Go Back  PPRuNe Forums > Flight Deck Forums > Tech Log
Reload this Page >

AF 447 Search to resume

Tech Log The very best in practical technical discussion on the web

AF 447 Search to resume

Old 10th May 2010, 08:38
  #961 (permalink)  
 
Join Date: Jan 2008
Location: uk
Posts: 857
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by OVERTALK
[...]
Large unintended rudder deflections during a high-speed LoC event may have partly detached the vertical stabilizer from its mounts, or even locked the rudder at a significant deflection.
Only problem with that is that the mounts didn't fail (or rather the VS didn't fail around the mount points, as it did in with aa587) - they were still attached to the VS, but ripped out of the fuselage.

Something clearly happened to take the a/c back to where the search is now though. My feeling is that it was in the air, possibly with control partially recovered, for longer than thought - but I don't have any way to explain loss of satcom to back that up.
infrequentflyer789 is offline  
Old 10th May 2010, 08:57
  #962 (permalink)  
 
Join Date: Jun 2009
Location: Paris
Posts: 691
Likes: 0
Received 0 Likes on 0 Posts
Hello HN39,
Originally Posted by HazelNuts39
takata;
as shown in the diagram produced by mm43 #946 (my thanks to mm43 for that) the negative relief valve is not controlled by the CPC. Intended as an 'emergency' backup, it is often no more than a simple mechanical device that requires no more than a negative pressure differential to open. As explained by Chris Scott, that happens when the aircraft overtakes (catches up with) the cabin, which is unlikely to occur above 7350 ft aircraft altitude.
Edit: Ok about the mechanical device, but this alert is unrelated to it.
takata is offline  
Old 10th May 2010, 09:07
  #963 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
takata;

just a 'dodgy conclusion?'

HN39
HazelNuts39 is offline  
Old 10th May 2010, 09:08
  #964 (permalink)  
 
Join Date: Jun 2009
Location: Paris
Posts: 691
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by infrequentflyer789
Something clearly happened to take the a/c back to where the search is now though. My feeling is that it was in the air, possibly with control partially recovered, for longer than thought - but I don't have any way to explain loss of satcom to back that up
No power.
No engine = no power = no Satcom.
That's my opinion also.
S~
Olivier
takata is offline  
Old 10th May 2010, 09:12
  #965 (permalink)  
 
Join Date: Jun 2009
Location: Paris
Posts: 691
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by HN39
takata
just a 'dodgy conclusion?'
HN39
Well, not so dodgy as this alert is clearly associated to the CABIN PRESSURE CONTROLLER. Mechanical devices operates by themsleves without regards with CABIN VERTICAL SPEED control. You don't need to switch them to auto/manual.
S~
Olivier
takata is offline  
Old 10th May 2010, 09:22
  #966 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
Cabin v/s advisory

Originally Posted by takata
this alert is clearly associated to the CABIN PRESSURE CONTROLLER
could it just be associated with cabin pressure?
HN39
HazelNuts39 is offline  
Old 10th May 2010, 09:31
  #967 (permalink)  
 
Join Date: Jun 2009
Location: Paris
Posts: 691
Likes: 0
Received 0 Likes on 0 Posts
Code ATA is about controller.
takata is offline  
Old 10th May 2010, 09:36
  #968 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
takata has published some spot depths, and the following graphic shows the general bathymetry in the area.



mm43
mm43 is offline  
Old 10th May 2010, 09:44
  #969 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
takata;

There was a post a page or two back where it was implied that the BEA got it wrong about loss of SATCOM. However, take a look at this graphic where the LOS is right when the a/c probably passed through the center of the mesoscale event NNW of ORARO.

mm43
mm43 is offline  
Old 10th May 2010, 10:22
  #970 (permalink)  
 
