AF 447 Search to resume (part2)

[auraflyer]

Garageyears wrote:

What will take time is ensuring the exact transcription is provided, since in many cases the audio is not exactly the clearest, particularly when things are not exactly going as planned.

Pardon my ignorance, but is the CVR data in "stereo" (ie is there input from multiple mikes, positionally offset) or is it just "mono"?

[Chris Scott]

auraflyer,

Even on the BAC 1-11 circa 1980 (UK-registered), the CVR had 5 channels:
each pilot's mic "hot", i.e, live even when pilot not transmitting on radio or intercom;
input to each pilot's headphones (mono);
area microphone.

Quality of audio was excellent (playback by AAIB, Farnborough). Area microphone(s) may not be as good as the others.

[jcjeant]

Hi,

Jean-Paul Troadec interview

Google Vertaling



Original source:
Vol*Rio-Paris.*"Toute la lumière sera faite" selon le directeur du BEA - France - Le Télégramme

It is concluded that the means were good but we did not look in the right place. So we explored the rest of the circle starting with the area where probability find the plane were the greatest

I dont understand this statement ...

So BEA know where find the plane were the greatest ... but anyways and contrary to this knowledge ... they searched firstly in other aeras .....

[bearfoil]

jcjeant

I don't know how it is done in Europe, but when an a/c flies into a Thunderstorm in US, we look with 99 percent certainty directly underneath the storm, not miles away. Our hurricanes only get their start in the ITCZ, they finish up in Louisiana or North Carolina. In the midwest, it is thunderboomers. I never could follow the logic of the search for 447.

[JD-EE]

Originally Posted by robertbartsch

Would anyone like to speculate on how these boxes survived in 10,000 feet of water for so long? I would have thought the pressure alone at that depth would have made this unlikely.

As noted it is designed to withstand even greater depths. It is a multi-layer package of very strong materials. It may have some electronics within that pressure vessel package. All the electronics on the outside would be toast. But the contents of that pressure vessel, as long as it is not breached by an impact during the crash, should be OK.

The description given of reading the device indicated they mounted it on a new set of electronics and discovered it was fully readable. I am not sure if that is the "memory chips" were remounted in a new Solid State Disk package with its electronics or whether they just used the pigtail sticking out of the package to connect to the new external electronics box is not 100% clear. I am inclined to believe the latter hypothesis given that it came together remarkably quickly.

I did notice that the pigtail that connects to the larger electronics package appeared to have disconnected cleanly in the one case and was still connected in the other. So both should have been in useful condition for simply plugging it into a new box and proceeding. I'm not sure if they even bothered to open the cylinder to visually check the contents. I would probably have done that before applying power and risking there being water inside.

[slats11]

I never could follow the logic of the search for 447.

My understanding is that they did look around LKP from the very beginning. They found nothing on the surface initially until they searched further north. Nor did they hear the pingers.

So what would you have them do after this? Remember at that time they probably had more faith that the pingers were functioning than in their understanding of what had taken place and where the plane was. That is to say that they likely reasoned that the pingers were working elsewhere, rather than the plane was near LKP and both pingers failed. Yes this reasoning turned out to be incorrect, but I can certainly see what they were thinking.

Phase 3. They chose to search out NW, based on independent sophisticated drift analysis. This also turned out to be incorrect. But again, I can understand their thought processes.

Phase 4. Forget drift analysis. Forget pingers. Start at LKP with the group with the best credentials to do the search. And they succeeded.

I don't see a conspiracy here. To me, this reflects a determined search in the face of lots of unknowns, a few false leads, and the unexpected failure of both pingers.

Will there be lessons learned from this? Sure. But I don't believe you can question the methodology of the search.

Part of the problem here is that some of the involved parties have less that a perfect record. And so people are looking for any evidence of a repeat here. That is understandable, and on the whole it is a good thing. A degree of skepticism is healthy. Only a degree however, only up to a point. Beyond this it starts to resemble paranoia. Courts of law have to decide innocence or guilt on the facts presented - and not on the past record of the defendant.

