Twice now I have had a Z height error mid cut. Going along fine and suddenly plunges deeper. Both times using 1/4 downcut during roughing. Any ideas? bit seems stable but has cost 2 projects now.
@FinchCreations Welcome to the group. Before anyone here will be able to help, you need to provide much more information:
- What CNC machine do you have?
- If you are running gSender, what version are you using?
- If you are not running gSender, what code sender are you using?
- Is this a new issue, or have you always had this problem?
- If this is new, what has changed since your last successful project?
- How thick is the material that you are having this issue with? Could you be hitting the top of the Z travel?
Alt mill 4x4
Latest G-Sender
has happened twice on Walnut, only 1/4 downcut bit
1 3/4 piece, very random, and out of the blue, code all good as a rezero of Z works, but too late as i ut too deep randomly
With random locations rather than one specific spot, my first thought is depth of cut and collet issues. I had the same issue at one point. Are you cleaning the collet and bit between tool changes to make sure there is no dust getting in between the bit and the collet? Are you sure that you are tightening the collet nut enough?
Try running with a shallower depth of cut and clean the collet (inside and outside), the nut and the taper the collet seats into.
Note that collets are considered ‘consumable’ in case you are doing production level work (I have never run into this).
If it happens again, bring it back to Z0 and see if the end of the mill is where it’s supposed to be. If not, something likely moved as @Jens said. Put a mark on your endmill’s shaft to see if it moves inside the collet.
Another possibility, but I’m not sure if this can happen with the Altmill’s closed loop steppers, is that a glitch made the spindle plunge. One such source of glitches is static electricity, so make sure everything, including your dust collection system is grounded. Another is the use of USB instead of Ethernet. Ethernet is better and you might want to connect to the SLB-EXT using that protocol instead.
It goes without saying, your gcode looks like it should? There are no plunge moves in there?
When you say that a “rezero of Z works”, do you mean that if you stop the job when the problem arises and return to Z0, the mill is where is was in Z at the start of the job?