Toolpath moved without the board moving?

Frustrating evening, also pretty confusing…

Started a badge project for a local pd. Finally thought I had everything right and I’m looking at the toolpath being laid down and something doesn’t look right.

I let it finish and go back to the previous toolpath that had initially ran. Runs the same path in a different position. I have no idea why… Checked clamps checked if it had moved, nope if it had moved as much as the toolpath was off I’d definitely notice.

Nope not a blurry picture. This is the same two toolpaths ran over 3 times. All ran in a slightly different spots with out the board moving.

Any reason why this would happen? I’m guessing it’s the belts or motors skipping or something? I’ve only been at this a couple months so I’m not sure.


Most likely this is lost steps. Please check the machine mechanically:

and belt tension:

1 Like

Alrighty, I figured it was something like that. I’ll check it tomorrow.

Just makes me sad… Finally got everything glued up right. Nope!

Live and learn I guess.

Yea for the most part the vertical lines didn’t move. The horizontal ones did though

One more question, is this something that just happens? Or is it from putting too much stress on the machine?

The Shapeoko and this entire class of machines is the perfect example of the “Wonderful One Hoss Shay” problem — each part needs to be balanced to the other, and trued up with them, but if anything isn’t in synch, it becomes the weakest point, and where things go wrong.

The great thing about CNC is the machine will execute an entire program when everything is set up and configured correctly. The bad thing about CNC is that a project will only come out correctly if everything is setup up and configured correctly and nothing goes wrong, and the hobby systems for the most part don’t have any sensors to determine when things go wrong.


  • always use:
  • always verify / preview toolpaths
  • always test toolpaths in scrap or inexpensive test materials
  • before committing to an expensive or irreplaceable material test thoroughly, put in a fresh endmill, and where possible, break up the project into discrete operations each of which where possible allows a bit of leeway for correction at a later step — only the final finishing pass (again, w/ a nice sharp endmill) should be a point of no return

Thank you for the answer! Always quick and informative.

Got everything tightened up. Essentric we’re freely spinning on left side and slightly loose on right.
Changed paths to 60 deg v bit and boom turned out awesome.


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