Typical Probe qualification Stdev tolerance

[[Jo...]]

Hi everyone,

Looking to understand what the typical probe qualification tolerances are. When I came to my new company, I noticed they are using a 0.001mm tolerance for the probe stdev. Is this a typical tolerance? In the past, I have used .01mm.

I suppose it could depend on your typical part tolerance.

 

Equipment: Zeiss Prismo Navigator, Vast Head

 

TIA!

[[Da...]]

The standard deviation is a statistical value for the amount of variability in a data set, such as a set of points. It does not show the range of a point set, which could be higher due to one or more outliers. It is an average value of the deviation of the point set to the mean value. When calibrating, this value should be well lower than a micron or 0.001 mm. A value of 0.01 is almost catastrophically bad and points to a problem with the probe system. 

[[Jo...]]

Hi

Please sign in to view this username.

,

Thanks for the feedback. From my understanding, this could depend on equipment and environmental capabilities. If you are not in a stable environment and not equipped with a machine accurate to that level, you could struggle. 

In my past life, we used .01mm but this was a Sheffield retrofitted with a PH10 and Calypso.  

What does your company use for probe Stdev tolerance, and what equipment are you using? 

[[Mi...]]

It depends....

Contura CMMs with XT Gold = 0.0003mm Standard Probes 0.0005 for our very long probes.

GageMax CMMs with XT Gold = 0.0005mm Standard Probes 0.0008 for very long probes

Our Crappy GageMax that only measures loose tolerance parts = 0.001mm for all probes

DuraMax/Eclipse with XXT = 0.001mm for all probes

Eclipse with RDS and TP20 = 0.003mm

 

 

[[Wo...]]

I would agree with Michael's values - that's what I'm seeing. The only exception is RDS CAA qualification in 12 positions - that tends to be higher around 0.001mm

[[To...]]

Please sign in to view this quote.

I often see as much .0017mm with CAA but I don't get overly concerned because, like you said, it's from 12 positions.

[[Mi...]]

Yeah my RDS recommendation are each position done individually with that Retrofitted Renishaw sensor... Not doing the CAA routine which I only recently learned was even possible. The current calibration method was already in place when I started and I've spent the last 5 years avoiding those machines as much as possible.... lol

[[Da...]]

Well, .001 is a bit less than .00004", so if you are expecting to measure to the nearest .0001", then yes, you need it. We measure more sheetmetal assemblies and levels where ±.010" tolerances are common, so my "quick and dirty" weekly calibration I do now to .005mm.

But my "super calibration" does the entire profile measurement on each stylus position, and on that I calibrate to .001, and I use this calibration on the rare occasions I need to check a precision CNC part (which is maybe 5 or 6 times a year).