Without looking I'd bet there is an FAA Technical Standard Order for data recorders. That document would points to the applicable specs. All the FAA TSOs are listed on the FAA.gov website. Other countries, for ex. Australia and EASA for (Europe) also have TSOs or TSO like documents for the aircraft products they certify.
] stated that the "black boxes" must withstand [among several severe tests] 5000-PSI and 30-day sea-water immersion [at 5000-PSI?].
NOTE: another article I just read stated that the pressure of seat water increased 1-atmosphere roughly every 10-Meters of depth.
So, for a depth of 7600M = ~760-A X 14.7 = 11,172-PSI... far greater that the crush requirement of the boxes [per the article].
The article also goes on to state that the ULB [pinger] must function at depths up-to 14,000-ft for 30-days... about 1/2-the depth possible in the area of the accident.
No argument from me, Wil. Although, I though the depth was more like 6100 m, but that's still well above the design requirement. But, that also doesn't mean that it can't survive. And even if the actual box is crushed, the mamories might still be OK if they pulled out fast enough. The bigger issue is that the ULB's transmission range is limited, and if it didn't survive the depth, then there'd be no signal to home on to recover the pieces of the box.
In the Pub thread, we were discussing a redundant FDR that would deploy upon a water landing.
Although constant information supply is difficult, airplanes rarely catastrophically fail without warning, couldn't when pilots first notice a significant incident set the plane for continuous air to land data supply.
cloa:
They could have an emergency 'button' or whatever and when it is pressed, all data is streamed to a remote source. Little or no bandwidth required.
I don't think that bandwidth is an issue. They have massive computer systems that are easily capable of handling the amount of data. It can be rendered relatively secure, I would think.
i'd reckon that when the brown stuff starts to hit the fan, the guys are too busy trying to understand the situation and figure out how they can save their (and their passenger's) a$$es to want to take time out to "push a button". the automatic transmit could work off the master, or any, caution light. sure you'd get a bunch of easily resolved situations, as most are; and sure, you won't get all the data (many situations develop from innocuous triggers) but it could be a workable backup to the existing black box.
IRStuff,
The CVR was an endless loop, (20 min.?)while the FDR was not. Had to be taken out at regular intervals for a read-out.
This is info from type courses, never having worked on the actual units. Pls correct if I'm wrong.
Dan
Current FDRs are effectively endless loops, they simply record over the oldest locations, not that different than the endless wire recorder. There is no requirement to remove FDRs. That would have been a bonehead design, if true.
They are required to pull FDR to verify compliance to requirements, such as the 30-day life of the ULB, and to verify that the memory hasn't gone south. Modern nonvolatile memories have something like 100k R/W cycle limitations, but there are random failures that would be annoying if they happened to be where the last bits of crash data was written to.
Docket No. SA-532
Exhibit No. 12
NATIONAL TRANSPORTATION SAFETY BOARD
WASHINGTON, D.C.
GROUP CHAIRMAN’S FACTUAL REPORT OF INVESTIGATION
COCKPIT VOICE RECORDER
DCA09MA026
(47 Pages)
Page 5
"The CVR had no apparent external damage, other than being wet. The underwater locator beacon (ULB Dukane Model DK100, s/n DM1661, battery expiration date October, 2009) did not function when tested. After shorting the center electrode to the case, no sound was detected using a Dukane Ultrasonic Test Set Model 42A12. The beacon was also tested using a Dukane Test Set Model TS100, which indicated “Open Probe/Batt.”
A burst of last-minute automatic messages sent by Air France Flight 447 includes one about a problem with a rudder safety device but that does not explain what sent the jet plunging into the Atlantic Ocean, an aviation expert says.
The industry official, who has knowledge of the Air France investigation, told The Associated Press that a transcript of the messages posted on the website EuroCockpit is authentic but inconclusive...
One of the 24 automatic messages sent from the plane minutes before it disappeared points to a problem in the “rudder limiter,” a mechanism that limits how far the plane's rudder can move. The nearly intact vertical stabilizer — which includes the rudder — was fished out of the water by Brazilian searchers.
“There is a lot of information, but not many clues,” the official said, speaking on condition of anonymity because he is not authorized to discuss the matter.
The official said jets like the Airbus A330 automatically send such maintenance messages about once a minute during a plane's flight. They are used by the ground crew to make repairs once a plane lands...
If the rudder were to move too far while travelling fast, it could shear off and take the vertical stabilizer with it, which some experts theorize may have happened based on the relatively limited damage to the stabilizer.
The industry official, however, said the error message pertaining to the rudder limiter did not indicate it malfunctioned, but rather that it had locked itself in place because of conflicting speed readings.
Investigators have focused on the possibility that external speed monitors — called Pitot tubes — iced over and gave false readings to the plane's computers.
clearly from these news posts (i got mine from yahoo) ther is already aquite a bunch of radio chatter between the plane and base, intended for maintenance not accident investigation.
i don't like the theory that the control system force the tail off the plane, though it could have. if the control system failed it could have commanded full deflection (it's supposed to go back to the last "known good" position; why not go back to neutral ??) but deflection is going to be limited by actuator power and at max speed deflections are going to be small. they can still exceed the design parameters, but it seems unlikely to me that the actuator attmts, the rudder structure, the rudder attmts are all stronger than the fin attmts.
i'd've thought that the system might have a structural fuse, like the brkts attaching the actuator to the rudder ... question is is an uncontrolled rudder better or worse than no fin ?
Seems to me that a rudder than can rip the fin off a plane in its operating envelope is attached to a fin that's grossly underdesigned.
That's like telling a new car owner that he should never turn his steering wheel to its limits because that might cause the wheels to fall off.
I'm having trouble grappling with the notion that Airbus would have designed a plane that would fall apart while only marginally exceeding its design operating range.
That's just literally an accident waiting to happen.
whilst i agree that it's unlikely for a rudder being driven by a command signal that max's out the actuator to rip the fin off, it doesn't have to be as dire as you post, IR.
the aero boys calculate the rudder force required for the different design cases, and the stress boys (finest chaps) design the structure to handle them. along the line they include failure cases. with everything working normally, everything's "just peachy". it the failure modes that make for interesting work ... the pretty much endless games of "what if". something has to break first, it's a matter of how much thought you put into it ... the guys sizing the actuator brkt may well think "let's never have a failure of the brkt" ... this'd cost the plane <1 lb and normally i'd call this "good excess weight". but the brkt could then be strong enough to allow the system to exceed the design envelop of the rudder ... maybe the rudder broke in two, one 1/2 unrestrained could cause the fin the flutter; i'd like to think that the rudder could handle this failure case.
what would the flight-path be like with the rudder hard over, without a pilot input ?? (sounds like a spiral into the ground, as the inside wing stalls and the outside wing generates a lot of lift).
it me the rudder is possibly a red herring ... if the pitot tubes iced up that'd affect the H. stab (pitch trim), though if the plane was close enough to stall, then the lateral trim would start to get a work-out. but we don't know the details of what they found with the V. Stab
