3 ms·

I would suggest to the regulatory/FDR/CVR manufacturers a change to the black box spec, which causes the pinger to only begin pinging 7 days (or some threshold)

I would suggest to the regulatory/FDR/CVR manufacturers a change to the black box spec, which causes the pinger to only begin pinging 7 days (or some threshold) after the crash. The first few days of pinging are generally useless anyway, because I have not seen any search where a sea vessel was remotely prepared to search in the right area or with the right equipment. Incorporate this delay period (in the absence of anything else), and at least you get an extra week out of the search potential time window.

dredmorbius 13y ago
Your assumption is false. There have been a few notable cases of flight-over-water crashes in which the correct zone wasn't found quickly, but in the case of flight 137, ships were over the presumed crash zone within hours of the aircraft turning up missing. In other cases, particularly where the crash itself was visually observed (US Airways 1549, Ethiopian 961, and others) where a multi-day delay in starting pingers would have been counterproductive. It's also useful to realize that any change you make on flight recorders applies to all equipment, needs to be replace on existing aircraft, and concerns a system which has to work with very high reliability.
- tzs 13y ago
  In some cases a delay in starting the pinger would delay finding the recorders, but does that actually cause any noticeable harm? It is important to find the crash site quickly, in order to help any possible survivors, but the pingers do not help with that, do they? Aren't survivors found visually or by the emergency locator transmitter? It could delay the forensics investigation, but that usually takes years to reach a conclusion anyway and depends on inputs other than the recorders, so it is not clear that getting the recorders is even on the critical path for that investigation. On a different note, I wonder if it would make some sense to have echoers instead of pingers? That is, instead of automatically sending out a periodic pulse, they would listen for a pulse and then respond.
  - dredmorbius 13y ago
    At best you might want to delay the system by a few hours. Which in the extreme case doesn't buy you much time. You've also really complicated the system, as it now has to determine how it's crashed and where and when to start the pingers. All parts that might fail. Simpler to either resolve the time component by 1) installing a larger batter (or more efficient pinger), or 2) by creating a jettisonable independent locator beacon, or possible a set of scatterable data recorders. In the latter case (say: a set of microsd cards), the problem still becomes "how do you find them*. The size, color, and beacons on existing data recorders all assist in their recovery.

dredmorbius 13y ago

Your assumption is false. There have been a few notable cases of flight-over-water crashes in which the correct zone wasn't found quickly, but in the case of flight 137, ships were over the presumed crash zone within hours of the aircraft turning up missing. In other cases, particularly where the crash itself was visually observed (US Airways 1549, Ethiopian 961, and others) where a multi-day delay in starting pingers would have been counterproductive. It's also useful to realize that any change you make on flight recorders applies to all equipment, needs to be replace on existing aircraft, and concerns a system which has to work with very high reliability.

tzs 13y ago
In some cases a delay in starting the pinger would delay finding the recorders, but does that actually cause any noticeable harm? It is important to find the crash site quickly, in order to help any possible survivors, but the pingers do not help with that, do they? Aren't survivors found visually or by the emergency locator transmitter? It could delay the forensics investigation, but that usually takes years to reach a conclusion anyway and depends on inputs other than the recorders, so it is not clear that getting the recorders is even on the critical path for that investigation. On a different note, I wonder if it would make some sense to have echoers instead of pingers? That is, instead of automatically sending out a periodic pulse, they would listen for a pulse and then respond.
- dredmorbius 13y ago
  At best you might want to delay the system by a few hours. Which in the extreme case doesn't buy you much time. You've also really complicated the system, as it now has to determine how it's crashed and where and when to start the pingers. All parts that might fail. Simpler to either resolve the time component by 1) installing a larger batter (or more efficient pinger), or 2) by creating a jettisonable independent locator beacon, or possible a set of scatterable data recorders. In the latter case (say: a set of microsd cards), the problem still becomes "how do you find them*. The size, color, and beacons on existing data recorders all assist in their recovery.