[Kosh]

I'm wondering if I have a problem with my z-axis.

1) When I use the CNCRP touch plate to zero a bit, I notice that when I say "goto zero", it is still visibly higher than the material which it should pretty much be touching.
( as an aside, I am still getting an inconsistent back off distance when zeroing. Somethings it retracts to 38.5 mm, other times 63.5. Can't see a pattern.

2) This week I surfaced a fretboard with it held in a vacuum jig. After I epoxied in the inlays, I put it back in the jig and told it to surface again. This should have got me inlays (and epoxy) totally flush with the rest of the fretboard, but I found they were still being left high. I then modified the program for negative "stock to leave" in 2 thousands of an inch increments. I had to all the way to -.006" before I actually removed all the high points.

Also in prep of 2, I also tried setting z-zero by just dropping the bit onto the surface and then tightening, but still no good. I suppose 2 could be due the FB getting sucked down a little lower in the vacuum jig this time (although it seemed to bottom out), but what about 1?

I am calibrated using the settings file that came from CNCRP.

Thoughts?

[lloydsp]

Shrug... 'sounds to me like your Z is losing steps. A simple test would be reduce the Z-axis speed by at least half, and see what results you get then.

If it becomes 'accurate' then, you know it's missing steps. If not, you have some other problem.

Lloyd

[ger21]

1) When I use the CNCRP touch plate to zero a bit, I notice that when I say "goto zero", it is still visibly higher than the material which it should pretty much be touching.
( as an aside, I am still getting an inconsistent back off distance when zeroing. Somethings it retracts to 38.5 mm, other times 63.5. Can't see a pattern.

If the actual position matches the DRO, then you may have an issue with your auto zero macro.
If the position does NOT match the DRO, then you are probably losing steps.

First thing you need to do is get your auto zero working 100% accurately, 100% of the time. If it doesn't, you'll always behaving issues.

[Kosh]

So, I took some measurements.

Ie, auto-zeroed, moved it up, then hit "goto zero".

1) first measurement: Z axis 4 thou too high
2) second measure meant: Z axis 2 thou too high

DRO always read Zero when it should

Then I realized the only different was the location where I decided to set Z zero.

At this point, I think I'm going to chock it up to having a less than perfectly flat surface for the touch plate to sit on (in this case it's an MDF jig which has been sprayed in polyurethane. It seems like it wouldn't take much unevenness for set that puck off just enough on one side to affect the Z setting. Also, I know for a fact that the MDF never measures equal thickness.

[LeeWay]

Also measure the actual height the the probe trips. Mine is pretty slow. I think it is what was recommended. It trips at just under an inch though, so that may be contributing some error in your ready.
Never try to calibrate on anything but a good or freshly machined surface for the best results. MDF around here can swell overnight minding its own business with the lights out. Like it stayed out drinking all night or something.