Join Date: Jun 2009
Location: Paris
Posts: 691
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by mm43
takata;
There was a post a page or two back where it was implied that the BEA got it wrong about loss of SATCOM. However, take a look at this graphic where the LOS is right when the a/c probably passed through the center of the mesoscale event NNW of ORARO.
I was not talking about the possible partial loss of SATCOM during ACAR transmission (neither infrequentflyer789 if I understood what he meant) but about the fact that she was not sending anything more after 02.14 while she could have still been in the air.
I don't buy the upset-recovery-upset-crash theory.
Documented upsets at cruise altitude are ending in a fairly short time pretty close to their initial position. Most of the displacement is on the vertical plan or at very high speed down to the surface. None would imply a possibility of impact like in AF447 case.
Overspeed could be deduced from vertical stabilizer state, it would have certainly left some evidences on the leading edge.
No evidence of structural failures are given by ACARs transmission.
I don't know where AF447 was when ACARs stopped, but I'm suspecting why they stopped and it is then possible to understand why she could have ended 40+ NM (still not confirmed) farther and back on its track.

Beside, this is what Seabed Worker is searching at nearly 4,000 m deep, in mountainous terrain, and it should not be spread over 8 NM of seabed:

takata is offline  
Old 10th May 2010, 12:47
  #971 (permalink)  
 
Join Date: Jun 2009
Location: Paris
Posts: 691
Likes: 0
Received 0 Likes on 0 Posts
I watched the BEA press conference but nothing very new:
- They do not want to speculate. It was answered to journalist question that the wreck probable location imply that AF447 turned back for unknown reason.
- 2/3rd of the search area was already covered, but exploitation is not made at the same time (drones are delivering data after 20 hrs of exploration).
- Two different signals were recorded by Emeraude sensors on 1st July. Signal analysis permited to acertain that they came from airborne distress balises: matching frenquencies, tempo, and modulations. So Navy/Thales people are pretty sure that they picked the right signals and it was confirmed again during another conference with experts.

S~
Olivier
takata is offline  
Old 10th May 2010, 13:02
  #972 (permalink)  
 
Join Date: May 2002
Location: GC Paradise
Posts: 1,100
Likes: 0
Received 4 Likes on 3 Posts
mm43
Thanks for answering the question. Reference to the Cabin Pressure Controller schematic shows that in Automatic Mode the barometric reference comes from the ADIRUs, but my take is that should that reference fail, then it selects Manual and the Safety and Relief valves will attempt to keep Cabin Pressure equal to the outside when descending below 8000/7350 feet. Here's the schematic for good measure -
Cabin pressure is normally controlled automatically but can be controlled by the aircrew in a manual mode if certain failures prevent the normal auto operation.

But, more importantly, if all else fails, the overpressure relief valves work on simple springs that operate autonomously should the pressure in the cabin go above normal differential limits. Also a negative pressure relief valve also works autonomously on simple springs to relieve pressure differential between inside the cabin and outside, if the internal cabin pressure becomes too negative.

I love your dedication, research and posts...many thanks and well done!
FlexibleResponse is offline  
Old 10th May 2010, 14:16
  #973 (permalink)  
 
Join Date: Jan 2008
Location: Blighty (Nth. Downs)
Age: 77
Posts: 2,107
Received 4 Likes on 4 Posts
Possible implications of CABIN VERTICAL SPEED ADVISORY (cont’d)

Thanks for the Pressurisation schematic, mm43. It shows that static (external) pressure is obtained via the Air Data elements of the ADIRUs, as expected (originating from the static ports). Unfortunately, it doesn’t show where within the pressure hull the cabin (internal) pressure signals come from. I think we can assume that any external probes are not involved? If so, any available Cabin VS data (and here we have one) are likely to be uncorrupted by issues such as external icing or excessive flight parameters.

By the way, I think the Safety Valve on the rear pressure bulkhead only opens to relieve grossly excessive differential pressure, not for inward relief.

Am I correct in inferring that the ECAM message could be triggered by either a negative or a positive VS exceeding 1800 ft/min, lasting at least 5 seconds? If so, a double engine flameout might result in a temporary positive (climb) VS of this nature, before the outflow valve had time to close. Once the Outflow Valves had shut, the cabin VS would fall well below 1800 ft/min. System design would not allow a simultaneous loss of both AC (engine) generators to disable both Outflow Valves in an open position. But power to the valve motors might be interrupted temporarily.

