AF 447 Search to resume

[DenisG]

As the majority of parts have been recovered on 15-JUN-2009 and 16-JUN-2009, this may appear the average stream distribution speed in order to derive the impact location.

Image: Debirs and bodies found on 16-JUN-2009.

[Hyperveloce]

Thank you for the last Seabed Worker's positions MM43: the 23:53Z one is well within the E-24 (red) square probed by the Emeraude (p 82, interim rep. #2). I also find it difficult to believe that this was an intentional departure from the initial route (or a U turn) as suggested by many newspapers today.
Jeff
PS) Hope that the BB are still colocated or close to big/visible debris of the plane !

[HazelNuts39]

Originally Posted by mm43

as the a/c would appear from analysis of the ACARS messages to have been in a deep stall for over 1min prior to impact

mm43;
for my own understanding, I would be most grateful if you could elaborate on that statement.

HN39

[DenisG]

Hmm, I am not familiar with that source. The BEA report stated a "bank angle close to zero at impact." (source: BEA-AF447-Interim Report, 17-DEC-2009, page 39)

Denis

[Pugachev Cobra]

GPS?

I've been thinking, I don't know about water crashes statisticis, but given the possibility of the tail fin floating above the water, wouldn't it be extremely low cost to deploy small gps trackloggers on the fins?

I know that planes aren't made to crash and it wouldn't work all the time. But with AF447, if there were such a device attached to the fin, with a battery and recording its position every minute or so, tracklogging its gps position up to i.e. 30 days, we could surely pinpoint the exact impact point and the surface drifting currents by the subsequent tracklogging.

It wouldn't necessarily need to be attached to the fin, maybe just a device that's easily detachable and floats on impact?

Just my 2 cents...

[DenisG]

Similar recommendations have been formulated in the 2nd BEA report. (source: BEA-AF447-Interim Report 17-DEC-2010, pp71f)
Denis

On the basis of this work, le BEA recommends that EASA and ICAO:

1. extend as rapidly as possible to 90 days the regulatory transmission
time for ULB’s installed on flight recorders on airplanes performing
public transport flights over maritime areas;

2. make it mandatory, as rapidly as possible, for airplanes performing
public transport flights over maritime areas to be equipped with
an additional ULB capable of transmitting on a frequency (for
example between 8.5 kHz and 9.5 kHz) and for a duration adapted
to the pre-localisation of wreckage;

3. study the possibility of making it mandatory for airplanes
performing public transport flights to regularly transmit basic
flight parameters (for example position, altitude, speed, heading).

In addition, the BEA recommends that ICAO:

4. ask the FLIRECP(19) group to establish proposals on the conditions
for implementing deployable recorders of the Eurocae ED-112
type for airplanes performing public transport flights.

I really liked the: "... le (!) BEA recommends..."

[mm43]

originally posted by mm43..
as the a/c would appear from analysis of the ACARS messages to have been in a deep stall for over 1min prior to impact.

HN39;
You are right, I was a little "loose" with my wording.

What I should have said was that the a/c on completing the left bank made 40NM in less than 4 min, yet impacted with a high vertical and low horizontal speed. The assumption is that the relationship between the vertical and horizontal speeds was as a result of a deep stall, where the loss of lift vortex created the wings level and small positive pitch on impact. To achieve both these conditions, i.e. covering the distance, yet impacting in the condition described, possibly resulted from an initial overspeed followed by a stall.

Not particularly wanting to revisit the "done to the death" ACARS messages -

FLAG ON CAPT PFD FPV and FLAG ON F/O PFD FPV (2 h 11)
This message indicates that the flight path vector (FPV) function is selected but
unavailable. In order to lose completely this function, which is elaborated by
the three IRs, in a way that is compatible with the CFR, one of the following
three conditions must be met for each ADR:
* barometric vertical speed higher, as an absolute value, than 20,000 ft/min,
* true air speed higher than 599 kt,
* measured calibrated airspeed lower than 60 kt.

