Anyone have any papers on ideal probe qualification frequency?

[[Ni...]]

I have asked a few pro's in the CMM world about how often is ideal for requalifying a stylus and it always "depends" but it's often suggested to perform once per 8 hour shift or once a day, etc. How do we know that this is ideal? Has anyone come across any papers that show a multi-factorial study on how measurements shift during different time lengths between probe qualifications?

I rather not do my own DOE if this information already exists somewhere.

Thanks!

[[Ga...]]

I'm not sure how long you can technically go without calibrating. However, if you measure a part and the results are bad, the first question the machinist will ask you is whether or not you calibrated beforehand. It saves time and having to go back and measure the part to just calibrate first thing in the morning and be able to say "Yes, the probes are calibrated. Your part is ACTUALLY bad, believe it or not".

[[Ni...]]

Thanks Gabe. Yeah I'm trying to proceduralize or set up preventative maintenance work orders for it, but I don't want to overkill it if I don't need to and I want to make sure that the operators who have to do the requal aren't overburdened. Figured hard data would shut down any naysayers.

[[Ca...]]

You qualify probes every day? I qualify them maybe once a week, depending on use. I use a probe qualification program, a full geometry qualification run takes about an hour and forty minutes. Most probes have 20-40 rotations to qualify.

[[Ni...]]

Hey Casey, my question is really about what you just said: "I qualify them maybe once a week, depending on use".

How do you know to do it more or less frequently?

[[De...]]

I have no studies or anything, however some things to consider include:

Temperature variation over time.
Type of Stylus systems / materials used.
Inspection requirements - accuracy required.
Wear on stylus tips.

The reasoning on these is pretty simple. Depending on what is used for your stylus system extensions, the temperature can cause thermal growth / shrinkage as temperature in the lab changes. The changes in length between stylus systems of different lengths will cause additional error on programs that utilize more than one stylus system.

This effect will be lessened if you are using extensions made from more thermally stable materials.

Depending on the type of material being measured and the stylus tip material wear can and will cause flat/worn spots over time.

Simply calibrating the probes will not do you any good if you do not put into place some method for checking the quality of the probes you are using. Limits on sigma on the calibration is helpful but can still miss problems with stylus form error.
Implementing methods such as scans of the reference sphere or roundness checks for checking the form of the stylus tips can help to identify problems with individual styli.

Finally the accuracy required based on your needs comes into play because if you are checking wide open tolerances none of this will matter much, however if you are chasing microns all of this will become far more critical.

[[Ca...]]

Derek makes some really good points. As for me, the lab si climate controlled pretty tightly, we have time/temp/humidity data loggers in all our labs as part of our quality control. It mostly stays within 2°F in here.

To Nima's question of how do I "know" when to qualify, it's tough to answer. I try to keep the probes clean, and only check clean parts. I'll let probes go a bit longer between qualifications if I know I've only been running parts made of stainless, or nickel, as the material won't really cause any build up on the probe. If I'm checking aluminum, I'll clean and qualify probes much more often. Obviously if there's any considerably collision, I qualify everything. Every now and then I'll put a setting ring on the plate, and make a quick program that checks the ring with a few different probes, just to be sure they're running true to each other. If the ring measures more than a couple tenths difference between probes, I'll qualify them.

[[Ow...]]

As everybody has mentioned, there are a lot of variables when it comes to setting up a probe calibration routine.

The UK's National Measurement Institute, NPL offers a lot of best practice guides and the best thing about them is that they are FREE.

https://www.npl.co.uk/resources/gpgs/all-gpgs

I'll attached 3 on CMM's that goes farther than, "it always depends" but, nothing conclusive for everybody or every application

NPL Good Practive Guide-43-CMM Probing.pdfNPL Good Practice Guide -42- CMM verification.pdfNPL Good Practice Guide -41- CMM stategies.pdf

[[Ri...]]

I think the biggest thing for me that I haven't seen mentioned is how often does the machine encounter a collision?