The way this investigation is running and the involvement of external agencies gives me a degree of confidence that we can trust the conclusions. Sure there can still be bias - and as someone else hinted, subconscious bias can be the most dangerous type of all. But bias does not equate corruption.

[RR_NDB]

Recorders info

auraflyer,

but is the CVR data in "stereo"

4 channel ("stereo"), look the specs here: Info on AF447 CVR

Observe the channel 4 has better specs

[JD-EE]

Originally Posted by Annex14

Would anyone like to speculate on how these boxes survived in 10,000 feet of water for so long? I would have thought the pressure alone at that depth would have made this unlikely.

And titanium.

[RR_NDB]

Salt water in PCB

Hi, JD-EE

I would probably have done that before applying power and risking there being water inside.

Much safer!

What you think about LF info yesterday?

[mm43]

jcjeant

Jean-Paul Troadec has responded appropriately to the question asked. He did mention that the area in which the debris was found was originally searched (US Navy TPL towed by "Fairmount Glacier") and nothing was heard. That was during Phase 2 and between then and Phase 3, the BEA engaged the Drift Group to try and narrow down the area to search.

The group came up with a area of probability that was to the NNW and around 30NM from the LKP and encompassed around 2,000 ^2 km, which was searched using towed side-scan sonar and AUVs equipped with SSS. Phase 3 was extended and the areas to the west and south of those already searched were covered by AUVs. There was an unfortunate interlude which lead the search to be moved over 40NM to the West by South of the LKP and this deviation was responsible for the debris not being found during the extended Phase 3 search.

Following Phase 3, the Metron Report made a statistical analysis of all the data they had at their disposal and concluded that the aircraft would be found within 20NM of the LKP.

Phase 4 under the operational control of WHOI used 3 AUVs and resumed searching from where they had left off at the end of Phase 3, ie about 10NM NNW of the LKP. Working methodically through the remaining area north of the LKP, it took them about one week to come across the debris.

The reason for neither of the pingers being heard remains to be found, and as far as I know, there has been no report of the FDR's missing ULB being located. I am sure that that line of enquiry will be run in conjunction with the ongoing FDR/CVR data analysis.

Overall, the initial air/sea searches for floating debris failed badly, but the ongoing planned secondary searches for the ULB pingers and then using alternative methods was done well.

[Machinbird]

I'm not sure if they even bothered to open the cylinder to visually check the contents. I would probably have done that before applying power and risking there being water inside.

Wasn't that why there was an extended drying time in a specialized oven before reading the chips? It wouldn't make sense unless they had opened the cylinders and taken out the cards holding the memory chips. I would think that the main thing holding the ocean out of the memory modules was a well situated O ring on an end closure.

[JD-EE]

Shadoko - regarding slammed cockpit door with everybody outside I'd simply comment that such a situation would be an unbelievable lapse of proper procedure or issuance of an official procedure with a glaring error. Three people can navigate around each other in the cockpit quite easily. So for changing crew and assignments there would be two procedures.

The first procedure simply substitutes one person in once seat. The new person moves into the cockpit over in front of the fourth seat. The person being replaced rises and exits past the new person. The new person emerges and enters the vacated seat. At that time the replaced person exits the plane. At no time is there nobody at the controls.

The second procedure puts the captain back in his seat when the person being replaced is the person in the other seat. Captain enters the cockpit and moves to the space in front of the fourth seat. The person in the captain's seat rises and moves to the door but does not exit. The captain moves into his seat, gets settled in. The person standing moves to the space in front of the fourth seat while the captain assumes control. The third person rises and moves to the door. The replaced person moves into the seat the third person vacated. Once everybody is comfortably at the controls the third person leaves the cockpit. At no time is there nobody at the controls.

It's a simple programming job, if you think about it. What on Earth would cause all three to vacate the cockpit? (Besides, doors I've seen of late have cipher locks on them so they can be opened from the outside. Before that I believe they could only be locked from the inside by a positive action by one of the cockpit crew. If that's still the case the person planing to leave the cockpit in the above two scenarios would exit the cockpit and the last crew member to sit back at the controls would lock the door from the inside.)

