News French plane lost over Atlantic

[Urwumpe]

Also, if I am correct, an airplane should be fine in case of lightning, since electricity should pass across the hull and the interior should be protected.

Faradays cage only applies to metal hulls. The many composite parts on aircraft are not automatically lightning proof.

Also, large amounts of electricity passing through a small bit of aluminum will also cause damage.

 

[Mustard]

The debris are localised at this area there are 1 or 2 hours ago.
Depth=4000 to 5000 meters (13 000 to 14000 feet)

 

[garyw]

Also, if I am correct, an airplane should be fine in case of lightning, since electricity should pass across the hull and the interior should be protected.

Not necessarily. There are actually two forms of lighting - One of which, yes, shouldn't be a major issue. The other type is MUCH more powerful and will melt a plane. http://en.wikipedia.org/wiki/Lightning#Positive_lightning

If such is the case, why airplanes fly across an unsafe path that may involve lightnings?

Weather radar will show where lightning is so that's a very good question.

 

[Moonwalker]

Weather Radar has certain limits and errors and is not a 100% safe guide through the sky alone. It is useful only in combination with an accurate meteorological briefing before flight plus experience and visual reference during flight, and of course proper usage. It can be easily possible to use weather radar improper and misinterpret the reflections. But it can also easily be faulty when you don't need faults right at a certain moment...

Anyway, I think it is likely that the area of cumulonimbus was just too big to scud running. The crew either had to return or in the worst case to fly through, which actually is not allowed but sometimes there are cases in which crews have to do difficult choices especially on long haul flights. To me it seems that it was an underestimation of the meteorological conditions enroute. At least the significant weather charts during briefing should have shown potential hazards. Once you got there your weather radar does not really help, it just shows the situation you are facing right in the middle of the Atlantic Ocean.

 

[Dambuster]

The debris are localised at this area there are 1 or 2 hours ago.
Depth=4000 to 5000 meters (13 000 to 14000 feet)

Can you please tell me where you got this information from?

 

[SiberianTiger]

We can only pray for at least some survivors would be found.

 

[Dambuster]

Against all the usual odds with these events, I reflect your hopes, too.

I agree. I don't tend to think about what's on the news this much, but this has been on my mind all day.

 

[Linguofreak]

An Airbus doesn't just have three computers. It has at least 20 for controlling multiple flight surfaces. the changes of a string of computer failures is next to zero.

Unless they all have a common cause, such as a lightning strike.

 

[Suzy]

Via The Guardian:

Pilots flying a commercial jet from Paris to Rio de Janeiro for Brazil's largest airline, TAM, said they saw what they thought was fire in the ocean along the Air France plane's route early yesterday.

Brazilian Air Force spokesman Col Jorge Amaral said authorities were investigating the report.


"There is information that the pilot of a TAM aircraft saw several orange points on the ocean while flying over the region ... where the Air France plane disappeared," Col Amaral said


"After arriving in Brazil, the pilot found out about the disappearance (of the Air France plane) and said that he thought those points on the ocean were fire."

 

[Urwumpe]

Also, don't forget: At the altitude, the airliner was flying then, we will most likely talk about positive lightning. That is an ion discharge and far more destructive than normal negative lightning. Until a few years ago, such positive lightnings had been considered pilot folklore.

Still, the question is where the lightning would have hit the plane, to cause such a damage.

http://en.wikipedia.org/wiki/Lightning#Positive_lightning

The computer strings on an aircraft are more or less independent and not even on the same electrical power system.

 

[SiberianTiger]

An equatorial thunderstorm's picture, as seen from a plane:

 

[Scrooge McDuck]

Update:
It looks like debris was sighted by Brazilian search planes.
http://news.bbc.co.uk/2/hi/americas/8079122.stm and http://news.bbc.co.uk/2/hi/americas/8077304.stm

 

[Urwumpe]

I heard in the radio, that the plane was without many major systems for three minutes before the crash according to the aircraft telemetry. I can't find much more about this, including internet links, but it sounds more strange now.

I can't say yet, which subsystems failed, how long each was offline and why this caused the crash. But it sounds more like a chain of unfortunate events, than just a single lightning strike. It would also be important to know which antenna system they use for transmitting the telemetry, if it was over satellite (SATCOM), the reception of such data would mean that the flight attitude was still fairly level.

Loss of cabin pressure is another one. If this event was really true (and not reported by the system because of a failed sensor), a lightning, regardless how powerful, would be ruled out. While a strong lightning can destroy parts of the wings and control surfaces, it should not blow holes into the cabin. And a power failure alone would never vent the cabin pressure (It would cause a slow decrease if the A/C compressor failed)

 

[Urwumpe]

This is the sad truth in almost all aircraft accidents.

Yes, but in some situations, the chain is shorter, in others long.

The currently best theory proposed by aircraft accident experts (in TV, not the people who are responsible for finding the real truth) is:

1. Plane flies inside thunderstorm or close to it, in very turbulent air
2. Lightning strikes, power transient sends all computers offline.
3. While computer reboot, pilots have to use the mechanic trim lines.
4. Heavy turbulence makes it impossible to control the craft with trim alone.
5. Aircraft gets thrown out of control and into a steep dive by turbulence.
6. Before the computers have rebooted or the pilot being able to recover the plane from the uncontrolled dive, it crashes onto the stormy ocean surface.

If the computers have been offline for 3 minutes and we assume they did not recover before crash (the news are cryptic about that), the dive from 35,000 ft took place at 11,000 ft/min or 195 ft/s.

