AF447

[BOAC]

TP - do you or PJ know where the assumed cruise CofG at "between 37.3 and 37.8%" sits with any limits? All the report says is 'within the limits'.

[BigFootDriver]

"the aircraft entered into a mesoscale Cb, it may have encountered moderate turbulences "

Huh??? Should read "it may have encounter extreme turbulences"

Punching holes in 55,000 foot TRW's will always produce way more than MODERATE turbulence.


Please join me in being scared of CB's above 30,000 feet.

BFD

[PJ2]

BOAC - the max aft cg that I show (and the 330 I'm pointing to may not be the same as the AF one), for 210k kg is just over 40% - fwd limit is under 17%. WRT TP's comments above, I show different slightly figures but the MTOW I'm referencing is around 230k kg and the AF 330 MTOW is 233k kg.

[BOAC]

So it looks as if it was 'aft' then. Are you aware of any 'guidance' on Cof G management on flights such as 447 had ahead of it? Weight-wise she appears to have taken off 243kg below max.

EDIT: to ask PJ if he has any observations on chosen level and speed. (I cannot find any determination of SAT deviation from ISA in the report.)

Originally Posted by BFD

Please join me in being scared of CB's above 30,000 feet.

- actually at any height. Going under a big one is not exactly fun and going through one below 30k, erm, no......

[wsherriff]

River City, a Fact.

Although not known by the "Industry", the swept wing aircraft design is subject to a pitch-up, in a strong, weather induced up-draft! (Personal experience in a Boeing 707) Management explained it away by saying I was outside the "Envelope"!!

AF 447 may have survived if it had been left to the crew's natural physical sense inputs and normal, trained, control responses! The FBW programmed control inputs, in response to the pitch-up attitude, created a complex of rapid attitude transttions beyond the "Scope"!

We must standby until the computer nerds (experts) absorb some Aerodynamics knowledge!



































































9

[phoenix leader]

Flight Data Recorder

All seems very quite at the moment, but is there any news on the search for the "Flight Data Recorder" (Black Box) and if not has anynone any information as to the latest on the search for it?

[Dysag]

phoenix leader

1) Yes, it is, quite.
2) No, not anynone.

[BOAC]

Not located as far as we know at 070720092104Z. I think most people assume the 'pinger' to be 'dead' due to the time factor and the chances of locating the tail section in the sea extremely remote. It would be a very lucky event if they found the metal.

Whether or not the search will continue...?

[takata]

Engines

@PJ2:
Thank you a lot for all those very detailed infos and for answering all my questions! You saved me many hours studying the FCOM Engine's Chapters as I was totally unfamiliar with this part of the aircraft.

I duely noticed that you had previously pointed an engine flame-out during a stall, but that stall would also imply a serious loss of altitude and much less distance could be flown after an eventual recovery. Nevertheless, 90-100 NM is fitting very well my bracket of [90-115 NM] corresponding to [0.40-0.50 m/s] drift speed from 0214? position in whatever direction (preferably to the Brazilian coast or F. de N.).

I was not suggesting that F-GZCP could have planned to ditch because it would be fairly suicidal by night and bad weather in the middle of the Atlantic, possibly without comms, and absolutely nothing is suggesting a single trace of preparation. So, I was wondering about a loss of control at low altitude during a desperate attempt to restart the engines, or a battery exhaustion...

I wondered also if an eventual TAT probes icing would not affect at some point the engine monitoring if a false and higher temperature would be delivered to the system. I have no idea of the consequences.

I'll study closely your answers and will re-read the engine icing documents. @Safetypee: thank you also for the complementary and very usefull informations.

S~
Olivier

[ClippedCub]

You mentioned aft cg as leading to a flat spin... from the report the a/c was close to the forward limit on departure.

Modern transports are designed with supercritical, or aft loaded airfoils to give thicker airfoils, (more fuel volume and reduced structural wing weight), compared to conventional airfoils. Think of these airfoils as conventional but with 5 to 10 degrees of flap deflection. Even with the long moment arms of the tail of transports, there can be up to a 20 count drag penalty to trim the nose down pitching moment produced by the aft loaded, cambered sections, at cruise. To mitigate this drag penalty, designers will trim at aft cg positions at cruise using fuel transfer. The pilots here can expand on procedure, and whether they would have been at max aft cg accounting for fuel burn.

We normally reference aerodynamic conditions at 25% MAC, so cg's approaching 40% are pretty far aft. Don't think even aerobatic airplanes are operated at 40% aft cg, but will check.

