What's new
What's new

Probe Calibration Sphere Size

allloutmx

Titanium
Joined
Mar 6, 2013
Location
Rochester, NY
Good Day All-

Is there any relevance to the size of a probe qualification sphere size versus accuracy?
Its been a while, but as an apprentice- I had been instructed to calibrate my micrometers to a gage block similar in size to the features being measured . Im not sure that the same logic applies to probe since we are telling it the dia it is suppose to be achieving but the question arises as we are probing a feature much larger in dia than the qual sphere itself.

thanks in advance.
:cheers:
 
The size of the qualification sphere is pretty irrelevant. The qualification sphere is telling the CMM where your current tip is in relation to the first calibrated tip, it also tells the CMM the size of the current stylus based on the known diameter of the qualification sphere. The CMM bases the readings of a part off of the scales on each axis, and any software compensation(temperature/material/etc).

Your micrometer on the other hand had a screw thread as well as a scale on the thimble and body. If the screw thread is messed up on one end but not the other it could read correctly at one end and not the other.
 
I'd say some but maybe small difference if the calib of normal size.
A larger sphere has less curvature so more closely looks like a straight wall to the ball on the probe.
Think touching a 1 inch vs a .020 sphere and induced error.
There is no need to have a near the same size at the features being measured as you do with gauge blocks and mics.
Totally different accuracy problem so do not link such thinking.
Sometimes "gold master" parts are used to check that the CMM is working correctly but not for probe calibration.

As above stated probe calibration tells the CMM where the probe ball is centered and comps out probe deflection or signal lag to come up with an accurate touch.
The working dia of your ruby tip is not the measured value of that ruby ball when you get really fussy.

As a check out I like to run the cailb 10 times and look at the the variation in numbers stored. Then I like to move the ball X/Y/Z and do it again.
"Trust everyone but always cut the deck".
One should know how much to trust your CMM. Unfortunately many will not like the answer and it is not what is published.

Never did a real study on this, a very reasonable question and looking for more input or real life experiences.
Bob
 
I'd say some but maybe small difference if the calib of normal size.
A larger sphere has less curvature so more closely looks like a straight wall to the ball on the probe.
Think touching a 1 inch vs a .020 sphere and induced error.
This is true and typically I have seen qualification spheres at Ø.750 or larger.

The error you see when the probe direction does not match the surface you are checking is called cosine error and while it cannot ever be fully removed, it can be minimized. On machines that run PC-DMIS the probe touches 3 points on the sphere before it begins its calibration. Those 3 points are enough to determine where the sphere is relative to the probe from where it started. There is a small correction and then it begins its qualification.
 
I think at the end of the day, we really just need to rely on the sphericity of the qualification sphere. When qualifying the probe, we are A) ensuring that the ruby (or silicon carbide, or diamond, or whatever) is round, but B) we are also trying to characterize the trigger characteristics of the touch trigger probe (or scanning head). We are accounting for non-linear deflection of the stylus build-up in various vectors. With a larger qualification sphere, we can take more points over a larger (physical) area, which allows the machine to better account for scale inaccuracies. But that's just my $.02 Cheers!
 








 
Back
Top