AF447

[Lemurian]

Ground Brick

It is possible, that SAT antenna was smcelectronic CMA-2102 SATCOM High Gain Antenna System.

It was.

[NARVAL]

Rudder Travel Limit Fault

Just to give a little information on the subject (A330) of the rudder :

Rudder : Yaw damping in turbulence after failure :in case of a total loss of electrical power there remains a backup yaw damper unit completely independant (it has its own gyrometer and the electrical supply is provided by generators driven by two of the hydraulic systems.
Rudder deflection : controlled by 2 limiter channels and two SECs. If both calculators fail, the max rudder deflection available remains what it was before the failure ; that is 4 degrees in cruise. The max deflection will become available again when slats are extended.

[Professorrah]

GPS antenna

No build in GPS, electronically steerable, phased-array antenna, 12-17 dB gain

for what it is worth, this antenna is 12-17dBi, not dBd, so it is not really a max of 17dB of gain - only against a theoretical radiator which of course does not exist, so the gain will be 2-3 dB less than the quoted gain.

[Paul2412]

Firstly, as this is my first post let me point out that I am NOT an airline pilot (although I wish I was). I do have a great interest in aviation but my experience is limited to Microsoft Flight Simulator...

I have a couple of questions:

1. If Airbus knew that the pitot-static tubes could become unreliable in a storm at high altitude (presumably, when you need them most) why did they not make the changes mandatory immediately? Thinking about the Aero Peru flight, although it was due to human error the consequences were disastrous. Even worse, if Air France knew about this why didn't they immediately replace the system? Do airlines like to gamble with passengers lives to save a little time and money? If it's proved that the pitot-static system was to blame, I can see very tough questions ahead for both Airbus and Air France.

2. I'm in IT myself, and find it increasingly hard to believe that a system that transmits data from the FDR back to the airlines HQ in real time or near real time cannot be developed. If the plane can send out warnings, why cant the critical data be sent to the ground? If this was in place, locating a tiny box in the middle of the second largest ocean in the world wouldn't be necessary.

I apologise if my lack of knowledge of real world flying has led to a ridiculous post, but I'm following this story very carefully and answers to my 2 points would be very gratefully received.

Regards,

Paul

[teropa]

Nobody has commented yet my suggestion of the encounter with very warm air.

Can somebody who knows how to interpret Tim's excellent met analysis well enough tell what the SAT would have been, had the airplane encountered a patch of very warm air inside the MCS. I am particularly interested in the idea that IF the A330 was not equipped with the Backup Speed Scale (AoA-based), that how difficult would it have been to control the airplane if at the same time warm air was present in a way that the aircraft effectively climbed several thousand feet (increasing stall speed in the process) ?

rgds,
Tero

[DC-ATE]

RWA -

"The crew [different flight, different time] 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........."

'From automatic to on.' I'd heard that the A330/A340 have automatic pitot heating, don't know if all airliners do nowadays? But if there's an automated system controlling the pitot heaters, could it be possible that the whole cascade of 'computer malfunctions' was caused by.......a computer malfunction........?

Someone please tell me WHY, if true, there is an "Automatic" position on the Pitot Heat Switch on this, or any aircraft.

[stadedelafougere]

What's the point for an airline to receive in real time the data concerning engine thrust, rudder movement and so one. This is of no use to the company which expects its aircraft to land safely on every journey.
Yes, when crashes like that happen, we would need this data, but it does not justify implementing on every aircraft live DFDR data transmission. Besides, many A/C do not operate ETOPS flight and remain isolated from the rest of the world for substantial period of time.
Finally, Airliners would have to modify their installations to cope with all these data.

[747guru]

Paul2412

Both your points are very valid indeed!

I would love to be a "fly on the wall" listening in to the conversations between the lawyers of AF and Airbus at the moment.......talk about a blame game!

I know little about IT, but as you say, there MUST be a way that modern aircraft can transmit "burst" transmissions on a regular basis (say every 20min) to home base including ALL the info that would normally be stored on the FDR/CVR?

OK a lot can happen in the 20min between transmissions, but surely a pattern may emerge leading up to a serious incident?

[merlinxx]

Please Mods

Airbus folks only SVP, crew, engineers etc., maybe those with similar systems.

[kennedy]

Paul 2412,

In response to your query, it's called tombstone safety.

Tombstone satefy means that although the manufacturers, airlines and regulatory agencies know it is unsafe, (usually because us pilots have told them) it costs to much money to change that system, until there is a fatal accident that forces the change!

Hence, examples like the continued use of HF in remote areas( while the pax in the back enjoy Satcom phones and internet inflight), last time I came out of Sao Paulo into the Atlantic, we did not have HF contact with Atlantico radio for over 2 hours ( they didn't have their CPDLC turned on that night!) , and all comms were Acars thou our operations for relay to them, including wx deviations.

