hexagon logo

Disk probe XY Correlation

Hey Guys,

Kind of a weird one that I haven't ran into before. I had a disk probe made by one of the guys on the shop floor to check a seal groove. To make sure it was reading appropriately, I had it check the datum bore that I align off of with the main probe (2x40mm). The diameter matches within .0001" but the XY location is off varying amounts. It calibrated fine, I calibrated the two probes at the same time just in case, I checked the probe shank with an indicator and it's in the head pretty straight (~.0015" drift across 50mm), but its location is jacked up by a few thousandths depending on me reloading it.

We tried adding a bit more of a radius/relief onto the disc but it didn't seem to make a difference from it being a straight diameter. Diameter still measures good, XY still off.

Also tried rebuilding the probe file, didn't seem to do anything.

A few questions:

First, the obvious: what the heck?

Second: Even if the probe was slightly crooked in its XY, would calibrating it not tell PCDmis that its XY doesn't exactly match the nominal and it would then compensate for that deviation?

Third: Have you guys had this happen with collet holders before? This is my first time really using one to hold a probe since we didn't have an M3 tap/die on hand. I've ordered one so I can use a normal probe holder but I don't see why I should have to do that. I did notice that one time it got pulled from the probe changer it was crooked. I don't know how. I unloaded/reloaded it like eight times in a row after that and it happened twice. This could be happening on an imperceptible scale but I tried checking it once when it measured poorly and I couldn't measure any level of crooked in it.

Edit: For what it's worth, this machine is a Global S 9.15.8
Parents
  • Yep. That was the solution. I had accidentally set a janky probe as the master. 95% of my programs use one probe and the ones that don't usually have wider tolerances (+/- .010+') so I guess the thousandth or two deviation was never caught.

    Remember to pay attention to your master probes people. Deviation between probes is less than .0001" now.
Reply
  • Yep. That was the solution. I had accidentally set a janky probe as the master. 95% of my programs use one probe and the ones that don't usually have wider tolerances (+/- .010+') so I guess the thousandth or two deviation was never caught.

    Remember to pay attention to your master probes people. Deviation between probes is less than .0001" now.
Children
No Data