AF447

[Flyinheavy]

Brazilian Air Force informs to have found first victim about 70 km from the last position that has been reported by the a/c. Whatever this last posn means.

More specifically they are talking 'where the last communication from the aircraft was received'.

[aguadalte]

Quote:
Sorry to disagree. Yes it will move. That is why you will get a "Use Rudder With Care" message in Status Page.

I think he means the LIMITER won't move (which means the rudder has full travel).

Max

Thanks Max, Sorry Swedish,

On the A330, if you get a double rudder limiter fault, the TLU may be left positioned in such way that it will not prevent the pilot from using more rudder than the structure can accept. (Full rudder can only be used below 160Kts).

[Tail Chase]

AF447 SAR effort

The Brazilian AF issued a number of releases earlier this afternoon, indicating that two bodies had been located and retrieved, along with a personal effects and (apparently) small aircraft components. The sequence of events this Saturday is as follows:
1.) 0300LT/06062009 Brazilian AF R-99B 6751 departed from SBFN to scour the debris field originally plotted early last Tuesday morning (02JUN2009), in which a passenger seat and a 7-meter long component were later visually sighted by one of the C-130Hs.
2.) 0814LT/06062009 Two bodies were visually plotted inside the search area
3.) 0930LT/06062009 The first body (male) was retrieved
4.) 1113LT/06062009 The second body (male) was retrieved
5.) 1200LT/06062009 A knapsack was retrieved, along with other smaller personal items

Cheers

[Pontius Navigator]

Originally Posted by SLFinAZ

I'm curious about the use of the term "bodies" vs human remains.

