Is repeatability required for equipment calibration?

greif

Involved In Discussions
We calibrate a lot of our own equipment. My question is, that in addition to measuring a calibrated, traceable, master scale at 20 points per axis, do we also (for each calibration ) need to make some sort of check of measurement repeatability (such as measure the same distance 20 times)?

Thanks!
 

Jim Wynne

Leader
Admin
Repeatability is a component of Measurement System Analysis, and GR&R in particular, not calibration.
 

greif

Involved In Discussions
My main concern is; can I be sure the machine is measuring correctly, unless repeatability is checked periodically?
 

Jim Wynne

Leader
Admin
My main concern is; can I be sure the machine is measuring correctly, unless repeatability is checked periodically?
I don't know what you mean by "repeatability" in this context. It has a particular meaning in MSA. You assure that the machine is measuring correctly by way of periodic calibration.
 

Ninja

Looking for Reality
Trusted Information Resource
Repeatability of the gage itself is most times assumed...the thing works or it doesn't. That being if you use a gage to measure a thing more than once, you get a "repeatable enough to not care about" result.... the 'caring about' being established by GRR...

What specific type of gage raises your concern?
 

greif

Involved In Discussions
I am thinking of a 2 axis CMM (x-y) in this case. Calibration is to insure a machine is measuring correctly. Something loose on the machine would make poor repeatability which would affect the accuracy of some readings.
 

Jim Wynne

Leader
Admin
I am thinking of a 2 axis CMM (x-y) in this case. Calibration is to insure a machine is measuring correctly. Something loose on the machine would make poor repeatability which would affect the accuracy of some readings.
Now you're in another different subject--device maintenance. If something loose on the machine might affect measurements, you should make sure that nothing is loose before measuring.
 

greif

Involved In Discussions
That is the point of the repeatability check- it would point out that something is loose.
 

Jim Wynne

Leader
Admin
That is the point of the repeatability check- it would point out that something is loose.
  1. What you're referring to as repeatability is not repeatability as the term is used in metrology. Avoid confusion by not using the term in an eccentric way.
  2. If you're afraid that your CMM is defective in some way, you should have that attended to.
  3. Do MSA as necessary to find out if there are issues with repeatability (and reproducibility).
 

Ninja

Looking for Reality
Trusted Information Resource
When something goes wonky with a CMM, it's usually fairly obvious in my experience... loss of focus (Z-axis motor), clicks, grinds, absurd data coming out...
If you're worried about it...simply keep a "known" at the machine and measure it when in doubt.

I had a MicroVu Vertex for 13years of daily use and the only time it needed repair for function was when a z-axis motor went out...it was pretty darned obvious with all the racket it made trying to focus.
Measuring 20x on the same part to verify seems like total overkill...just make a simple test regimen (2minutes or less) to verify function.

We had a program to measure ID and OD of O-rings...I just put my wedding ring on the stage and hit "go"...if there were any issue in X, Y or Z it would have shown up immediately. I would do that perhaps 2-3 times a year, typically while training someone new to the machine.
Other operators used it every day, multiple times a day...again, if something were wrong, it's pretty obvious on a CMM.
HTH
 
Top Bottom