FAA Mandates Major Aircraft "Black Box" Upgrade

coondoggie writes "Earlier this week the FAA mandated upgrades and updates to aircraft voice and data recorders within the US. The goal of the updates: to assist future investigations with 'more and better data' from accidents and incidents. The 'mandate means manufacturers such as Honeywell and L-3 Communications as well as operators of airplanes and helicopters with 10 or more seats, must employ voice recorders, also known as black boxes, that capture the last two hours of cockpit audio instead of the current 15 to 30 minutes. The new rules also require an independent backup power source for the voice recorders to allow continued recording for nine to 11 minutes if all aircraft power sources are lost or interrupted. Voice recorders also must use solid state technology instead of magnetic tape, which is vulnerable to damage and loss of reliability.'"

You'd think

[sleeponthemic]

That video surveillance would be part of the mandate.

Re:It sounds so easy but

[rabiddeity]

The difference between a $40 mp3 player and a flight recorder is that the flight recorder must be engineered to never fail, ever. If you plug the mp3 player into an outlet to recharge and a power surge hits, it will get fried. You expect that. You can buy another one. But the flight recorder has to withstand the aircraft getting struck by lightning repeatedly, and still continue to function.

In addition, every component must survive the severe stresses involved in a plane crash. The severe acceleration can cause large components to get ripped off their solder points. The device will likely be cooked to several hundred degrees as the plane burns around it, so all the components need to survive that (electrolytic capacitors will explode well before that). Heck, if the plane spontaneously breaks apart on a trans-Pacific flight, the box gets cooled to the outside air temperature of around -50 C before slamming into the ocean at high speed. Let's see your music player take that and survive. And I hope whatever software running the thing wrote the data out cleanly before everything went to hell, because if any of those stresses caused a hardware glitch that overwrites or erases the log, you get to tell the FAA that you really don't know why that plane crashed. Oops.

Re:Solid State is vulnerable to damage as well

[Khyber]

The point is that those vibrations you mention would destroy the solid state storage, thus rendering the data absolutely useless and null. True that tape drive motors would be severely affected unless the whole unit had a gyro stabilizer (which I think some models do) but solid state would shatter upon impact. You rarely find working electronic devices after a plane crash, except for military ones.

Re:It sounds so easy but

[mabhatter654]

like all good regulations though, they do many things very backwards. I've worked for a contractor too and many practices, while safe, are outright backwards given the leaps in technology. An iPhone and Wii controller are probably more advanced, and more reliable... not entirely fit for the job of a black box, but the direction it should be going... half the size and twice the function. The 50 year-old engineers that design this stuff are just plain out-of-touch with what technology can do now... flat out unable to understand it's application in many cases I've seen. Something like an iPod Touch has 16 Gigabytes of data... that's plenty of storage for what they need. As most instruments are digital (or should be) it should be easy to interface to the outside instruments rather than have so many enclosed as the quality of external instruments is much better now. So much has changed, an inline data trap with the fly-by-wire would be more in line, tried and true similar to any plain network logger... but aircraft people just don't think like that.

Question: why just record?

[martyb]

Here's a question that's been gnawing at me for a while... why is the "black box" just a recorder? I'd think of this question every time I heard that there's been an accident and the black box had not been found. OR, that they found the box but it was too badly damaged to make out all the data. Is this still a problem?

If a black box (BB) senses an anomalous event, why couldn't it transmit a [compressed] copy of the recorded data? Or, even better, besides recording it all, transmit all the data all the time. Maybe not to the airline, but to you at L3 Aviation Recorders, perhaps? With the recent talk about providing in-flight internet access, I could see this happening sooner or later.

Without internet access, just have a reserved frequency to transmit on. If transmit time becomes an issue, use multiple frequencies and transmit on each one of them in parallel.

I can't imagine I'm the first to think of this, so what am I missing here? Could it be it is only now that we could conceivably do this?

FAA Looking To Make Money From Fines

[AO]

I think the real reason for the new rules is to increase the money from fines.

