Cutting Deeper Than Commanded

How To Machining
1

Hi All,
I am cutting a board and the commanded depth is -0.1". It is actually cutting -0.2" deep, although the Z position in CM is -0.1" as shown while the job is running. I did zero the Z before running the job. What could be happening here?

Thanks,
Mark

2

Have you trammed/surfaced your MDF?

What material? Cutting using what tooling? At what feeds and speeds? With what sort of toolpath? How did you measure the stock? The cut feature?

Some possibilities:

  • tool pulled out of collet (tighten)
  • stock pulled up into cut (secure so as to prevent this or use a downcut tool)
  • machine pulled into cut (adjust cut geometry to reduce tooling engagement)
  • inaccurate setting of Z-axis zero
  • inaccurate stock measurement/variable thickness of stock
  • MDF not of a consistent height

Upload your file, step-by-step notes on how you secured your stock and set zero relative to it and managed all tool changes and a photo showing an attempt at cutting still in place on the machine with the machine at the zero position (or a specified offset from that)?

1 Like
3
  • Machine not calibrated. Move your Z axis a distance you can measure accurately, say with a 123 block, and verify it actually moves that distance.
    There are parameters in GRBL that set the steps/mm for the steppers you can use to fine tune the accuracy.
1 Like
4

Hi Will,
Thanks for the info on this. Tramming I did week before last, and you referred me to the procedure on that. It was slightly off fron to back, but it is right on now, as seen after flattening the waste board.

The carve is a simple pocket on insulation foam that I use for testing. Method for securing workpiece is screws into the wasteboard due to the size of the piece. There is almost no load at all on the 1/4" end mill. There is some inherent tension in insulation foam such that after you machine it some, it can move, sometimes a lot. This is not the case here as the piece was flat on the table. All of the other mentions, except the actual Z zero are ok, including stock thickness. Is there a possibility that somehow there was a Z offset being applied?

Thanks,
Mark

5

Thanks Tod,
So far all of the other carves I have been doing have been the exact dimensions I am calling out, in X,Y, and Z. I suspect the calibration is fine in this case. I am just wondering if somehow the Z had an offset being applied. Is there a way to initialize only one axis, say the Z axis?

Thanks,
Mark

6

I don’t know about initializing a single axis, but you can certainly zero one at a time.

How are you setting Z zero? Are you using the BitZero or touching the surface? And is that with a probing pin or a bit?

After you set Z zero and the machine touches the bitsetter, have you jogged back to the material to see that the positions look correct?

1 Like
7

Hi Cullen,
In this case, I am using touch method on the surface of the board and manually zeroing the Z, then showing 0.000. After that manual zero, the bit goes over to the bit zero and does that normal routine.

Another question that came to mind is how the machine keeps track of where it is at, without encoders. What if the axis loses some steps? I don’t think this new machine would have issues with that and past experience tells me that open loop motion systems can get screwy when they lose position.

Thanks,
Mark

8

Have you verified after the bitsetter measurement that Z zero is still correct?
I have had trouble with bits slipping in the collet. I suggest putting a sharpie line around the bit where it touches the collet after you tighten. Then you can easily see if there is a problem with that after the cut or by pausing during the cut.

These machines are opn eloop as you say, so they do not know if they have lost steps.

Is your material measurement accurate? All dumb questions I know, but it helps to narrow down to the root cause.

9

Thanks again for the reply. Material is measured with very good quality calipers, and the bit didn’t appear to slip. I will mark and check on a future carve to make sure. When I manually zero’d the Z, I used the paper method and Z read 0.000 after clicking the button. I should also then return to the same point after bit zero measurement, to make sure the Z still says zero when touching the material in the same spot again.

10

Right. That would allow you to verify that things are not getting messed up with bitsetter. One more thing to rule out.

11

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