Cost is now the main driver of safety in the modern aviation world, and will only get worse in the future!

[DC-ATE]

Don't know who wrote this: -

We tried this in the sim:
FL 350 M .81 normal law,
Quickly apply full rudder, the a/c rapidly banks to 50-60 degrees bank, with only 4 degrees of rudder travel available.
Conversely, an engine failure in cruise is counteracted with the lightest touch of rudder, maybe 1 inch and can barely be seen on the flt controls page.
Respect the rudder, it is a very powerful control.

I'll ask this again: WHY are you messing with the rudder in cruise with or without turbulence? [Transport pilots only, please] Or, is this an Airbus technique? While it is 'proper' to try and keep the wings level, it is not necessary to keep the heading constant. Trying to maintain heading can lead to serious overloads.

[parabellum]

On the Boeing's that I have flown the only time you touch the rudder at cruise speeds is during the recovery from unusual attitudes, top rudder to compensate for extreme bank angle, other wise, once past, at the latest, 210kts, leave the rudder alone. It would appear that somewhere in the transition from general aviation to jets the lessons about the rudder, it's uses etc. and how to apply it are being forgotten.

[Paul2412]

stadedelafougere:

I understand that a large amount of data would be worthless, and as a result when the aircraft parks up at the gate successfully a simple signal to identify the end of the flight is sent and the airline can choose to delete the data while possibly keeping anything flagged as unusual so the maintenance guys can take a look.

Also, if the data was kept it would be easier to diagnose the start of a problem that may eventually become serious enough to cause concern. For example, a problem develops in flight for some reason, that information is now logged with the airline. If 7 days later a flight is forced to make an emergency landing, or worse still is lost the investigators can see clearly that a problem occurred at a particular time. As I understand it, the FDR overwrites itself every 30 minutes so at the moment, that particular fault would never be detected.

The argument regarding data size and cost is not relevant, I work in the IT industry and can assure you that storing TB's of information does not cost a great deal at all.

Basically, the point is that if the technology already exists for the airplane to send messages to the ground, why not take all messages that the FDR flags as unusual straight to the ground regardless of whether it caused a problem or not? Prevention is better than cure.

In response to the tombstone mentality, I understand that its a risk vs cost industry as is every industry. However, I run an IT business hosting many clients valuable data. If I receive an advisory that may cost me in the short term but will protect me in the long term I'll take it and implement it straight away as I know a total failure of my system could cost me dear. You would think that when your talking human life that could be lost it would be the same!

[TP-51]

I know little about IT, but as you say, there MUST be a way that modern aircraft can transmit "burst" transmissions on a regular basis (say every 20min) to home base including ALL the info that would normally be stored on the FDR/CVR?

OK a lot can happen in the 20min between transmissions, but surely a pattern may emerge leading up to a serious incident?

Great idea, but as long at the Pilot Unions are worried about airlines reviewing this data and then taking action against pilots after flights - this will never happen.

Perhaps all data could be sent to a '3-party' organisation such as IATA for storage for a period of say 24 hours or 7 days then automatically destroyed.

[lgw-morph]

Lifted straight off of Bloomberg:

Air France Probe Suggests Plane Broke Up in Air, Estado Says





By Laura Price
June 12 (Bloomberg) -- The Air France plane that crashed June 1 may have partly broken up in the air before hitting the Atlantic Ocean, O Estado de S. Paulo reported, citing investigators it didn’t identify.
Most of the 16 bodies examined in preliminary stages of the probe into the flight from Rio de Janeiro to Paris were found naked or with minimal clothing, suggesting the wind may have removed the garments, the newspaper said. The possibility of an explosion or fire in the jet is also unlikely because the bodies showed no sign of burns, Estado said.
Almost all of the bodies had multiple fractures, the paper reported. Investigators haven’t found water in the victims’ lungs, which would indicate drowning, Estado said. Bodies were found 85 kilometers (53 miles) apart, which may also indicate the Airbus A330-200 broke up before reaching the ocean, Estado reported.
Representatives from Brazil’s legal medical institute, which is conducting the body examinations in the northeastern city of Recife, weren’t immediately reachable when Bloomberg News called for comment before regular working hours.
To contact the reporter on this story: Laura Price in London at [email protected]
Last Updated: June 12, 2009 07:09 EDT

[BOAC]

Bodies were found 85 kilometers (53 miles) apart, which may also indicate the Airbus A330-200 broke up before reaching the ocean,

- to me it does not 'indicate' this at all. I keep on seeing this - can someone explain how an aircraft, supposedly 'breaking up' badly enough to spill passengers, can travel on for 85k? We are looking at around 6-7 minutes in a straight line, and I would suggest we should assume the ocean has spread this debris. Case remains open for me.