- the second condition is possibly the trigger in this case.

mm43

[DenisG]

Also at 2.11h, from BEA-AF447-Interim Report-17-DEC-2009, p. 37

ISIS (22FN-10FC) SPEED OR MACH FUNCTION (2 h 11)

• a CAS higher than 530 kt without the Mach value exceeding 1. This
  condition implies that the aircraft was at an altitude comprised between
  about 4,000 and 14,000 ft;

Could somebody explain to me, why it can be derived that this can only have occurred between 4,000 and 14,000 feet? Why not lower or higher? And why then still so unspecific?

To mm43:

• "at 2 h 10 min 34, the last position received was latitude +2.98° (North) and longitude -030.59° (West)" (BEA-AF447-Interim Report-02-JU-2009, p.48) or to us 'last known position'.
• Still, the first 4 ACARS occurred in the 20 seconds before this position, including TCAS off.
• After that we have receiced the above speed or mach function, implying that the a/c was already pretty low at this point (between 4,000 and 14,000 feet, at 02.10.54h), making the slowest average decline guess at around 60 feet per second after that.

Q: This also means that the decline must have begun before the first TCAS messages arrived at 02.10h, right?

Thx.
Denis

[mm43]

Update: "Seabed Worker" - position

Take note of the positions - indentical for 2 hours!

7 May 2010 12:12z Hdg 272.5 Spd 0.4 2°42'37"N 31°12'36"W
7 May 2010 10:33z Hdg 262.8 Spd 0.2 2°42'37"N 31°12'36"W



Suspect the ROV is in use.

mm43

[takata]

to DenisG & mm43:
So you are both making the assumption that everything was fine/working with probes and:

..ISIS (22FN-10FC) SPEED OR MACH FUNCTION (2 h 11)
..FLAG ON CAPT PFD FPV and FLAG ON F/O PFD FPV (2 h 11)

were triggered by real pressure indications?

What if pressure (=speed, altitude, etc.) was wrongly calculated by frozen probes?
Condition for system faults doesn't mean that real flight conditions were actually matching those cases. In fact, it wasn't the case for all aircraft displaying similar faults in similar flight conditions (probe freezing). Just notice that most systems feeded by probes were displaying errors at the same time.

[DenisG]

Hi takata.

Not necessarily.

Assuming that the sequence of events must not correspond to the sequence of incomings ACARS (as noted by BEA), but should be correct within a minute (BEA), I wonder then:

If ISIS (22FN-10FC) SPEED OR MACH FUNCTION (2 h 11) implies the a/c was between 4,000 and 14,000 by now, the impacting event must have occured before those first ARCARS were received, including the 1min error. Coming down from 35,000 to 14,000 should take a few seconds more before that event.

But I can only follow on, if I understand, why these 4,000 - 14,000 can be established from the ACARS.

I am trying to think through the possibility that those ACARS and probably pitot tubes incl. were perhaps one result of the decline, but not the cause of the decline, if the decline had begun some minutes before that, as an altitude of max. 14,000 at approx. 02h 10min (- 1minute) (if correct by BEA and I do not understand yet) would reasonate.

Denis

[CONF iture]

The new search area seems to be 40 NM west south west of the last known position.

After all it is not that far away from the possible impact position as mm43 estimated it in its post #739

[takata]

Hello,
Last July, I've made the assumption, based on body recoveries that the actual plane crash might have been South of last known position (0210), and that the search was not primarily conducted at the right place (mostly because wreck was not located).

In fact, the primary searches were focused on the "upset theory": loss of control followed by crash very closely following the last ACAR sent.

But another possibility existed: that the end of ACARs was not due to aircraft crashing but to loss of NAVSAT. Consequently, this aircraft could have crashed somewhere and sometime after this last ACAR was transmited. Nothing in those ACARS was telling us that this aircraft was in total perdition at 0214. It might have continued for some time on its flightplan, then turned back due to bigger failure than probe icing, but of course, one causing NAVSAT failure. This also ruled out a critical airframe damage as this aircraft should have been able to fly much longer in this case. The main difficulty in order to verify this hypothesis was addressing body drift which seems to be quite difficult in this area.

