Quote:
In Reply to Parent Post by Mark Paul
Many qquestions:
Several people here have different thoughts of whether testing with a profilometer should give Cpm as a result in Minitab.
|
Cpm, the machine capability index, is a dubious statistic to begin with, but a machine changes things that are put into it, and a profilometer measures them. Two different things.
Quote:
In Reply to Parent Post by Mark Paul
We wonder if the 32 pc. SPC should be bounded by 1.6µM max with zero as the LSL?
|
I don't know what you mean by "32 piece SPC" or what you're trying to accomplish. SPC, in general, involves sampling and measurement of characteristics during production.
Quote:
In Reply to Parent Post by Mark Paul
What then is the tolerance that is put in the MSA? We are using 10 customer pieces x 3 x3. Someone said we need to use just 5 standards on the MSA but that doesn't use the customers parts on the MSA.
|
Imagine that someone comes to you and says, "I need for you to screw a piece of metal to a table top," and you, thinking it a simple task, agree to do it. Then you find out that the "piece of metal" is a blob of mercury. In trying to do MSA for a profilometer on customer parts, that's pretty much what you're trying to do. It's also probably the reason that someone suggested doing something with standards rather than parts, because that's the only way you can hope for any kind of repeatability.
If you want to give it a go, use the total tolerance (1.6µM)--for GR&R it's the total amount of tolerance available, not how the tolerance might be divided.
Profilometer GR&R and MSA - Cpm as a result in Minitab?
Hybrid Mode
