2 ms·
See the Ford/Cessna VR story here: https://news.ycombinator.com/item?id=37776842 https://news.ycombinator.com/item?id=37776842 That’s paperwork and an ink stam
by sokoloff 3y ago
See the Ford/Cessna VR story here: https://news.ycombinator.com/item?id=37776842 https://news.ycombinator.com/item?id=37776842
That’s paperwork and an ink stamp as the only difference.
There are other cases now where NDT is used on high-grade bolts and Heim joints. Parts coming off the same line are sold with a -M suffix (“this bolt was individually magnaflux tested and can go on an aircraft”) or not (“this part can go on a tractor”).
Where no failures are ever found in NDT, I think most practical engineers would agree that the parts are physically identical.
- littlestymaar 3y ago> Parts coming off the same line are sold with a -M (“this bolt was individually magnaflux tested and can go on an aircraft”) or not (“this part can go on a tractor”). I think most practical engineers would agree that the parts are physically identical. Wait, of course they are identical… unless there is a defect! But that's the point of testing: when you can tolerate 1% defect then you can just sample your production line, but if your “safety demonstration”[1] relies on the fact that no screw is going to fail catastrophically under load because there's a defect, then you have to check them all. It's not “That’s paperwork and an ink stamp as the only difference”, exhaustive testing vs statistical testing. The work needed for the testing process is actual work, it's not about filling the paperwork. [1]: not sure this terminology exists in English, I'm just doing a literal translation of a nuclear Engineering concept I know in French only.
- peteradio 3y agoThat story said it was easy enough for that particular part to have 1 line rather than 2, its not generally applicable to all parts.