Probe Calibration Accuracy

[[Ch...]]

I would like to know which probe calibration technique is the most accurate.

I have a Contura with a RDS CAA XXT head.
In previous divisions I have always calibrated the probes using the 17/12 step process that is built into the Calypso software.

In the division that I am currently at, each angle of the probe is a called a different tip. i.e. like a star probe. Thus if I want a different angle, I have to create a new tip, then calibrate it.

When I asked why it was done this way, I was told that it is more accurate.

My question is: Are they correct?

[[Je...]]

I believe you are referring to CAA vs discrete angles (non-CAA mode). Discrete angles should be at least as accurate but most often it will be more accurate as each angle is measured directly rather than mathematically calculated.

[[Ch...]]

Then theoretically you would need greater than10k discrete tips per probe to get every conceivable angle. Correct?

YIKES!! 😮 That would take all day to calibrate just one probe!

What is the accuracy when calibrating using the calibration routine built into Calypso? .001? .0001? .00001?

[[Da...]]

Charles, are you going to use all 10K angles??? I'm pretty sure you won't. When I had a rds caa I would calibrate each angle (that I used) and calibration was pretty much automatic after set up. I think we averaged maybe 5 or 6 positions per probe, it only took a few minutes. For the gain in accuracy, it was well worth it!!!!

[[Je...]]

Please sign in to view this quote.

The most I've ever used was 150 discrete angles on a single stylus. I had 4 styli that each had between 50-150 discrete angles qualified. A full geometry requalification routine for all 4 styli took just over an hour. This could be broken up into styli specific programs to spread it out over the day between part runs. Could also have the full routine set to run at the beginning of the day before most CNC machines complete their warm-up cycles. Keep in mind, this was also for an inspection program that took ~8 hours to run.

I would say that having more than 15-20 discrete angles per stylus is uncommon. Each discrete stylus angle should take about 30 seconds to qualify. Closer to 3 minutes each under a Passive Qualification but that routine should only be needed at initial creation or if the stylus (or sensor) is physically altered in any way.