I may be revisiting ground discussed previously, but − when considering the possibilities − could we add the above scenario to the more likely one in which the aircraft has descended rapidly to the cabin altitude, and then takes the cabin with it? Has a double engine flameout at high altitude already been ruled out?

Chris
Chris Scott is offline  
Old 10th May 2010, 14:19
  #974 (permalink)  
 
Join Date: Jan 2008
Location: Blighty (Nth. Downs)
Age: 77
Posts: 2,107
Received 4 Likes on 4 Posts
Hi takata,

Quote from takata (Today, 0545z):
“This alert may be related to purely sensor/ADIRUs errors; it may be related to a real aircraft descent rate above 1,800 ft/mn from whatever altitude (in case external pressure is still feeded to the controler by another circuit - but whilch one?); it may be related to crew actions during troubleshooting...”

Can you explain what “external pressure” has to do with cabin pressure (cabin altitude) and cabin VS? What am I overlooking?

Alternatively, why would an ECAM warning of excessive cabin VS be spurious? The creation of the warning implies that cabin pressure data were still available. That being so, what would cause them to be corrupted?

Chris
Chris Scott is offline  
Old 10th May 2010, 16:20
  #975 (permalink)  
 
Join Date: May 2010
Location: MA, USA
Posts: 126
Likes: 0
Received 0 Likes on 0 Posts
Pinger Range

MM43 wrote:

Many thanks for the path loss calculations. I am in agreement with what you say, the only additional problem in deep water searches is the salinity inversion layers - one at 250 - 350m and if deep enough another near 4,500m.
I generally agree. I certainly should have mentioned the additional complications presented by the sound speed profile on the propagation of the pinger signals. The effect of variations in sound speed with depth is well known and long studied. The actual profile is highly dependent on local conditions and it changes over time. The primary influence in the warmer climates results in a higher velocity near the surface (10-300m) due to the warm surface water, giving way to lower velocity in the colder water at 1000-2000m, and gradually increasing again at greater depth due to the constant (2-4C) temperature and the ever increasing pressure. Salinity also plays an important role. These variations (and there can be several or many layers) bend the acoustic rays like lenses or mirrors. However, on any given day, only a sound speed vs. depth measurement will reveal what the present conditions are and what bending/reflection of rays will be in effect.

Generally, these sound speed variations are horizontally stratified over a wide area, and therefore have negligible effect on the propagation along a direct vertical path, perpendicular to the layers. Thus one would expect the path to be unobstructed when the receiver is directly over the source. However, the if the path is only 10 or a few 10's of degrees from vertical, there can be significant attenuation, or, oppositely, regions that concentrate and boost the sound. One common situation, once the receiver is well off the vertical path, is total internal reflection of the sound, creating large shadow zones. These effects will further limit the radius from the pinger where the sound will remain detectable.

The sea is also full of fronts and internal waves that refract and channel sound horizontally in ways that are current topics of research. These effects may play a role in this search also, but are probably less of a factor, because once the angle from the receiver to the source is far from vertical, the range will have reduced the signal strength below detectable levels, anyway.

You might like to share your ideas on what form the Thales analysis probably took.
Signal processing is not my specialty, so there is not much point in my speculating about that. I have some idea where I would start on that problem, but it would be a brute force approach and probably would not yield useful results. The problem is made more difficult by the fact that the frequency and timing tolerances of the pinger are quite loose; it obviously doesn't use a crystal, if I'm looking at the right data sheet.
auv-ee is offline  
Old 10th May 2010, 17:05
  #976 (permalink)  
 
Join Date: Jan 2008
Location: Herts, UK
Posts: 748
Likes: 0
Received 0 Likes on 0 Posts
mm43, good work continues from the original inspirational stab, it will be a year soon; many thanks to you, takata and others of course

Excuse my ignorance but I can't trace back to find a legend for the depth countours

Is red shallow and dark blue deep? or vice versa?

===
Any chance we can forego relentless discussion of CPC systems etc. I fear there may be little further to learn now. As we said in the old thread... we just need the boxes now, because whilst the ACARS msgs may indeed be factual they might nevertheless be misleading until we have some real boundary conditions for WIHWH and WHEN

No? Ok, then...

