build. What is on your Z read out prior to issuing a probe command, and what exact probe command are you issuing?

[ttabbal]

I tried again just now. Z was at -89 (machine). I was trying to use the probe command in UGS platform. The console output looked like this...

>>> G21 G91 G49
>>> G38.2 Z-30 F5
>>> G90 G21
ok
[PRB:-594.000,-464.000,-79.000:1]
ok
ok
>>> G21 G90
>>> G43.1 Z-99
>>> G90 G21
ok
ok
ok


I just ran the MillRight CNC logo gcode file, it worked great, ignoring the flaking due to the random piece of chipboard I slapped on the machine. That's the material's fault, the machine performed flawlessly. I'm new to CNC gcode, so I'm not sure what the command for a probe would be if I type it in.

[Derek the Admin]

Congrats on a successful first run.

The reading of the Z in the work coordinate system is important when probing. The Z value in the WCS has to higher than the Z parameter of the probe command.

[ttabbal]

So, should I zero the WCS before trying to probe? I can make it work with feeler gauges etc., but I would like to get it working.

[Bruce]

It might be easier to jog the bit close to the zero area and the set the work coordinate to zero there.  Then "fine tune" to zero with the probe.
Here is Derek's video on the whole subject.


Homing and Work Coordinate Systems using Grbl with Millright CNC

[ttabbal]

Yeah, the video doesn't cover the probe though. That's where I'm going wrong. I'll try to get it close and zero it. I don't think I zeroed before.

I can, and did, use the feeler method. Just paper, but it was just a test. I have real feeler gauges that I can use while I sort out what I've done wrong with the probe.

[Derek the Admin]

Just let’s try to zero the Z position prior to issuing your Z probe command. Unless we have a crossed wire on our end of things it’s just a matter of making sure it is above the position in Z to which it is being told to probe downward to.

[ttabbal]

Thanks. I zeroed out the WCS and tried the probe. Instant trigger.

So I thought it might be wiring related. I removed the clip from the endmill. Now the probe runs as I would expect, running until I touch the clip to the plate. So it appears there is something off causing it to ground through the frame or something like that. Could the touch plate wiring be reversed?

[Derek the Admin]

It could be (or couldn’t be). There’s a couple ways to flip them. You can literally cut and reattached the leads to the opposite wire (probably easiest) or you can go in the box and do it (box must be u plugged from the wall). Looking inside the box the left most wire set coming off the plug has two reds and two blacks. The bottom red and bottom black are probe signal and probe ground. You can check continuity between the pins and the wires coming off the back plug to find which is which and swap them. Or you can email [e-mail] with some pictures of the wires running into the breakout board and that back 4 wire plug and we can send something flipped the opposite direction.

[ttabbal]

That did it! Thanks! I crimped some connectors on the wires and swapped them. Now the probe works properly. Glad it was an easy fix.

[Derek the Admin]

Awesome. Sorry about that. That honestly had me scared we had a wider problem because this happened a couple times many months ago. I actually went into the shop and checked the built boxes worried all of them were like that. They checked out. Looks like you were the unlucky victim of our goof.

[ttabbal]

That's my luck all right! No big deal, I honestly thought I was doing something wrong. Now I can get to making some stuff once I figure out the software a bit more. I think I have the clamp file from the website set up in Fusion360 now. Perhaps I'll try that once I get a spoilboard set up.

A bit in the future for me, but could a switch be wired up in line and used as a probe for something like www.scorchworks.com/Gcoderipper/gcoderipper.html ?