Edit: For the Yak 54, cg can approach 37%, but their forward cg limit is 29%. Must keep it aft for maneuverability. Abbreviated paragraphs,

Personnel from the Yakovlev Design Bureau provided the following information: The airplane is rated for nine positive Gs, and seven negative Gs. It has a roll rate of 340 degrees per second. Yakovlev Design Bureau personnel reported that according to their flight test data, at an aft center of gravity (CG) of 37 percent mean aerodynamic chord (MAC), the airplane exhibited neutral stability, but was easily controlled. At a forward CG of 30 percent MAC, the stick force per G is 4 kg (8.8 pounds). At an aft CG (37 percent MAC), the stick force per G is 1 kg (2.2 pounds). The airplane has an aerodynamic buffet that occurs before a stall. The altitude needed to recover from an upright stall was 820 feet, and 984 feet for an inverted stall. The airplane did not demonstrate any tendency to enter an inadvertent spin during stall testing.
Additionally, Yakovlev Design Bureau personnel reported that upright and inverted spins were steep and stable. The angle of attack during a normal spin was 40 to 50 degrees, with a recovery in 1/2 turn. The altitude loss for a 3-turn spin was reported to be 1,969 to 2,133 feet. The altitude loss for a 6-turn spin was 2,625 to 2,789 feet. The load factors during spin recovery did not normally exceed 2.5 Gs. A flat spin had an angle of attack of 60 degrees, with 1 turn for recovery. The altitude loss during a 6-turn flat spin was 2,297 to 2,461 feet.

The AOA for normal stall can be reduced by relaxing back pressure slightly, but still keeping it stalled. Speeds rotation. Notice that even with low mass at the airframe extremities, it still takes one turn to recover from a flat spin. The Yak-52 takes 2-3 turns. An A330 with distributed fuel load and passenger/cargo load would be extremely difficult, if not impossible, to recover from a flat spin even with larger controls.

[Art-Deco]

Hi again.
Some reflections from a sat-ops (not a pilot).
Some people here think that the 30 second blackout in the ACARS-txm:s is related to a violent turning on the a/c-body.
There could be a much simpler explanation, ACARS works in microwave band and as the a/c passes some very dense cb:s, these ones has so much high damping factors that the link budget to the satellite is disrubted. I have work for years with disrubted satellite signals when a huge cb occurs, the damping factor has varied from 10-25dB depending on the clouds formation.

I have taken the Tim Vasquez erlier calculations and added the timeframe of the ACARS. As there was a 31 second lost carrier it should be possible to calculate the exakt size of the cb:s worst area, I got approx 6km with around 780km/h, -but I may have got the wrong speed of the a/c here.

I have noticed that the a/c should have kept a straigth course until it hit the real bad CB center to get a 31 second loss of ACARS-txm:, after that the a/C came out of the "black-spot" and the txm:s resumed. If the a/c after a minute started to make a hard right (east) move, it would make further txm:s hard as it then was on the border to a heavy cb with a lot of damping on the microwave L-band.

Enlosed is Tims erlier work that I used to plot the real-time acars pos.
The clouds may have a litte differnt shape as the sat-images were taken some minutes erlier. If you the compare the debree-map, it looks like it spread like rings on the water with a little drift.

Enclosing BEA:s ACARS LOG again to compare with the images below:

