Originally Posted by
Hyperveloce
I see. But we have:
1) an intrument or a device (can be a Pitot-Static system, a plane)
2) a signal/indicator monitoring the performance of 1)
3) relevant procedures when 2) is activated
4) crew training/reactions to 2), given 3) and their observation of 1)
and it seems that all the 4 contribute to the flight safety ? If the regulations only specify the first 2, are they sufficient to ensure a probability on the safe continuation of flight ? (How is the "acceptable relation" between 1)-2) & 3)-4) & the overall flight safety impact defined or evaluated ?). Are there specifications about false alarms rates in normal/degraded modes ?
Jeff
The regulations go a bit further than just requiring failure rates to be commensurate with the hazard.
The failure rate of systems (including their warning systems0 is as noted previously, covered by 25.1309, and specifically 25.1309(b)(1)and(2).
The provision of appropriate procedures is required per 25.1585(a)(2)and (3) for abnormal and emergency cases.
That the procedures provide "continued safe flight and landing" for the various failure cases is determined in accordance with various handling requirements (for example) with various precedents and guidance material concerning, say, degraded handling acceptable for such cases.
I suspect that, certainly at initial cert, the basic probability of total loss of the air data system was already so low that compliance was achieved without worrying about the procedures, and any procedure provided was above and beyond the basic cert requirements.
To answer the question about false alarms, a false warning is in itself a system failure and would be assessed like any other for its consequences versus the probability.