We must be careful attibuting English elegant variation to translation from foreign sources.

~~~~

From the position of the bodies it will be possible to backplot using wind and current data to determine a point of entry into the water. A US expert used the term Hindcasting (sic?).

[Tail Chase]

AF 447 SAR Effort

Further to my previous post, Brazilian Navy corvette Caboclo is currently carrying out retrieval duties. After collecting the two previously mentioned bodies, the ship´s crew retrieved passenger oxygen masks, a blue-upholstered passenger seat (with serial number) and a large number of personal effects.

The two bodies were found some 900 km from SBFN

Cv Caboclo will remain on site to search and collect whatever it finds until late this afternoon and Sunday morning, when it will transfer all that it collected to the frigate Constituiçăo. The frigate will then sail to a point 250 km E of SBFN, when it will launch its Super Lynx helicopter to SBFN with bodies and whatever material was collected by the Caboclo. Bodies and material will then undergo preliminary identification and will subsequently be flown to SBNT.

Cheers

[Grunf]

AF not replacing Pitot tubes

If this is true it is weird - AD is COMPULSORY within the time compliance requirement (which was 2003).

So not complying with an AD is a really weird thing, if true.

Cheers

[Squawk_ident]

Bodies recovered

French media are announcing the same thing.
A seat, a bag with a laptop inside and an AFR ticket. Two male bodies have been recovered and are on board a Brazilian vessel.
A German military satellite would have located an aera with possible debris of the plane.
A French submarine is on its way and will try to locate a signal from the bottom of the ocean. Unfortunately it will take a while before it reaches the area.
Regards

[rer47]

Earlier mm43 calculated aircraft speeds and positions based on an assumed speed at INTOL of 540kts.
The references I have seen give the speed at INTOL as 453kt (see reference 15 in the Wikipedia article).
Assuming this ground speed, the results change considerably:

INTOL 0133Z
453kts
EPODE 0157Z
453kts
ORARO 0205Z
576kts
FINAL 0214Z

Assuming FL350 and an OAT of -43C (from Vasquez (ref 25), but ignoring his estimated 10kt headwind), 453kts corresponds approximately to M 0.78, evidently the recommended speed for turbulence. 576kts corresponds to M 0.98, which is impossible. Therefore they must have increased speed long before ORARO. This would be consistent with overspeed and destruction of the aircraft due to turbulence.

As reported by others, two bodies have been reported found about 900 km (revised to 800 km - BBC) NE of Fernando de Norohna (refs 53, 54). The final ACARS location is 858 km NE of that island.

rer47

addendum: A Brazillian news paper reports "470 miles - about 900 km from Fernando de Norohna" and "69 km from the final location communicated". This would be nm, and the last location would be the 0214Z message (FINAL).

Aeronáutica confirma descoberta de destroços e corpos - Estadao.com.br

[philpop]

New elements

More details of the ACARS messages have become available on June 5th and suggest following events while the airplane was in cruise (note, there is no message regarding electrics, hydraulics or engine problems):

02:10Z: Autothrust off
Autopilot off
FBW alternate law
Rudder Travel Limiter Fault
TCAS fault due to antenna fault
Flight Envelope Computation warning
All pitot static ports lost
02:11Z: Failure of all three ADIRUs
Failure of gyros of ISIS (attitude information lost)
02:12Z: ADIRUs Air Data disagree
02:13Z: Flight Management, Guidance and Envelope Computer fault
PRIM 1 fault
SEC 1 fault
02:14Z: Cabin Pressure Controller fault (cabin vertical speed)


There have been at least two similiar incidents preceding AF-447 (dates of both flights are unknown):

First incident: An Air France Airbus A340-300, registration F-GLZL performing flight AF-279 from Tokyo Narita (Japan) to Paris Charles de Gaulle (France), was enroute at FL310, when the airplane went through a line of thunderstorms. The captain's air speed indication suddenly dropped to 140 knots, the systems issued an alert regarding disagreeing speeds (NAV IAS DISCREPANCY), the navigation display showed a tail wind component of 250 knots. The captain released control of the airplane to the first officer and tried to switch his display from ADIRU1 to ADIRU3. 2 minutes later autopilot and autothrust disconnected and the fly by wire changed into alternate law. The crew noticed icing conditions (static air temperature [SAT] -29 degrees Centigrade) and switched anti ice including pitot heating systems from automatic to on. The speed indications became normal again and agreed again, the autoflight systems were reengaged and ATC informed of severe icing. ATC reported, that two flights had just passed the location without problems. When the crew attempted to reset and reengage ADIRU 1 two times, the system again brought the message "NAV IAS DISCREPANCY" on both attempts, although the speed data appeared consistent. The crew suspected polluted pitot tubes.

Maintenance found, that the drainage holes of all three pitot tubes had been clogged, rendering it very likely that weather combined with the clogged drainage holes caused the incident. Maintenance had reported more clogged drainage holes on A330 and A340 aircraft in the past to Airbus Industries. Airbus Industries was aware of the problems, changes had already been introduced to the pitot tubes on the A320 family, where similiar problems had occured. A modification of the A330/A340 pitot tubes was already planned by AI.

