Practical lesson in CMM Accuracy

[[Ow...]]

Measuring a 20 mm sphere (in pic) using ten 430° circles covering at least 75% of the sphere, 5 mmps speed, 8 mm ruby, Low-pass Spline with 2.5 wavelength filter and std outlier.
Probe S-Value is 0.0003mm
Anybody want to guess why it was measuring it 0.010mm undersized, with good form?

[[Jo...]]

Because that standoff is so long that it actually pushed off or because that little thumbscrew wasn't holding the foot down tightly enough?

[[To...]]

Please sign in to view this quote.

To add to John's answer, a loose connection between any of the components.

How about an erroneous radius value in the Reference Sphere properties?

[[To...]]

Worn probe?

[[Ri...]]

I'm going to go with the L shaped connection to the R&R sub-plate.

[[Cl...]]

I had a similar problem once. The 8mm stylus was too large for the sphere it was measuring.

[[To...]]

He said there there was good form, so I feel the sphere can't be moving.

[[Ow...]]

Andrew got it right first but, even though the thumb screw did have it pretty tight, you could still move it by hand with maybe 2 to 5lbs of pressure.

It just wasn't rigid enough and the probe using 200mN of force easily moved it around, while keeping good form.

Reduced the pressure and dynamics to 100 and 50% and it got about 5um bigger (still under), reduced it to 50 and 10 and it got 7um bigger but form wasn't as good.

I suspected as much when I put it on for a reference artifact and needed it high enough to clear a few other fixtures at the time.

Took the top level off and put it in a chuck and BINGO it was dead on 20mm with the standard 200mN of force.

Just thought I'd share for of an example how important proper holding/fixturing is.

[[To...]]

hmmm One would think if the sphere was moving in X and Y and probably not moving in Z, or at least not as much as it got higher in Z, that the form would be affected.

[[Ow...]]

I agree Tom.
I said it was good, within what I thought it might be at 0.003-0.004mm and got as bad as 0.008 with the 50/10 settings but, not close to the 0.001 I got when it was secured. I would have thought it would get better with the lighter settings but, apparently size was ill-compensated with the pressure and maybe being slower on the reduced dynamics had a weird effect too.?
So, moving around did affect it, I just thought it would be even worse.

[[Jo...]]

Thanks for posting this. Anyone else have any "lessons learned" to share?

One big thing I've learned is that even though the ISO 5459 switch is turned on, it only assures mutually perpendicular datums when they are individually selected in the characteristic. It doesn't affect the Base Alignment or any secondary alignments. I know you can force this but before i knew about it, I thought it was automatically happening with any DRF.

[[Da...]]

Mine's pretty solid 🤣 🤣

Ref sphere.jpg