Do YOU want to?
weisusu said:
I have several welding processes where I need to validate the testing method. Basically I have three welded parts where I use exactly the same tensile tester, same testing speed, same testing fixture, same program for testing. For testing validation standpoint, do I need to perform three MSAs or one is sufficient to justify since all testing parameters are identifical?
weisusu
If you are in Autotmotive, you probably have to, but...do YOU want to?
Do you see any benefit?
If you don't want to, just do one generic test and submit it clearly labelled as such.
You might get called on it right away (apologzie and do the other tests ASAP), or it might be accepted for years (smile inwardly).
I find the customer person reviewing the data is even busier (and possibly) less knowledgeable than me.
I have sent in MSAs with incorrect math, MSAs done on gage blocks, and family MSAs for years. As long as the R&R was under 10%, there was never a problem (all they understand?). It's my little customer test.
I did get caught with the gage block study by an astute customer SQE. He asked if I knew that it was incorrect to do studies on gage blocks. I said I did, but didn't know if he did. I said I'd fix it if he issued a formal Corrective Action so I could show my boss I needed time to fix it. It did not come up in his closing meeting.
So, there is "what is required", and "what you can get away with".
I suppose there is also "what are you comfortable with".
These are 3 very different things.
I hate MSA, but I always learn something from them. If you don't do it you'll never know. Does it matter? To who (whom?)?
Please let us know what you decide to do