Abnormal behavior during job

CNC Machines Shapeoko 5 Pro
1

I’m still learning this machine, so not sure what is the source of this behavior.

I have a pretty simple design, using foam. I have a relief (centered) with the intent to then do a cutout.

The relief went well, and the cutout was running fine, then ‘out of the blue’ it offset, did a cut, offset did a cut, until I hit the kill as I wasn’t sure where it would end up.

red lines in photo are the “unintended” cuts…
foam_test.c2d (900 KB)

PXL_20260829_193940796
PXL_20260829_1939407961920×2805 1.09 MB

Carbide Create Build 853 (OSX/Apple Silicon), Carbide Motion 664 (Linux/ARM)

2

Usually, if a design does not cut to match the 3D preview:

image
image1920×1152 63.6 KB

it’s because of some mechanical aspect:

  • cord or hose interfered
  • linear rails/blocks need maintenance/lubrication
  • machine hit a clamp

Video on this:

If the material was something other than foam, I’d go on my spiel about cutting tool engagement and using the new Cutout toolpath rather than a contour, but I doubt that’s the case here.

For some information on cutting foam, see:

1 Like
3

Excellent video… I haven’t seen it before and. I was surpirised at how many of Winstons points and suggestions I already do like checking the endmill length against the material. Actually did the exact same check today on a #102 endmill that I was using to cut through 21mm walnut.

4

Other suggestions that I should check?

No cords or hoses connected that could interfere
no clamps, it was foam and stuck down with double sided tape

The maintenance is a possibility but I bought the unit in June and I haven’t done a lot with it as of yet so would be surprised if that were the case…

5

The machines are pretty basic when one considers each element in sequence:

  • software writes out G-code for a command to move
  • Grbl interprets the G-code converting it into electrical signals to the motors
  • wiring conducts the signal (check the connectors and wiring)
  • motors rotate based on the current received
  • motor shafts turn couplers which turn ball-screws (check the couplers)
  • ball-screws turning cause the linear blocks to move (check lubrication)
  • machine moves unless something prevents the motion (physical object, feeds and speeds not right for tooling/material)

If you still can’t find anything obvious, check w/ the folks at support@carbide3d.com

6
  • …and the kerf comes out here :smiley:

BTW the “up picture” moves (green here) look consistent, the “down picture” moves (blue) seem to be where the problem is…

image
image540×581 55.3 KB

1 Like
7

You’re says that this suggests the the “down” (X right) is what’s being interfered with or failing, correct?

8

Yes, exactly. Here is a little better picture about what I mean (I don’t have CC so I can’t open your workfile to get which axis is which)

image
image1920×2805 1.09 MB

1 Like
9

The C2D file has toolpath specified for the pocket, how did you define the perimeter cutout?

I wonder if the cutting bit was pulled away from its initial height position (after binding with material)

There is a significant difference in material removed left vs right, particularly where the lowest red horizontal line is

10

I used the ‘cutout’ tool path (vs a contour (CCPro 853))

It’s a pretty light density foam, and didn’t see any binding. I have another piece, and am going to run it again…