Faulty KP3 Probe ?? <Solved, Tighten stylus>

[tblough]

I would run your indicator test the same way the probing routines work: fast feed to the work, back off, and then slow speed to the final touch. I'm pretty sure this is a machine dynamics issue and not a probe issue.

[RJS100]

Thanks Tom. I have thinking about your comment on machine dynamics, and one thing comes to mind. When I measure tool length, if I remeasure the same tool repeatedly, I get the same results each time. If you have a few lines of g code you think would illuminate the problem, I would be happy to try it.

[suntravel]

Main difference using a dial indicator vs probe is that the CNC knows where to stop with the indicator, with a probe it stops if the input is set.

I would try another probe and/or another input.

Uwe

[RJS100]

I'm moving inputs around to accommodate my ATC project using the ether1616. I will swap the probe to another input and see what happens. Easy enough.

[suntravel]

Do not use the Ether1616 for a probe, it is too slow. Use the onboard inputs.

Uwe

[RJS100]

SOLVED! I swapped inputs for the probe tripped.... Same results... Made no difference.

Then I tightened the stylus a bit. Probed the top of my vice 10 times with a variance of just .0002. The probe spec that came with the serviced probe for Z repeatability was .00017. If my home-built converted machine is only off by .00003.... I am a happy camper!

Hell... I just finished redesigning and machining my x, y and z ball screw mounting brackets

[suntravel]

Big surprise, if done ok it works like it should

Uwe