2:10:10 - .1/WRN/WN0906010210 221002006AUTO FLT AP OFF
2:10:16 - .1/WRN/WN0906010210 226201006AUTO FLT REAC W/S DET FAULT
2:10:23 - .1/WRN/WN0906010210 279100506F/CTL ALTN LAW
2:10:29 - .1/WRN/WN0906010210 228300206FLAG ON CAPT PFD SPD LIMIT
2:10:34 #0210/+2.98-30.59
2:10:41 - .1/WRN/WN0906010210 228301206FLAG ON F/O PFD SPD LIMIT
2:10:47 - .1/WRN/WN0906010210 223002506AUTO FLT A/THR OFF
2:10:54 - .1/WRN/WN0906010210 344300506NAV TCAS FAULT
2:11:00 - .1/WRN/WN0906010210 228300106FLAG ON CAPT PFD FD
2:11:15 - .1/WRN/WN0906010210 228301106FLAG ON F/O PFD FD
2:11:21 - .1/WRN/WN0906010210 272302006F/CTL RUD TRV LIM FAULT
2:11:27 - .1/WRN/WN0906010210 279045506MAINTENANCE STATUS EFCS 2
2:11:42 - .1/WRN/WN0906010210 279045006MAINTENANCE STATUS EFCS 1
2:11:49 - .1/FLR/FR0906010210 34111506EFCS2 1,EFCS1,AFS,,,,,PROBE-PITOT 1X2 / 2X3 / 1X3 (9DA),HARD
2:11:55 - .1/FLR/FR0906010210 27933406EFCS1 X2,EFCS2X,,,,,,FCPC2 (2CE2)
/WRG:ADIRU1 BUS ADR1-2 TO FCPC2,HARD
2:12:10 - .1/WRN/WN0906010211 341200106FLAG ON CAPT PFD FPV
2:12:16 - .1/WRN/WN0906010211 341201106FLAG ON F/O PFD FPV
2:12:51 - .1/WRN/WN0906010212 341040006NAV ADR DISAGREE
2:13 - .1/FLR/FR0906010211 34220006ISIS 1,,,,,,,ISIS(22FN-10FC) SPEED OR
MACH FUNCTION,HARD
2:13:14 - .1/FLR/FR0906010211 34123406IR2 1,EFCS1X,IR1,IR3,,,,ADIRU2
(1FP2),HARD
2:13:45 - .1/WRN/WN0906010213 279002506F/CTL PRIM 1 FAULT
2:13:51 - .1/WRN/WN0906010213 279004006F/CTL SEC 1 FAULT
:14:14 - .1/WRN/WN0906010214 341036006MAINTENANCE STATUS ADR 2
2:14:20 - .1/FLR/FR0906010213 22833406AFS
1,,,,,,,FMGEC1(1CA1),INTERMITTENT
2:14:26 - .1/WRN/WN0906010214 213100206ADVISORY CABIN VERTICAL SPEED

[singpilot]

Page 18 of the prelim report references this situation. AF447 had a FCMS (Fuel Management System) that used a 'trim tank' to move CofG aft for cruise. This transfer begins in climb, and, as noted, keep target CofG within 0.5% of target MAC. 37.8% was the estimated aft CofG for this segment of the flight, and actual regulatory aft limit would have been 'behind' that.

[Will Fraser]

I will continue to await the synthesis of a straightforward analysis of the only important evidence that exists to date, and that having been actively ignored, the Forensic Pathology Report. Anyone satisfied with 'apparently well preserved' ? At some point in time, "we haven't seen the Data" must become "We refuse to look at it." There remains quite a disconnect between Brazilian reports inferring a breakup and BEA's "conclusion" of an intact a/c at impact.

[steamchicken]

Without intending to do so, I may have hyperfocused my disagreement with BEA terminology. In and of itself, the phrase en ligne de vol is reasonably benign. To take issue with it involves a very strict 'raison d'etre'. It may be a non charged description of horizontal for whomever wrote the piece. The writer isn't responsible, the group who approved the text is. As Surplus1 so eloquently points out, this report is composed by parties who have well defined and difficult to disguise interests in the ultimate understanding of the piece by all who read it. As such, it is a rehash of all the information already known, exclusion of information that would lay responsibility on an interested party, and is essentially useless.

Without substantiating a conclusion of hull integrity, they claim it by 'visual inspection'. No supporting engineering, no metallurgy, no independent or reviewed analysis. They rely solely on their authority to conclude to do so. This report is opportunistically political.

Your evidence of this appears to be: that you don't speak French very well. What would you have said about a Frenchman who made similar assertions about an FAA/NTSB report entirely because he didn't understand an unusual phrase in English?

[Will Fraser]

I would tell you that it wouldn't matter to me, both authorities have divergent missions from what is 'expected' by the public. Besides, it doesn't matter what I think, what do you think of it? My source as a non French speaking person is a phD in French linguistics, whom I trust completely. Steamchicken, if you read my quote (above) for comprehension, you will note that I have softened my initial reaction, and am giving BEA some room here, and for the reasons you state. But it isn't only en ligne de vol that is concerning. There is not a chauvinistic bone in my airframe, all are vulnerable to a healthy scepticism. Do you doubt there is an element of 'circling the wagons' here? I see it, for one.
WilyB- willdothat

[WilyB]

Will Fraser:

Please re-read page one of BEA's report.

SPECIAL FOREWORD TO ENGLISH EDITION

This interim report has been translated and published by the BEA to
make its reading easier for English-speaking people. As accurate as the
translation may be, the original text in French should be considered as

the work of reference.

This part may also be of assistance:

This document has been prepared on the basis of the initial information
gathered during the investigation, without any analysis and - given the
continuing absence of wreckage, the flight recorders, radar tracks and
direct testimony - without any description of the circumstances of the
accident. Some of the points covered may evolve with time. Nothing in
the presentation of this interim report or in the points that are raised
therein should be interpreted as an indication of the orientation or
conclusions of the investigation.

