Ahh... Yes, I see that now
On one hand I look at HP/MPI as a procedure that finds the parts that are flawed from the get go. But, that HP is the key. If one never suffered an overpressure condition in their rifle I expect that a number of that 3% would likely serve through their normal expected service life without failure.
Then, you have to figure that a certain number of that 3% would "fail" a MPI inspection at some point but not in a way that was user detectable or catastrophic, and may still serve out their expected service life or at least not be drastically reduced.
Then, some part of that 3% would fail early, detectably, even catastrophically. Those are the ones I don’t want

.
Yes, it would be interesting to know the details of the 3%. How many are catastrophic failures? How many are detectable after the HP without the MPI? Stuff like that. Even more interesting and I suppose more to the point of what TehLlama was getting at would be to take a significant number that "failed" HP/MPI testing, but only in a way that was detectable with MPI, stick them in guns and see when they failed in an obvious/detectable manner.
Bookmarks