So far, everything is confirming this hypothesis. The airframe was possibly intact at impact time; the bodies drifted North-East, and if the wreckage is to be confirmed at this spot, it means that it turned back and was very possibly still under crew control after 0214 for quite some time:
- we still don't know at what point decision was made to flight back.
- we still don't know what could have caused NAVSAT failure, but we have a suspect called ICE.

It is now more and more improbable that any decision to turn back could have been made before this last ACAR was actually transmitted: because of probe icing, the following system failures, and due to flight conditions into this thunderstorm zone. This could have ruled out any change of direction until something much more critical happened. By 0214, this aircraft was almost out of the thunderstom zone, it seems hard to decide to come back on it or to attempt any maneuver being surrounded by high conv cells.

NAVSAT failure is either NAVSAT related, then certainly non critical for flight safety, either, it is power related. Ice (the only one proved suspect so far) doesn't make power supply to stop until both engines are shut down. As a matter of fact, Ice may really cause dual engine flameout. See here: Engine icing & flameout at 38,000 feet/cruise (PIC)
No more ACAR recieved would imply that they would have been simultaneously shut down.

It would be nice to know from an experienced pilot, what he would do in such a case:
- severe icing observed, loss of speed indication and systems, alternate law2, while being in the middle of a thunderstorm zone (supposed cleared by deviation but aircraft surrounded by it).
- event lasting more than 4-5 minutes of flight, then dual engine flameout.
- what will he do next ?

Last, question about FDRs:
- If power was lost, what happened to FDRs records? are they stopped also in this aircraft?
If so, and even if the FDRs are retrieved, recorded data between 0210 and 0215 would be difficult to figure out due to probe icing, and nothing more to expect after this point.... same for voice recorders.

S~
Olivier

[jcjeant]

Hi,

Methink it's some emergency eletrical power sources in those planes ... like batteries ..
I can be wrong .. but seem's I read it somewhere ......
I suppose they are not only for the passengers courtesy lights
Edited:
Yes indeed .. I forget the RAT (the green power )

[NeoFit]

Emergency power supply = Ram Air Turbine (RAT)

[takata]

If one doubt that could have happened to A330 @ FL400, dodging tropical thunderstorms above ocean, see this article bellow:

As a Qatar Airways flight dodged thunderstorms on approach to Shanghai in 2006, it encountered a problem that, until recently, was considered virtually impossible: nearly four miles above the earth, both engines of the big Airbus A330 shut down at the same time.
The engines quickly restarted and the pilots managed a safe landing. But the incident, along with similar ones before it, set off alarm bells throughout the industry because of the cause: ice inside the engines.
Modern jet engines long were thought to be impervious to internal icing. But airlines, regulators and weather scientists now think otherwise, and have been scrambling to figure out how to handle the hazard. Despite some progress, the shutdowns keep happening.
On Monday, the Federal Aviation Administration will propose new safety rules that are expected to apply eventually to about 1,200 widebody jetliners world-wide, including Boeing 747 jumbo jets. Pilots of those planes will be required to turn on engine anti-ice systems more frequently during descents, to reduce the chances of sudden shutdowns and to increase the likelihood that engines that quit will restart.

Full article here:
Airline Regulators Grapple With Engine-Shutdown Peril - WSJ.com

Neofit: "Emergency power supply = Ram Air Turbine (RAT)"

Right, but RAT doesn't power NAVSAT, and the process in order to restart the engine is speed/altitude related... something not very easy when one doesn't get reliable speed indication, neither particular training at doing it without such indications, neither being in the middle of a real thunderstorm...

S~
Olivier

[takata]

Originally Posted by jcjeant

Methink it's some emergency eletrical power sources in those planes ... like batteries ..
I can be wrong .. but seem's I read it somewhere ......
I suppose they are not only for the passengers courtesy lights
Edited:
Yes indeed .. I forget the RAT (the green power )

But of course, failure of power doesn't mean that this aicraft would crash, but unability to restart the engines will certainly do the job pretty well. An A330, out of fuel, already glided about 100 NM and landed. But from this place and direction of flight, there was no other option than to try and restart the engines, and certainly to do so while turning back.

