Re: How do I determine a correct Sample frequency for SPC features?
The problem is if you are precision machining, there is no sampling rational that can make the X-bar/R work. I have case studies that prove that not only is X bar-R OR I-MR wrong, they give the exact opposite signals to the true condition of the process - i.e. if the process is out of control on X bar/R it is actually in control and vice versa. Those are the absolute worst charts you could ever use for precision machining - bar none. You are better off using paper charts than confusing your operators with an automated system that is wrong. In fact, I did that, replaced a computerized SPC system with paper charts - in a union plant - and they actually appreciated it because finally the charting made sense.
You may want to approach QC-Calc and ask them is they have correct charting (X hi/lo-R) for machining, since they claim to offer SPC software. You get more traction as a customer than other people might. It's sad if they don't know about it - the technique has been published for over 10 years and implemented much longer than that. And....it is correct!
So where do you stand on Bob Doering's approach. Although I understand his approach, I'm handcuffed with using QC-Calc (Xbar / R charts only). Please expound on how you would approach should be used for sub-grouping if your only option was Xbar and R charting (what would be a rational sub-group for precision machining?)
You may want to approach QC-Calc and ask them is they have correct charting (X hi/lo-R) for machining, since they claim to offer SPC software. You get more traction as a customer than other people might. It's sad if they don't know about it - the technique has been published for over 10 years and implemented much longer than that. And....it is correct!