In accordance with Annex 13 to the Convention on International Civil
Aviation, with EC directive 94/56 and with the French Civil Aviation
Code, the investigation is not conducted in such a way as to apportion
blame or to assess individual or collective responsibility. The sole objective
is to draw lessons from this occurrence which may help to prevent

future accidents or incidents

[xcitation]

VHF Radio problems

(page 18 of BEA AF447 prelim report, english version)

1.6.4 Condition of the aircraft before departure

On arrival of the Paris-Rio de Janeiro flight the day before the accident, the Captain reported a problem at the level of the VHF1 selection key on RMP1. The aircraft has three RMPs:
RMP1 on the left-hand side, RMP2 on the right-hand side and RMP3 on the overhead panel.
The ground engineer had switched round RMP1 and RMP3 to allow the aircraft to leave, in compliance with the regulations (departure covered by a MEL). The departure covered by this MEL item did not have any operational consequences.

The 3 Brazil ATC requests were not replied to and the VHF radio had been repaired prior to take off. Coincidence? Cockpit non-response is hard to believe as they were continuing radio comms made seconds earlier. ATC tried 3 times. Also no further AF447 VHF comms after this point for 30 mins (breaking the off radar 10 min report in SOP?).
VHF is triple redundant however if one fails how do you know unless you do a test? Perhaps the earlier MEL was incorrectly identified as RMP1 switch but was actually an issue with VHF1?

Only a minor item but I want to ensure all the valid data points on the table.

[ VHF radio description:
Each RMP (Radio Managment Panel) is associated with a particular VHF (Very-High-Frequency) Radio. RMP1 is associated with VHF1, RMP2 is associated with VHF2, and RMP3 is associated with VHF3. All VHF Radios can be tuned by each RMP. When the VHF Radio selected by the RMP is not associated with that RMP, the SEL Light on that RMP as well as the RMP associated with the VHF Radio illuminates. For instance, if RMP2 has selected VHF1, the SEL Lights on RMP2 and RMP1 will illuminate.]

[singpilot]

xcitation;

A careful read of the radio contacts section shows the last voice comms at 0153 was on HF radio with ATLANTICO. To include a SELCAL check. Normal procedure was to take (noisy) HF off of speaker/headset at that point. A voice only recall from ATLANTICO requesting TASIL update would not have been 'heard'. There has been no explanation as to why ATLANTICO did not use SELCAL again to get a response from AF447.

But all of this last comm was on HF, so any involvement with VHF was moot at that point.

Also, the 'usual' setup was that the F/O was doing all the comms, and the swapped 'control panel' was between station one (Capt) and station 3 (Overhead panel). Station 2 (F/O) was never involved with swap.

Hope this helps.

[einhverfr]

Ice particle icing in engines and restart alt.

PJ2: My impression of the ice particle icing paper was that engines could not be restarted usually above 10k feet because of the time required for the ice blockages to clear. Note in the paper it mentions that engines that were not shut down recovered function at about the same altitude. So while I don't doubt that A330 engines could behave differently, I don't think that it necessarily follows that they would provide power at a much higher altitude.

[JD-EE]

HazelNuts39, "Does this help?"

Wow, yes. Thank you!

You mention forty degree angle of attack, too. That probably would interrupt ACARS transmissions or at least make them less likely to succeed. It also extends the search range more along the reverse path and to the West. (Takata has good points about the currents and drift that cannot be totally ignored.)

The extreme angle of attack still means 120 degrees (round numbers) of relatively safe turn in clear air in a minute. I suspect in turbulence the more conservative figures would be used. So we do end up with a small donut hole in the South through and South East portions of the region around 02:10:34's report.

The next serious question is whether or not they thought they had a reason to head backwards. I can think of many. But more experienced pilots would do better than my guesses and imagination. It MIGHT be simpler to presume they decided to go around the storm to the left at the last moment and banked sharply for as much as a 90 degree turn. That puts them 5 miles West at the end of the turn as well as 5 miles further North (sort of rounded upwards and kept at convenient numbers for a SWAG.) Then there are (at least) three more minutes to account for. Plotting those kinds of estimates along top of takata's data will probably help form better estimates. Since takata has the originals I hope this plotting is added and reposted. (chuckle - not that I'm about to swim out there and try to find the plane myself or participate in the effort. It just might help figure out what the pilots are likely to have done during the period they may have maintained control. There might be a dollop of education that can come from it. At least it might start to dispel what begins to sound like conspiracy theories by declaring some scenarios are flat impossible.)

JD-EE