[Towhee]

Thanks for the diagram of antenna locations, Saigon Lost. It gives evidence that the vertical fin was intact throughout the period of the ACARS reports, else there would have been HF Fail reports, too. The VOR antennas are not monitored, but the HF antenna couplers, in the leading edge of the vertical fin, are active LRU, Line Replaceable Units, that are powered, and provide fault reporting to the transceivers, which would, in turn, report to the ACMS, and from there to ACARS.

Awesome post. A perfect example of why this board rocks! You just crossed the early breakup hypothesis off the list

only assuming ACARS reports are complete

[positive climb gear]

Rudder Travel Limit

Apologies,I am an A320 driver but was wondering if any A330 pilots might comment on this:
FCOM 1 states
For the A320 the deflection of the rudder and the pedals is limited as a function of speed. Each channel of the limiter is controlled and monitored by its associated FAC. If both FACs fail, maximum deflection is available when slats are extended.
Hence, up to 160 Kts CAS, the maximum rudder deflection of 25 degress is available. This then reduces (non linearly) to a maximum deflection of 3.4 degrees for speeds greater than 380 Kts.
One rudder surface controls Yaw. The yaw damping and turn coordination functions are automatic. The ELACs compute yaw orders for coordinating turns and damping yaw oscillations, and transmit them to the FACs.

As it seems on general concensus that there was a problem with the probes leading to erroneous speed indications, what if the FACs allowed a greater rudder deflection during their normal dutch roll damping manoeuvre, since they sensed a much lower erroneous speed?

If the speed then did increase (possibly as quickly as it dropped), the FACs would recognise that the rudder deflection was greater than allowed, would the computers assume a FAC failure? Of course the secondary effects of rudder are roll, thus making handling even more difficult.

Hence I hope the previous response for transport pilots to comment on their use of rudder- in the A320 the rudder is being operated by the FACs without your knowledge- as there is no feedback on the pedals!

Is it a similar design on the A330.....?

[Gergely Varju]

ILS27LEFT: I tried to avoid unncessary posting there, since the thread is a bit chaotic as is, and if admins decided to keep it as thread and not as a temporary subforum where theories can be discussed in an organized way (and merged into a thread at later point, and moved back to this part of forum), I think it is easy to reduce their workload by posting less. Yet, your post makes me wonder, if it is a good decision or not.

First of all, you ignore some key thing. GPS wasn't lost, since the a/c knew its own coordinates.
GPS isn't useful to calculate airspeed, since you don't know about wings, so GPS signal has nothing to do with airspeed data.
(Even trying to track something that moves with the air would be better indicator for speed... but sadly it would be pretty hard to say the least)
Some people on a hungarian forum suggested, that if you have attitude data and EPR you can calculate the speed, but it would again ignore some of the effects of weather...
Some people suggested metal plates and various devices to determine airspeed, such devices aren't tested or installed, so it doesn't work.

While it is theoretically possible for a computer to calculate some estimated airspeed from various data, and try to help the pilot to keep the a/c in safe limits, it would require some pretty complex computer software, which isn't tested or installed yet. Worst of all: It isn't even implemented in a way that would let us use it on an Airbus FBW aircraft. And measuring if you are withing safe limits can be easier anyway.

The serious faults you speak about are consequences of each other. You also ignore what kind of "serious faults" we are speaking about. If you count the alternate law message, the autopilot and autothrottle messages, etc. as serious faults, not simple consequences of pitot icing, then it could make sense.

But you also gladly ignore the fact, that weather, etc. can be considered common, yet many flights cross the atlantic safely. Why? We can say, pilots try to avoid Thunderstorms. We can say, that the aircraft are certified and should be safe on such trips. We can also say there are many options to avoid such events.

About Coffin corner aviation: We can speak about Coffin Corner aviation for many pages, and about how various systems on the aircraft can determine what speed is safe, and how various factors like any "obstacle" under the aircraft (terrain, CB, traffic) can influence what can be safe, but it isn't that simple as you want to portray it.

[matkat]

Paul2412, It is not for the manufacturer ie Airbus to mandate a service bulletin and in fact they cannot in any case it is for the state of design to do this so only EASA would have been able to mandate a change from a SB to an AD. In earlier posts I have seen the term "mandatory service bulletin" used though manufactures often use this term only the state of design can change this from an optional SB to a mandatory airworthiness directive, so in summary from a legal standpoint Airbus did issue an SB but EASA never mandated it so I cannot see why there would be an issue as in reality AF really did not have to change anything as all service bulletins are optional until changed to an AD.