Second incident: An Air France Airbus A340-300, registration F-GLZN performing a flight from Paris Charles de Gaulle (France) to New York JFK,NY (USA), encountered brief turbulence while enroute. The autoflight systems dropped offline, "NAV IAS DISCREPANCY", "NAV PRED W/S DET FAULT" and stall alerts were repeatedly issued during the following two minutes. The airplane continued to JFK without further incident. A review of the policy of retrofitting pitot tubes was recommended and authorities informed.

[AMF]

ByteJockey

http://www.faa.gov/other_visit/aviat...overy_Book.pdf

ByteJockey,

I have no idea why your post would be deleted. That FAA link should be required reading for anyone interested in thinking about possible causes for this accident and every driver operating at high altitude. For the uninitiated, it's filled with useful facts and info like this relating to high-altitude flying....

"In another high altitude situation, the crew decided
to use heading select mode to avoid weather while
experiencing turbulence. The steep bank angle that
resulted from this mode quickly caused slow speed
buffeting. The crew’s rapid inappropriate response
to disconnect the autopilot and over-control the
airplane into a rapid descent in poor weather exacerbated
the situation. These real world examples
provide evidence towards the need for more detailed
training in high altitude operations."

While it is all pertinent, of particular note should be areas of interest that could be directly related to the conditions the AF was operating at the time and possible factors; High Altitude Aerodynamics, Thrust Limited Condition and Recovery, Maneuvering Stability, Effects of Wx that could cause Stall or Slowdown at High Altitude, Use of Anti-Icing on Performance, Inflight Icing Stall Margins, Flight Techniques of Jet Aircraft, Human Factors and High Altitude Upsets, Multi-Engine Flame-out.

Understanding those basics of high altitude flight, and transposing what is known or probable about AF at 35,000' (which was still relatively heavy operating near the top of it's weight-defined envelope), that conditions were turbulent and possibly severe, that the A/P disengaged itself and the crew were hand-flying (without numerous protections), that there were airspeed anomalies or miscompares whether they be vertical gust, icing, turbulence or a combination of all-induced within or (in the case of vertical gusts and turbulence) in the clear above quickly developing convective activity, and the possibility that the aircraft entered an area of rapidly increasing temperatures aloft as it traveled laterally through uplifted warmer moist air and it's evident that there is no reason an aircraft need be broken, malfunctioning, or giving erroneous information to the crew to find itself in a perilous and rapidly deteriorating aerodynamic (and possibly untenable) situation if recovery is thrust limited and the attempted recovery executed in severe conditions of turbulence, icing, vertical gusts, and possible engine problems associated with loss of aerodynamic flow/control.

It doesn't matter if it's an Airbus, Boeing, or the latest whiz-bang corporate jet...no aircraft escapes these Laws of Aerodynamics or magically isn't affected by what Mother Nature can hand it at those altitudes just because every system is working.

No aircraft is designed to fly at or near the upper limit of it's envelope without strict adherance to load-factor/bank, and AOA limits, and the difference of being near the edge of that envelope as opposed to even 5,000' lower in terms of riding through turbulence/limits on maneuvering inputs to observe them, and having excess thrust to overcome rising-temp or turbulence/vertical gust-induced airspeed degradations/fluxuations are night and day.

If the previously posted QRH numbers are anywhere close...that for that altitude and approximated weight, and assuming the crew slowed the aircraft to below its Maximum Turbulent Air Penetration Speed when the pilot messaged 10 minutes prior to the ACARS message flurry, the margin for control between overspeeding and min clean speed was merely 15 knots at that altitude and weight, and only 25 knots between overspeed and min speed/stall buffet. And within that small airspeed window load factor/bank limits must be observed or the low-speed buffet boundary rises rapidly.

If the bank angle gets too excessive (whether turbulence or pilot-induced, and at high altitude "too excessive" is simply a standard bank at low altitude) it's possible for the low-speed buffet boundary to rise (and it's rise corresponds to the rate of roll...i.e. very quickly in turbulence) the point it crosses over the aircraft's current speed which will produce an immediate stall. If the bank angle continues this low-margin buffet can cross-over past it's Max Turb Speed and over the aircraft's Mmo/overspeed and the "corner" available for flight doesn't exist. At high altitude the pilot's ability to counteract turbulence-induced roll to stay within load-factor AOA limits is seriously hampered by sluggish response to control inputs. It's entirely possible that vertical gusts acting unevenly on the aircraft or roll axis occillations associated with varying turbulence amplitude can't be counteracted quicky enough by pilot inputs, and in severe turbulence, guaranteed, because the aircraft experiencing at least momentary loss of control is part of what defines severe turbulence.