Last edited by HarryMann; 10th May 2010 at 17:30.
HarryMann is offline  
Old 10th May 2010, 17:43
  #977 (permalink)  
 
Join Date: May 2010
Location: MA, USA
Posts: 126
Likes: 0
Received 0 Likes on 0 Posts
Pinger Range

auv-ee wrote:

As for the ambiguity in location, I doubt that it is due to navigational uncertainty, but rather due to acoustic localization uncertainty. I have no idea what signal processing was done to actually extract the signal from the noise, but it would likely have involved integration over long periods of time: as long as the sub, traveling at, say, 3-6knts to minimize self-noise, would have been within a few of km of the pinger. After a long integration, one would only be able to guess the location with a resolution of roughly that integration "distance" (time, times the sensor transit speed), i.e a few km.
Had I bothered to read the relevant sections of the two interim reports prior to posting, I could have avoided guessing the speed Emeraude transited while listening: 6-10 knots is listed in the second report.

Normally, a 20db signal-to-noise ratio is required for simple detection, resulting in the commonly quoted 2-3km detectable range.
Also, I see that the second report lists 2000-3200m as the assumed detection range for Emeraude, consistent with the above conclusion.

The report also cleared up something I was wondering about: why were the TPLs from the US Navy not towed deep enough to hear? In fact they were towed deep, but they covered only the northern "ALPHA zone". If the Emeraude analysis is correct, they were simply towed in the wrong place.
auv-ee is offline  
Old 10th May 2010, 18:00
  #978 (permalink)  
 
Join Date: Jun 2009
Location: in a plasma cocoon
Age: 53
Posts: 244
Likes: 0
Received 0 Likes on 0 Posts
"Le Figaro" reports that the two data recorders are probably not colocated but separated by a 1.5 to 4 NM range.
Le Figaro - France : Le vol AF 447 livre ses premiers secrets
Jeff
PS) on the bathymetric maps, red is ~ -1000m deep (slightly above), blue is ~ -4000 m (slightly below)
Hyperveloce is offline  
Old 10th May 2010, 19:19
  #979 (permalink)  
bearfoil
Guest
 
Posts: n/a
The "turn back" was broached quite some time ago. Engine(s) loss was discussed as well. Early separation of the VS seemed problematic due to antenna location. If the black boxes are indeed 8 km apart, disintegration in flight looms back into view. Then again, Figaro tells me the Muslim "problem" is overblown, and 7 kilos of Heroin were intercepted, not a gold standard of journalism, imo.

Chris Scott reminds us of the Cabin P alert and loss of the pressure vessel seems a possibility. Since it was (ACARS) last, a descent to and through 8000 feet seems supportable (with cabin integrity), and the alert also could be saying "We're lower than the Cabin, and can't keep up." The alert also suggests that pressure inside was way low, and 8000 feet cannot be maintained, something that could happen at any altitude (seemingly) from 37k down to 8000, if indeed the warning was genuine and based on true sensing.

Although PJ2 didn't agree that a turnaround was done "Immediately", it is conceivable that had the Pilots turned to miss the Wx and experienced an uAS in concert with the turn, they may have made it into a 180 rather than a 30.

For keeping the thread going with brilliant data and graphics, much is owed to mm43 and others. It is a good thing that plenty of folks are keeping up with the investigation, for many reasons, not the least of which regards mitigating the factors that caused this tragedy.

bear
 
Old 10th May 2010, 19:26
  #980 (permalink)  
 
Join Date: May 2010
Location: EU
Posts: 2
Likes: 0
Received 0 Likes on 0 Posts
location of pingers

The uncertainty comes from : the sub's navigation uncertainty added to the bias on sound propoagation. Considering that reckoning the actual sound velocity profile in such environmental conditions proves rather difficult, the expected detection range will greatly vary. therefore, when eventually you have something on your sonars, you never know whether your are at the limit detection range or whether you just happened to cross a specific sound ray. Most of the time, you'll assume that you were at the limit detection range, which increases the uncertainty.
Also, if you want your position record to match your recordings, in order to be able to determine where you were sailing when the detection occured, you need a very accurate time measurment on both logs. This adds too to the final estimation.
savu is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.