From TFA
 

* By January 1, 2005, retrofit all airplanes that are required to carry a cockpit and data recorder with a system that is capable of recording the last two hours of audio; and is fitted with a 10 minute independent power source that is located with the device and that automatically engages and provides 10 minutes of operation whenever power to the recorder ceases, either by normal shutdown or by a loss of power to the bus.

* Require all aircraft manufactured after January 1, 2003, that are required to carry a cockpit and data recorder be equipped with two combination cockpit voice and data recording systems. One system should be located as close to the cockpit as practicable and the other as far aft as practicable. Both recording systems should be capable of recording all mandatory data parameters covering the previous 25 hours of operation and all cockpit audio and controller pilot datalink communications for the previous two hours of operation. The system located near the cockpit should be provided with an independent power source that engages automatically and provides 10 minutes of operation whenever normal aircraft power ceases. The aft system should be powered by the bus that provides the maximum reliability for operation without jeopardizing service to essential or emergency loads. The system near the cockpit should be powered by the bus that provides the second highest reliability for operation without jeopardizing service to essential or emergency loads.

As I recall, this is 2008, all year long.

Yeah, good luck with that.

[Mr. Roadkill]

Voice recorders must also use solid state technology instead of magnetic tape, which is vulnerable to damage and loss of reliability

Okay. Good luck with splicing together itty bitty fragments of flash memory chips. Good luck with pulling information out of flash memory chips that have been under a couple of miles of salt water, and had the briny deep seep in between the legs and the epoxy and into their inner goodness. I hope they've got all kinds of grinding machines designed to allow them to separate individual chips off busted boards and prepare them for reliable connection to special test jigs, because the chance of them being able to play back from a flight recorder that's just fallen from 40,000 feet must be pretty slim.

I'm not saying you couldn't build a solid-state flight recorder that could survive most conceivable crashes, but surely tape and solid-state should be viewed as complementary technologies - current, perhaps improved magnetic recorders for the current timeframes (so you've got at least the last half hour on something you can piece together and pull an analog signal off, if need be) and the whole flight on an ever-improving series of solid-state recorders that would have to consider mil-spec as a starting point for where they need to head.

Re:It sounds so easy but

[eonlabs]

To be fair, my gamecube is dead. In its wake, I've been playing my super and regular nintendo.

I beg to argue that older technologies have stood the test of time compared to our modern works which last two years or less.
I would be far more interested in a black box that works reliably, even with some moderate internal hardware failures.

I should also note, the regular nintendo we've been using is split in half and missing a large chunk. The gamecube could be mistaken for new.

Re:Question: why just record?

[Starker_Kull]

Actually, something very similiar to that is already done at many airlines. The data is downloaded at various intervals, and examined for any 'unusual' events - the pilots involved are contacted in a 'non-jeopardy' fashion and asked to explain why something occurred. It has already led to significant improvements in maintainance replacements, and highlighted a few non-optimal procedures that tend to put a crew in a worse place than they started.

The key is that it is non-jeopardy, otherwise the pilots wouldn't speak candidly about the situation and what led them into it, and you might get little or no clue as to what was actually occurring. We call it FOQA [wikipedia.org], and I'm sure various others have thier own names...

Re:It sounds so easy but

[Lumpy]

and the pilots just keep pushing the gas. The plane hit the ground, perpendicular, at 1200kph.

as a amateur pilot it blows my mind that a commercial pilot would freak out about such a failure and continue to throttle up. You have a large number of other indicators you can use. Even in pitch black night and thick fog you have some indicators they teach you in flight school to make it so you dont hit the ground at full throttle.

Re:It sounds so easy but

[hjf]

At the time, the company was in a severe debt (still is but they're much better now, they ordered a couple A380's). Pilots received no training, no simulator, and were forced to work in "if you don't fly, you're fired" conditions. The commercial aviation in Argentina was (and I think it still is) under Argentine Air Force regulations, one of the most corrupt forces. Airports were privatized from the airline desks to the door, but behind that it was still the same. Traffic control wasn't privatized: we have no radars in most airports (the narcs keep it that way), only in the Ezeiza airport. Other radars are powered off at certain times (night, ideal for dropping marijuana loads coming from Bolivia in the middle of the night).