Momentarily losing control of the aircraft in severe turbulence at lower altitudes means that although for a moment the pilots couldn't counteract the environmental, outside force acting on the aircraft the aerodynamics remain far more sound, and once the outside force abates (if the aircraft isn't structurally damaged) control can be quickly regained with the help of authoratative control response and excess engine thrust availability. Since at low level the performance envelope between high and low speed buffet boundaries is far wider and bank angle/load factor limitations far less restrictive, the aircraft may never go outside the associated larger operating envelope.

This is not the case at high altitude where the aircraft is operating near to the top edge of it's envelope where speed buffet boundaries and roll limits small even in smooth air. Due to these thin margins, turbulence or vertical gust forces and fast temperature changes can quickly place the aircraft outside any or all of these limits (possibly at the same time) but when it abates it leaves the aircraft well outside it's envelope and fully in the realm of unsound aerodynamics where flight control inputs are degraded, and overcontrolling or wrong input can quickly exacerbate the problem. To make matters worse, engine thrust available to help regain the envelope is minimal or non-existant if a flame-out occurs.

For the non-pilots who have an interest in this;

It's hard to explain to the uninitiated just how differently an aircraft flies when near the top of it's operating envelope compared to just a few thousand feet lower, and what these degradations and thin margins mean. If you're near the top end of that envelope, even something as little as a unforecast, sudden rise in temperatures aloft of 5 to 10 degrees can require immediate action (descent to a lower altitude because engine thrust isn't available in excess at altitude) because the top of the envelope is lowering...crashing down if the temp rise is quick enough.

The same goes for encountering even moderate turbulence with regards to bank/load factor limits. Turbulence at altitude is not simply a matter of comfort or breaking the airplane like it is at lower altitudes...its a matter of the aircrafts ability to remain flying as the low-speed stall margin rises up to approach your current airspeed every time the rolling action is induced. The greater/quicker the bank, the higher/quicker the rise. This is why at high altitude autopilots automatically limit roll to about 1/2 the bank (appx 15 degrees) of what it normally rolls the airplane to at lower altitude. This roll-limiting feature for most autopilots usually becomes active automatically when climbing through 30,000 to 33,000 feet and the aircraft enters the higher-altitude realm of small-margin aerodynamics.

Coupled with this of course is the whole necessity of coordination with using rudder and it's limits, dutch rolling and yaw tendencies at altitude for every type of aircraft, and how ocillations can develop during turbulence with receding and leading wing local velocities that (due to small buffet boundary margins) can actually approach low-and-high buffet at the same time on difference sides of the aircraft if they increase in amplitude.



When hand-flying, a pilot must still observe the critical limits even while counteracting turbulence/fluxuations that are trying to force (or has already forced) the aircraft past them. It's my understanding that with the A/P disengaging the way it did on the AF flight as indicated by ACARS message, even as a Fly By Wire aircraft that normally has overbanking protections built-in while being hand-flown, while operating in Alternate Law (as it was this case as indicated by another ACARS message) this particular overbanking protection (among others) is removed, and it will essentially fly and maneuver like every other aircraft.

At high altitude, heavy, and near the top of its evelope, that means "poorly". Basically, high altitude flying is potentially one of the biggest can of worms in aviation.....high altitude flying in severe weather quickly opens it, and nothing on the aircraft need be broken or malfunctioning for it to be perilous.

And if aerodynamic control is is lost and can't be regained, Air data readings...which are designed to read accurately when the aircraft is flying soundly, not in the middle of lost aerodynamics situations or possibly wildly fluxuating outside conditions as found in extreme weather/updraft/downdrafts.....will predictably be inaccurate, fluxuating, and conflicting/miscompare between systems which could easily produce some of the subsequent ACARS messages that people are assuming mean either component source failures or sensing malfunctions.

[innot]

what I find unreal is the temps to go from -46 to -18c in fell swoop

It is not as unreal as it sounds. Last year on an A340, while flying through what was probably the remains of previous days CBs, we had some spectacular St. Elmos fire. The whole airframe was glowing, lightning coming out of the radom etc. Also the TAT increased dramatically, from -30 to almost 0° within approx. 5 minutes. I was sure this was caused by the electric discharges heating up the TAT probe, but nevertheless the upper and lower limits on the the speedtape came uncomfortably close.

As we did not want to find out what would happen if the apparent temperature pushed as out of the envelope (overspeed and stall protections at the same time?), we were about to start an immediate descent when we suddenly left the area and the TAT fell back to -30° within seconds.

