Dear Dallas,
The repeatability test is a simple demonstration of the capability of a gage to perform a required task with the degree of "uncertainty" being less than the tolerance band allowed for the process. For example: If you have a peening callout of .010 - .014A intensity you would want to make sure that the gage (and the operator) were capable of reading a consistent number within the tolerance band. Generally 10% or less variation is acceptable, 10-30% is tolerable if you have efforts toimprove the situation, and above 30% is inpermissible, you must STOP. Thereason for this guidline is to prevent operator (or gage) errors fromcorrupting the gage measurement process.
We have a booklet on "How to conduct a gage study" which we will send toyou. It describes how you number 10 Almen strips and measure their archeights and record the information in a table. You repeat the measurementstwice. Ideally each time you read a certain strip you would get EXACTLY thesame number. This is the "Repeatability" of the "R and R" study.Then a second person, using the same gage and same ten strips measures thestrips, again three times. Next, a third person does the same. By comparingthe results of three people you now have the "Reproducibility" portion ofthe "R and R" study.
It would be a shame to have an operator actually doing a great job but not get credit for it due to bad gage "R and R" performance. Gage "R and R"should be done on a periodic (usually annual) basis.
Jack Champaigne