GRBL settings for jog distance?

Software
1

So I moved, and my machine has been off for a while.

Got it back up and running, and GRBL settings are back to defaults.

When I first got the machine, I had to change the Z axis GRBL from 40 to 42.8 to get the Z movement right.

I did some test cuts, verified I still had to make the same kind of change, and changed my 102 value set in GRBL.

I set my height for my next test cuts with the touch probe. I then jogged my router up to get the probe out, started it up, and hit “go” on my test.

…… and my system cut into my clamps moving to X/Y/Z. I checked things again, and it thinks Z is now underneath the work surface (by a little).

So the jog distance manually jogging is smaller than the new Z distance for automatically moving…. I think.

Where can I modify the Z jog distance? Or is this just some crazy fluke, and reset everything and try again?

And thanks in advance!

2

Jog distance increment and movement increment should be the same.

Usually difficulties such as you describe are caused by a disconnect in how the origin is specified in the file and how the zero is set relative to the stock, but that’s a greater adjustment than I’ve typically seen.

If you set Z origin at one point, then move away, then go back to zero you should return to where it was set, invariant of if the movement otherwise was greater or shorter than expected.

Please review:

If you continue to have difficulties please describe step-by-step:

  • what you did
  • what you expected
  • what actually happened

and we’ll do our best to work through this with you.

1 Like
3

Just commenting since I am surprised by such a large steps per mm factor. Back when I bothered to calibrate my axes, I rarely had to adjust outside of the 39.9 to 40.1 range.

I suggest troubleshooting this problem with the default GRBL values initially.

Any chance you might have positioned the probe on the corner of the stock and then probed Z only ?

2 Likes
4

I second the thought that there may be a translation/transposition/rounding error somewhere in the math to figure the steps, 42.8 is the largest value I have ever heard of to be entered in the $102 setting… Like @Julien, I would be wary of any value outside of the range of 40.20 to 39.80…

2 Likes
5

Is my math off then?

Here’s the rundown:
Stock thickness 0.627 (MDF)

Set pocket depth as 0.5
Remainder left: 0.163
So 0.5 depth is only cutting 0.464
Per:
https://shapeokoenthusiasts.gitbook.io/shapeoko-cnc-a-to-z/x-y-z-calibration
0.5/0.464 = 1.07758

40x1.07=42.8
(rounded down so I can run another test and increase if needed.)

6

And I’m measuring by cutting the pockets, then slicing off the end of the board so I can get exact measurements of the depth.

[Album] imgur.com
7

The “deeper” pockets in the above pic are set to 0.627, and therefore should be the whole way through the board.

I’ve also probed XYZ with the touch probe, then put the probe on the stock and re-probed Z only. I get same cut depth.

I believe last time I had to make this adjustment is was the same 1.07 value… about a year or so ago.

8

I would suggest a pair of pockets, one just deep enough to be certain it is cutting, the other some even metric measurement down, but not through the board.

Please write us at support@carbide3d.com and let us know about this — I think we should send you a new belt.

9

That math is not off, but any inaccuracy in measurement (or rounding) would likely be amplified by the tiny depth of cut being measured (similar to falsely extracting statistics using too small a sample size)…

Better to use a method like this:

1 Like
10

Are you sure you don’t have a mechanical issue causing the shallow pockets? There should be no difference between a jog motion and any automatic motion.
If you want to calibrate using pocket depth, I would first face the stock so you’re starting from a perfect zero, and then be conservative with your depth of cut to rule out lost steps.

2 Likes
11

@MichaelStone,

To figure out whether it’s lost steps or belt slipping/stretching during an actual cut, you could first check what distance the router mount travels when jogging 0.5" up or down, with a ruler next to it. A difference of 0.04" (like you got in that pocket cut) will be visible to the naked eye, but I would not be surprised that when jogging in the air you get very close to 0.5" actual travel, and the problem only occurs during cutting.

2 Likes
12

This topic was automatically closed after 30 days. New replies are no longer allowed.