Gage R&R MSA - Variation, Tolerance, Variance by different methods

[bobdoering]

Uncertainty is great for "protection" of the measurement when used for accept/reject. It does not give as good of an understanding of the statistical resolution the gage truly provides the way ndc does. It is important to calculate ndc for variation to assure you have enough resolution in the dimensional zone you are charting (which is a function of variation). For charting, the ndc truly should be >10 to avoid "chunky" data.

 

[ncwalker]

Agreed.

And again the point, it's measurement systems ANALYSIS, not measurement systems ANSWER.

All these numbers describe a system that takes more than one number to describe.

 

[bobdoering]

Re: GRR MSA: Variation, Tolerance, Variance by different methods

Thanks Miner,

And you are right, that is my situation,

The reason is that I select my Parts to spread across the measuring range of the gage, for example a caliper 0-150mm

Is this a process where you have historical capability data? If so, you will get a more accurate ndc and GRR %Variation (in my opinion) by utilizing that variance than the variance presented by the 10 parts used in a study. See AIAG MSA 4th edition page 121. They are not real clear on what they are trying to say, but conceptually it makes a lot more sense to me.

 

[Miner]

Re: GRR MSA: Variation, Tolerance, Variance by different methods

Is this a process where you have historical capability data? If so, you will get a more accurate ndc and GRR %Variation (in my opinion) by utilizing that variance than the variance presented by the 10 parts used in a study. See AIAG MSA 4th edition page 121. They are not real clear on what they are trying to say, but conceptually it makes a lot more sense to me.

Agreed. % Process Variation (PV) is always better than % Study Variation (SV).