Applicability of R and R study for gauges

[Andrews]

Intent of R and R study is to find out whether there is any variation in the method of checking of a particular parameter using a particular instument / gauge.

How many of you agree with this? If so why should we check R and R for gauges like plug gauges where there is not going to be any variation in the method of checking ?

Thanks,
Sonu

 

[LI Yan]

Pls. check in MSA 3rd edition manual. You can treat Pin gauge as attribute data. There are some method in this manual to stuty G R&R for attribute data gage and why to study.

 

[wmarhel]

If so why should we check R and R for gauges like plug gauges where there is not going to be any variation in the method of checking ?

There is always variation, the question is whether or not the amount present is acceptable. In the case of something like a plug gage, maybe the variation is due to the temperature of the gage when the measurements are is taken varied from one set to the next.

Another potential source of variation is that the operator just doesn't have a fine enough "touch" and so feels that the next size down is required. The next size down may only be 0.001", but that could still be different from what the other two operators are finding.

The purpose of a MSA is to verify that the equipment and the people are able to operate within acceptable limits of variation.

Wayne

 

[Caster]

It's those d*** people

Intent of R and R study is to find out whether there is any variation in the method of checking of a particular parameter using a particular instument / gauge.

How many of you agree with this? If so why should we check R and R for gauges like plug gauges where there is not going to be any variation in the method of checking ? Thanks,Sonu

Sonu

I have said the exact same thing myself whwn faced with this requirement, so I thought I'd just go ahead and prove how stupid this requirement was.

So far I have RARELY had an attribute gage pass a "good" R&R test. For example using real shop floor operators, hiding previous results, samples just at the edge of fail, etc.

Things I have seen fail...leak tester, thread gage, and amazingly a simple plug gage.

People always astound me with their ability to use even a simple gage incorrectly in the most amazing, creative and unexpected manner.

Why not try what used to be called the short method attribute, 10 parts, 3 people, 3 repeats. If they all agree waste no more of your time....and can we hire some of your people?