Bias, Stability and Repeatability - What is an acceptable percent % ?

[Larry Wade]

Hi I am the quality engineer and calibration supervisor of our QS9000 registerd company. Up until now we had performed only Gage R&R and calibrations on our gages and equipment. Our auditor has since asked us to do some form of repeatability test before calibration. I have looked in the MSA book for QS9000 at the examples of Bias, Stability and Repeatability. I believe what he is looking for is actually more along the lines of Stability or Bias. My problem is what is concidered an acceptable percentage. I can not find anything in the MSA manual.

 

[TheOtherMe]

You have to relate your results to a percent of tolerance, as I understand it. That with gage R&R. The MSA manual talks about looking at it all as a 'system'. If I understand it correctly, they want to make sure you are considering all the system elements. You may get away with not doing linearity or bias calculations if you can argue the precision of the measurement device far exceeds the tolerance (extreme case --> using a cmm to measure a 0.9 +/-0.1 meter rod). But then, you always know the linearity and bias and stability and uncertainty of the cmm simply because the manufacturer provides them for their products due to their very nature.

 

[Marc]

Also see www.qs9000.com/ubb/Forum4/HTML/000005.html