(How do you put an elephant in the refrigerator? How do you put a giraffe in that refrigerator? Think about it....)

[JD-EE]

Originally Posted by RR NDB

We don´t deserve!

As if you Europeans didn't work hard to earn it?

(If you're going to get political, so can I. )

We'd best drop this line of discussion.

[JD-EE]

Shall we mark 2011 May 16 as the date deSitter gave up flying anything other than antique airplanes or third world airlines? They'll be all that still has direct control in a few more years. And I presume that broad statement of yours indicates you do not like autopilot at all. Didn't even the DC3 have some form of autopilot?

[JD-EE]

bearfoil, how many times do you remember me commenting about how unfriendly it is for the computers to toss up their metaphorical hands and deliver control to possibly surprised and unready pilots, especially when it could raise a flag and continue flying by the same rules the pilots are supposed to adopt in that situation?

I remember getting shouted down, too.

Of course, if THAT is what happened, AB would be into liability up to its neck unless it was following some higher authority's order. In that case the higher authority (the government?) would be at fault. Oh, my, would THAT be a picnic.

(Since when have governments been able to perform engineering? They do it nevertheless with the arrogance that comes with "POWER". "Because we can and we have a 'God given' right to do it.")

[JD-EE]

Originally Posted by bearfoil

If ACARS ended at altitude, I think it may be time to reinvigorate the proposition the a/c lost some parts, on the way down, even just post the last vertical speed message.

Yeah, the hamsters died of hypoxia when the compressors slowed down. They then fell out of the engines at altitude.

Bear, you should know me well enough by now to know I really have a hard time passing up straightlines....

[JD-EE]

CogSim, regarding two seated pilots the Chinese fire drills I posted can be tweaked to permit both of the seated pilots having had a chance to empty their bladders as needed during the interchange without having only one person at the controls for more than 20-30 seconds at a time. And when there were only two in the cockpit both would be seated at the controls. All they need to do is work through the procedure to see it can come together naturally.

With the bad wx zone so close I am sure they'd be all settled in for the duration of the crossing. Then they'd work out whatever was needed for emptying bladders when it was safe again.

Of course, conversation regarding radar here seems to indicate they might not have had the foggiest idea there was bad weather ahead before the fit hit the shan.

[JD-EE]

GarageYears, I think "synching" in context would be more like deriving a full sequence of events: A leads to B leads to C and D. C leads to E ... D leads to H, I, and J... Once that is done they have to figure out what the precise timing of each of these events in the cascade mean.

Everything is in order with even times. Now, how do those times really evolve into an event tree? Did B really cause C and D or was something else the real cause? This is where the brain burns out from the complexity that can evolve. (I presume they have procedures derived from long experience to mitigate the complexity exceeding brain capacity. Computers will help. But without the human brain and its experience computers might give quite inaccurate results even accounting for reaction times.)

[JD-EE]

Originally Posted by wes_wall

Anyone care to comment on when BEA might think summer is? Officially, it is the day of the year when the Sun is farthest north (on June 20th or 21st). This day is known as the Summer Solstice.

Have you ever heard the quaint phrase "Microsoft Minutes"? It comes from the presentation of estimated time until a task, say a large file copy, is complete. You often see it sit at 3 minutes for several minutes; you see it skip backwards even by large amounts; you see it skip forwards even by large amounts; it's all "Microsoft Minutes".

I propose a similar measure "BEA Summers". (Did they say WHICH summer?)
I want to see the results almost as much as those most anxious. But I am not going to break a sweat over it until at least the end of the warm portion of the year in the Northern Hemisphere.

[JD-EE]

gums, I'd be inclined to add another point to your fine set. There should be a procedural change that says a human pilot must control and fly the plane for at least one hour out of every 10 in the air in stretches no smaller than half an hour. That is probably not enough to give the pilots experience for what they will be handed if things get too rough. But, I suspect it will help.