[albrewer]

> expensive metallurgical analysis involving destructive testing, spectrometers, and electron microscopes

I used to work in a pressure vessel fabrication shop (for customers like Shell and Exxon). We had a few handheld mass spectrometers for exactly this purpose. Destructive testing was achieved with what we called a "coupon", a piece of metal that ostensibly went through every treatment the base part did. The coupon was destructively tested, then etched and examined with a metallurgical microscope. This level of inspection is achieved by every ASME BPVC VIII compliant fab shop in the US and Canada; many of which are very, very small.

Boeing is outright negligent here if they didn't qualify their parts.

[dr_orpheus]

The article mentions that the CoC may have been falsified, but I also wonder if part of this is they had falsified coupon testing/inspection documentation (or likely pulled a "good" coupon test and said it was for that batch). They definitely did not test any coupons after receipt though since the testing by Spirit after the fact confirmed that "the material passed some of the materials testing performed on it but failed others"

I cannot imagine (I say hopefully) that there is not some level of testing here, but I wonder if they were relying on supplier testing and the authenticity of that. But in that case I would also assume that there would be some source inspection of the supplier. These might all be bad assumptions, unfortunately, but this is coming from my experience working in aerospace on the space side of things.

[hinkley]

CoC -> Chain of Custody for those out of the loop.

That’s how you make sure Honeywell actually made this particular part, that your QA signed off on it, and that this particular one was used for stress tests and thus must never, ever end up in the spare parts bin.

[albrewer]

> they had falsified coupon testing/inspection documentation

This is accomplished by specifying a separate testing house that you trust for this, if you don't have your own equipment. Many manufacturers don't have a tensile test specimen puller, Charpy impact test machine, fatigue test machine, mass spectrometer, x-ray machine, ultrasound technician, or metallurgical lab technician on staff to verify all this. But what you don't do is blindly trust documentation supplied by the vendor.

Not to say you verify every little aspect of everything documented- at some point it's not economically viable. But everything I've mentioned above is pretty reasonable to do, especially as reliability in the end application becomes more critical.