There is no regard for security in air transportation. LAPA 3142 was completely destroyed after aborting takeoff, hitting the fence at the end of the runway, crossing over a busy highway (crushing a Chrysler Neon on its way) and finally crashing into a gas thing. Yes: the runway points straight into a highway and in the middle there are underground gasoline and gas pipes.

In the movies "Fuerza Area S.A." (Air Force Inc.) and "Whiskey Romeo Zulu" (LAPA 3142 was LV-WRZ), former LAPA pilot Enrique Piñeyro explains the causes of both accidents and the situation of aviation in Argentina. Fuerza Aerea is a documentary, WRZ is a movie (based on the true story).

Now, take both movies with a grain of salt: Piñeyro, as a pilot, tries to defend other pilots. But I, personally, think that if you're not trained to fly in other-than-ideal conditions, or if you don't know what to do when alarms flash, you should not fly. The same if planes are not in condition (in LV-WRZ, Piñeyro asks the maintenance staff about the engine fire extinguishers (IIRC), and the guy tells him "Just fly carefully"). But pilots never went on strike or anything. Piñeyro justifies everything on the fact that "pilots didn't receive adequate training" and "airplanes were not in 100% condition". And he gets angry when people call it a "Pilot Error" (just listen to LAPA 3142 CVR, you'll hear "beep beep beep beep ..." and the pilot asking "what's that?")

Re:It sounds so easy but

[LeadSongDog]

Now, put that Ipod nano inside an old sneaker and try again. The Ipod might fail, but the chip inside will survive and be readable.~~~~

Re:It sounds so easy but

[CharlieG]

In reality, the only part that really matters is the memory, the rest is a luxury. First thing you do, make sure you use g-10 glass boards and ceramic package chips. Next. You talk to your ME, and he calls say, Dow Corning to talk about potting compounds. Depending on different electrical, cooling, fire and other needs, they pick out a potting compund. Of the top of my head, it's probably be one of the glass bead filled compounds, as I can't see a memory chip, or a dozen needed serious cooling capability. At this point, your ME designs an enclosure and connector - probably some fairly bulky ampex type. The entire enclosure, with the memory inside is potted (don't forget to pull a vacuum to get the gas bubbles out) and the enclosure sealed - often a metal can soldered closed. Yhen that MIGHT go in another can - and then that goes inside the rest of the recorder, and it goes in it's own can

During all of this, he's been consulting with the guy who runs the environmental test lab, who probably has more experience in what really happens in the tests - what tends to fail, and what tends to work (in a flight data recorder, I'd be worried about the ingress/egress points of signals - ditto the "memory block") - connectors tend to shear. Moving parts are "bad" (hence the FAA wanting to get rid of magnetic tape, with it's motors etc)

Your environmental test guy then either takes the prototype/early production unit to his lab, and beats on it per the spec, or, more likelike for prototype acceptance testing, calls one of the dozen or so places around the country (such as http://www.daytontbrown.com/ [daytontbrown.com] Dayton T Brown or http://www.aeco.com/ [aeco.com] American Environments (Both on Long Island due to the fact that there used to be 2 Airplane Mfgs here, plus a lot of electronics companies), and you have THEM do the testing for the spec. BTW Your test guy and your ME will probably work with your internal machine shop to build the mechanical test fixtures, and an the test guy will work with the EE and the prototype wiring shop to build the electrical test fixtures - so that all these fistures survive the testing environment

While it not "simple" or "every day", it IS almost routine. I probably used to put something through some sort of acceptance test 1-2 times/year (and tests could take weeks to months). Sometimes things don't work - and it's back to the ME/EE and saying "OK, here is what failed" - and why - and doing a re-design