Moreover, the attitude of the aircraft at impact might be related to an attempt at restarting the engines that failed due to proximity of the sea. Water was hit unexpectedly. It looks like the aircraft was either stalled or either at the botom of a ressource and nobody seems to have been prepared for a ditching attempt.

The other point is that ice may also damage the engine. It never happened that such a case was recorded but from the number of failures already documented, it looks like all the perequesites for such an incident might have been present for this flight (even the period of the year: Spring-Summer).

S~
Olivier

[PJ2]

takata;

If power was lost, what happened to FDRs records? are they stopped also in this aircraft?

The DFDR is powered by the AC2 bus which would be lost in the Emergency Electrical Configuration, (loss of both generators). I believe the CVR is similarly powered. The recorders will have stopped operating the moment AC1, AC2 and DC1 buses were unpowered. The ATSU [Air Traffic Service Unit] contains the ACARS and CMS [Central Maintenance System] as well as other functions and has inputs from the FMS [Flight Management System], FWC [Flight Warning Computer], DMC, [Display Monitor Computer] and SDAC [System Data Aquisition Concentrator], SATCOM control and #3VHF input would also stop once these buses were unpowered.

Regarding any loss of airspeed information and disconnection of the autopilot/autothrust, the immediate key in responding is to do nothing initially except to pull the thrust levers out of the CLB detent and set the EPR/N1 where it was just prior to the failure, and strictly maintain the same attitude. This would be challenging in heavy turbulence and darkness with no horizon.

Flight in Alternate Law is just not an issue. The A330 becomes an ordinary aircraft, easily controlled providing one is, as with any heavy transport at high altitude, gentle with the controls. The sidestick does provide enormous power vice a control column/wheel however.

If the thrust levers were left in the CLB [Climb] detent with loss of the autothrust system, the aircraft would slowly, (very slowly) accelerate towards VMO but it would take a long time...10-15 minutes; there is not a lot of reserve thrust remaining at these altitudes and weights. That assumes still air of course. In heavy turbulence, strongly-rising columns within a TCu we cannot say anything that is certain.

Also, I do not believe a full turn-back was attempted though an altering of course is entirely possible. A turn-back is a very drastic course of action with many other mitigating actions available before such a decision would be indicated, (this assumes that radar was on, used and correctly interpreted). I say this because of the time involved for such a maneuver. A 180deg turn, discussed last August on the second original thread now in Tech Log, would take just under seven minutes and cover a diameter of just under 14nm, (463kts TAS, 25deg bank, still air). The max bank achievable on autopilot using the heading select is 20deg so anything steeper requires a disconnection and hand-flying. Under circumstances of heavy turbulence, darkness and a degrading autoflight system with associated warnings, that too would be a challenge.

It has been observed the ending of the ACARS messages does not necessarily mean the end of the flight so we don't know the timings.

We know that the radome was not damaged in flight because of the damage pattern it does exhibit, (lower area shattered/missing, upper area intact but damaged - I do not recall any other damage such as hail damage, being cited), so we can assume the radar was functionable though we cannot say it was on, even though that is a good assumption.

PJ2

[takata]

Hi PJ2,

Thank you for your reply.
So, if the power failed after last ACAR, the recorders would shut down also and we would be left with a big blank if the aircraft did not crashed right after that... In some way, it may be only an indication of the power failure if recorded data could be clear enough to assume that aircraft was still in control when recording stopped.

But you did not say exactly what you will do if you were left with a dual engine flameout and a controlable aircraft, assuming the same position, meteo and heading ?

S~
Olivier

[HazelNuts39]

Altitude, CAS, Mach

Originally Posted by DenisG;#866

a CAS higher than 530 kt without the Mach value exceeding 1. This condition implies that the aircraft was at an altitude comprised between about 4,000 and 14,000 ft;

DenisG;
At 14000 ft, 530 kt CAS, the Mach number is 1. Therefore the condition 'CAS higher than 530 kt without the Mach value exceeding 1' can only be true below 14000 ft (fundamental physics). I do not understand the addition of 'above 4000 ft.'
HN39