Does a CMM (Coordinate Measuring Machine) require Gage R & R study?

[SGquality]

Dear quality experts,

I am a newbie here - have seen the huge number of threads with enormous body of knowledge. Thanks all.

I have one question and would appreciate your views - we have a CMM for measuring part dimensions. Does this need Gage R & R study ?

Also a general question, which measuring equipment require the Gage R & R study and which type of measuring equipment do not.

Thanks in advance

 

[NumberCruncher]

Re: Does CMM require Gage R & R study ?

Hi SGquality

The answer to your first question is "Yes". If it is a fundamental part of your production process, it will probably need an MSA study.

The answer to your second question is contained in the answer to your first question.

Simply ask yourself, "Is this measurement a fundamental part of the production process?"
If the answer is yes, you will probably need to do an MSA study.

(I say 'probably' simply because I know the people who participate in this forum will almost certainly manage to find an exception!)

A few simple examples.

A CMM used to measure parts. Presumably you are not just measuring these parts for fun. If you are using the CMM for sorting good from bad, monitoring your production, as part of an approvals process, anything where you could reasonably state that "without this measurement, we would find it difficult / impossible to make / check / quality assure our products."

Balances, vernier calipers, micrometers, plug gauges (ie go-no go gauges), eyepiece graticules on microscopes, thermocouples, hardness testers, tensile testing machines, if they are used as part of your production process or QA system, will need MSA studies at some point.

The thermostat in the air conditioning unit in the offices, whilst very nice, even useful, is probably not essential to your production process and will not require an MSA study. Nor will the small plastic ruler on your desk that you use for swatting flies and making a "Boing!" noise with.

NC

 

[Jim Wynne]

Re: Does CMM require Gage R & R study ?

Dear quality experts,

I am a newbie here - have seen the huge number of threads with enormous body of knowledge. Thanks all.

I have one question and would appreciate your views - we have a CMM for measuring part dimensions. Does this need Gage R & R study ?

Also a general question, which measuring equipment require the Gage R & R study and which type of measuring equipment do not.

Thanks in advance

I don't know what customer requirements you might have to satisfy in this regard, but from a practical standpoint you should be interested in understanding how measurement error affects your processes. MSA (GR&R in this case) isn't done on devices alone; it's done on measurement sytems including devices, operators and parts. The goal should be to verify that a given measurement system is appropriate for measurement of specific part features.

 

[falconer65]

Re: Does CMM require Gage R & R study ?

The complex answer is yes and no. The yes is, as Jim pointed out, a GR&R is a measure of the interaction of the part, the measuring instrument, and the person making the measurement. So the no is from a CMM requiring a GR&R study, because a CMM would be required to be calibrated.

To the second part of your question, yes it would be nice to have GR&R studies of everything. With a CMM a GR&R is more a measure of the programmer, and an MSA Type 1 study can be more appropriate. The reality I often face is limited resources and only what the customer asks for gets MSA, of which a GR&R is a subset.

 

[Jim Wynne]

Re: Does CMM require Gage R & R study ?

The complex answer is yes and no. The yes is, as Jim pointed out, a GR&R is a measure of the interaction of the part, the measuring instrument, and the person making the measurement. So the no is from a CMM requiring a GR&R study, because a CMM would be required to be calibrated.

To the second part of your question, yes it would be nice to have GR&R studies of everything. With a CMM a GR&R is more a measure of the programmer, and an MSA Type 1 study can be more appropriate. The reality I often face is limited resources and only what the customer asks for gets MSA, of which a GR&R is a subset.

Setup (fixturing) of parts is also important in most cases, which is where potential operator error comes in.

 

[SGquality]

Re: Does CMM require Gage R & R study?

Thanks for the responses.

I forgot to mention earlier that we do annual calibration by Carl Zeiss and before we started using the CMM, we had successfully completed Measurement Uncertainty and Correlation with Customer's CMM.

Similar the the earlier responses on Yes and No for doing the GRR for CMM, would my additional information be able to decide if Gage R & R is required ?

 

[Jim Wynne]

Re: Does CMM require Gage R & R study?

Thanks for the responses.

I forgot to mention earlier that we do annual calibration by Carl Zeiss and before we started using the CMM, we had successfully completed Measurement Uncertainty and Correlation with Customer's CMM.

Similar the the earlier responses on Yes and No for doing the GRR for CMM, would my additional information be able to decide if Gage R & R is required ?

The question, it appears, is not whether GR&R (or some form of MSA) is required. You decide what's required and what's not, in the absence of customer or regulatory obligations.

What you need to decide is whether MSA would be helpful. If you think that measurement error (regardless of the calibration status) is a significant risk, you should do your best to understand that source of variation. If, on the other hand, you determine that the risk of measurement variation is negligible, and that MSA won't add any value, you should probably not do it. The decision is yours.

 

[falconer65]

Re: Does CMM require Gage R & R study ?

Setup (fixturing) of parts is also important in most cases, which is where potential operator error comes in.

True, even the best program cannot make up for a bad setup.

But a decent setup with a good program will nullify operator error with iterative alignments. By "finding" the part, then "checking" to see if the part is where the cmm "thinks" it is, operator error in placing the part on the fixture is reduced to a fraction of a percent of the total variance.

 

[Mary - QA Manager]

Does anyone have an excel spreadsheet that might help me with my CMM gage R&R?
The customer has very specify requirements.
One part inspected 10 times by one worker. I have up to 15 dimensions to check.
Thanks!
Mary

 

[Ripitup65]

If you have a CMM program written, which includes a part holding fixture the answer to "Does a CMM require a GR&R" the answer is obviously "NO."
The coordinate system written into the CMM program will always establish where the part is prior to taking dimensional readings. This would eliminate the Appraiser by Part error and the Repeatability error portions of a GR&R. This leaves you with Equipment error. This can be answered by the CMM's calibration record since it will give you all axis of error, linear bias, and accuracy from a known standard.
If you are a CMM person and you do a lot of of free state dimension checks this is different. But in this case it would mostly likely be a one part check.
Hope this helps.
Ripitup65