Obligatory Preface: I've never worked on defence projects but...
The issue of initial quality and provenance of the chips aside, I don't understand how this wasn't picked during Functional Performance Testing (FAT) of the system before final acceptance.
For such a critical safety system you would expect that all tests were performed and that it was operating correctly within it's design parameters. You would expect whoever was responsible for procuring (either principal or contractor) would have witness points were a suitably qualified person from their side (or third party independent verifier) would witness and sign-off on the test being performed to spec, and the results were within expectations. This is why these processes exist.
NB: The test in this case might not even need the actual parachute, just testing that the circuit pulses at the right time in the right place, or that a solenoid activates.
Then again, maybe it did pass all these checks but the poor quality of the parts meant they degraded at an accelerated rate because they weren't designed to operate in those conditions. Who knows. As things get smaller and more abstract (digital v. analogue) it's much more difficult for us to rely or even expect traditional methods of "trust but verify" to satisfy our requirements. It's becoming more and more incumbent on the end-owner to know their asset as it becomes harder and harder to understand it.
The issue of initial quality and provenance of the chips aside, I don't understand how this wasn't picked during Functional Performance Testing (FAT) of the system before final acceptance.
For such a critical safety system you would expect that all tests were performed and that it was operating correctly within it's design parameters. You would expect whoever was responsible for procuring (either principal or contractor) would have witness points were a suitably qualified person from their side (or third party independent verifier) would witness and sign-off on the test being performed to spec, and the results were within expectations. This is why these processes exist.
NB: The test in this case might not even need the actual parachute, just testing that the circuit pulses at the right time in the right place, or that a solenoid activates.
Then again, maybe it did pass all these checks but the poor quality of the parts meant they degraded at an accelerated rate because they weren't designed to operate in those conditions. Who knows. As things get smaller and more abstract (digital v. analogue) it's much more difficult for us to rely or even expect traditional methods of "trust but verify" to satisfy our requirements. It's becoming more and more incumbent on the end-owner to know their asset as it becomes harder and harder to understand it.