There was nothing on the wx-radar and the turbulences were only light to moderate.

I would not completely rule out something like this happened to AF447, but I think it is very unlikely and other theories seem more plausible.

Just my 2 cents

Thomas

[Junkers388L]

TCAS fault due to antenna fault

Could anybody confirm that this is really the correct interpretation of the error code 3443005 given in the ACARS message?

Since the Air France ACARS printout has this as "WRN", I would believe this is a consecutive error resulting from the 341115 Failure ("FLR") error code reported also at 02:10Z, right?

Potential explanation of a consecutive error could be the TCAS dependence on pressure altitude. With respect to the ACARS message time stamps, I think we need to compare them to the logical interdependence of the systems to give more substance to the analysis.

Would be great if Greenspinner et al. could shed some light on this, since I have no access to A330 material at this level of detail....

[S.F.L.Y]

14 years old issues

OPERATORS HAVE REPORTED AIRSPEED DISCREPANCIES WHILE FLYING UNDER HEAVY PRECIPITATIONS OR FREEZING CONDITIONS, WHICH SOMETIMES LED TO THE A/THR AND THE AP DISCONNECTION, THE E/W "F/CTL ADR DISAGREE" AND "F/CTL ALTN LAW"........

STRONG CUMULO-NIMBUS CONTAINING A HIGH DENSITY OF ICE CRYSTALS CAN BEE ENCOUNTERED, PARTICULARLY IN THE INTERTROPICAL CONVERGENCE ZONE (ITCZ)
IN SUCH AND ICY AND TURBULENT ATMOSPHERE, THE A/C AIR DATA PARAMETERS (PRESSURE DEPENDANT) MAY BE SEVERELY DEGRADED, EVEN THOUGH THE PROBE HEATERS WORK PROPERLY.
IT HAS APPEARED THAT THE CHARACTERISTICS OF SUCH AN ENVIRONMENT COULD EXCEED THE WEATHER SPECIFICATIONS FOR WHICH THE PITOT PROBES ARE CURRENTLY CERTIFIED.

[Safety Concerns]

Thats why it is strange that they are talking about known issues. These issues were dealt with and a new probe had to be in place by 2003. It was a mandatory sb.

I think the BEA and Air France need to clarify what they are saying.

Also a fleet replacement programme has ETOPS consequences so I am sure Air France are staggering the replacement and not replacing all 3 at once.

I ask this because there is still an open issue of blocked drain holes on the probes which could lead to serious consequences particularly in stormy weather.

[S.F.L.Y]

Maybe that the 2003 new probes were still inefficient. Now the questions are:

• Why airbus, civil aviation authorities and operators decided to allow operations knowing these issues (especially the latests in 2008)?
• How many crews were informed of these issues?
• How many aircrafts with these probes are still flying?
• Will these aircrafts keep flying tomorrow and with which safety guaranties?

[sequ]

Wasn´t this 330 manufactured in 2005?

Whatever was required by this SB should have been complied with during manufacture.

I think this is a moot point.

Saludos,

SEQU

[barrymung]

02:10Z: Autothrust off
Autopilot off
FBW alternate law
Rudder Travel Limiter Fault
TCAS fault due to antenna fault
Flight Envelope Computation warning
All pitot static ports lost
02:11Z: Failure of all three ADIRUs
Failure of gyros of ISIS (attitude information lost)
02:12Z: ADIRUs Air Data disagree
02:13Z: Flight Management, Guidance and Envelope Computer fault
PRIM 1 fault
SEC 1 fault
02:14Z: Cabin Pressure Controller fault (cabin vertical speed)"

Whilst I can see that blocked pitot tubes could cause all manner of spurious error messages, I fail to believe it'd cause things like "TCAS fault due to antenna fault" or even "failure of gyros".

[Swedish Steve]

Barrymong, your quote is someones interpretation of what happened.
It is not accurate.
What is a pitotstatic port? A330 has pitot heads, and static ports. They are metres apart.
ISIS doesn't have gyros, it has a solid state sensor.
ADIRUS did not fail, the AD part of one disagredd with the others. There is no evidence of the IRU part failing.

Try a different source before you jump to the wrong conclusions (or better still read the whole thread)

[Safety Concerns]

2 from 3 pitot probes supplying erronous data would account for 23 of the 24 messages. The only message that it wouldn't cover is the IR2.

[PJ2]

SEQU;

Wasn´t this 330 manufactured in 2005?

Whatever was required by this SB should have been complied with during manufacture.

I think this is a moot point.

Fully concur.

The AD was issued to be effective in November of 2002. The AD applied specifically to -300 series A330's and not the A330-200. Here are the applicable paragraphs:

AD/A330/1 Pitot Probes 12/2002

Applicability: A330-301, -321, -322, -341 and -342 series aeroplanes, all serial numbers without either Airbus Industrie modification 44836 or 45638 embodied during production or Airbus Industrie Service Bulletins (SB) A330-34-3038 or A330-34-3071 embodied whilst in service.

Requirement: Remove Rosemount pitot probes part number (P/N) 0851GR and replace them with either BFGoodrich Aerospace P/N 0851HL probes in accordance with SB A330-34-3038, or by Sextant P/N C16195AA probes in accordance with SB A330-34 3071.

Note: DGAC AD 2001-354(B) refers.

Compliance: Before 31 December 2003.

This Airworthiness Directive becomes effective on 28 November 2002.

You are correct in your observation, that this AD would have been incorporated in the manufacture of the aircraft in April of 2005.

There may be other factors which are exhibiting issues related to pitot tubes and information becoming available, but focussing on the notion that the accident sequence was "initiated" by AF not changing pitot tubes is incorrect.