That is no controlled glide and also hardly explained by a single turbulence in the region. It would also not happen if the plane was flying level, the flight simulator buffs might verify my extravagant claim, but even after being send into a dive by a "mother of all downwinds", its aerodynamics should pull it level again. Even a tail-heavy flight would not cause that rapid descent.

 

[Notebook]

BBC did report that one of the telemetry messages was reporting a large rate of descent. I obviously don't know what the value is, but it must have been outside the normal for that part of the flight.

N.

 

[Moonwalker]

It would also be important to know which antenna system they use for transmitting the telemetry, if it was over satellite (SATCOM), the reception of such data would mean that the flight attitude was still fairly level.

I think it was SATCOM. ACARS normally transmitts via VHF3 and automatically switches to SATCOM when VHF3 is not available. I can't imagine VHF3 was available that far out above the Atlantic Ocean.

Just to give an idea for those who want to know:

Loss of cabin pressure is another one. If this event was really true (and not reported by the system because of a failed sensor), a lightning, regardless how powerful, would be ruled out. While a strong lightning can destroy parts of the wings and control surfaces, it should not blow holes into the cabin. And a power failure alone would never vent the cabin pressure (It would cause a slow decrease if the A/C compressor failed)

It is even difficult to imagine that a complete loss of power could have happened, including the two batteries, the two engine driven generator units and the RAT, and the potential use of the APU (we don't know yet how busy the crew really was and what they decided for, or what happened at all).

But on the other hand, if they directly flew right into a heavy thunderstorm, which are ususally rather heavy on that route due to the intertropical conversions, there is a good chance of a total loss. It is almost like an unwanted suicide. There is nothing worse for the structure of an airplane like those turbulences that you are going to experience in such storms. Combined with the strong lightnings, I can't imagine they got just one in such intertropical storms, I think multiple systems failures and a total loss as a result becomes very likely. To be honest: I think it was a real nightmare for the crew and passengers. Even more, it was a night flight. Scary is not even the right word...

But, well, I fear we will never know details. The depth of the ocean at the potential crash site is between 4 and 6 kilometers. We all know what this very likely means for the black box recovery...

 

[Linguofreak]

If the computers have been offline for 3 minutes and we assume they did not recover before crash (the news are cryptic about that), the dive from 35,000 ft took place at 11,000 ft/min or 195 ft/s.

That is no controlled glide and also hardly explained by a single turbulence in the region. It would also not happen if the plane was flying level, the flight simulator buffs might verify my extravagant claim, but even after being send into a dive by a "mother of all downwinds", its aerodynamics should pull it level again. Even a tail-heavy flight would not cause that rapid descent.

Aerodynamics might pull it level again, but in a steep dive you might end up overspeeding the airframe and breaking up before you recovered.

Not to mention that damage to the wings or tail during a lightning strike might cause a loss of control that could overstress the airframe in any one of a million different ways. Also, if your wing damage included a fuel tank being ruptured, things could get ugly really fast even without a loss of control. If an engine was damaged in the course of a lightning strike, the compressor or turbine might come apart, which could cause trouble as well.

---------- Post added at 04:33 PM ---------- Previous post was at 04:23 PM ----------

From the Wikipedia article on the event:

The last contact with the aircraft was at 02:14 UTC,[8] four hours after take-off, when its avionics automatically transmitted several messages via ACARS indicating multiple systems failures. [10] The first of these messages, at 2:10 UTC, reportedly indicated that the autopilot had disengaged and the fly-by-wire computers had switched to an alternate program used in the event of multiple system failures. Next, the aircraft transmitted several messages indicating failures of the Air Data Inertial Reference Unit, the Integrated Standby Instrument System (a backup system providing primary flight instruments), and two of the three flight control computers. The final message received, at 02:14 UTC, indicated a possible cabin depressurization at location 3.5777°N 30.3744°W.

The cabin depressurization was reported in the last message, which came four minutes after the first. So it could be that the cabin depressurization indicates the time at which the aircraft broke up.

 

[Moonwalker]

1. Heavy turbulence makes it impossible to control the craft with trim alone.

Actually the recommendation is to disconnect the auto pilot and let the storm "fly" the aircraft. Of course, that is just a theoretical pilot training advice, just like a return is not recommended once you find yourself in the middle of a thunderstorm. But reality looks different and it is actually an open secret that intertropical storms can mean death very likely, even to a bird like the A380.

From that point of view I really hate crossing the Atlantic Ocean. Not a miracle that most pilots do not like that special Rio to Europe route...

 

[Urwumpe]

But, well, I fear we will never know details. The depth of the ocean at the potential crash site is between 4 and 6 kilometers. We all know what this very likely means for the black box recovery...

We need to call James Cameroon?

Honestly: In such a depth, in such situations, we have only a chance for getting the flight recorder, by starting searching as soon as possible and with the best tools as possible.

The FR emits a sonar signals as long as it has battery power, but in such a depth, deep below the eternal thermocline, you can't find it with normal recovery gear. You would need to get serious equipment - towed arrays, ROVs, in the worst case semidivers. Regarding the circumstances, this search would be the nightmare for any ocean recovery team - If the plane had high speed during impact, the plane got pulverized and the 4000-6000m down to the bottom for the FR would be impacted heavily by currently uncharted currents. Even if you would find wreck parts at a point, the flight recorders could be dozens of kilometers away.

 

[Moonwalker]

From the Wikipedia article on the event:

The last contact with the aircraft was at 02:14 UTC,[8] four hours after take-off, when its avionics automatically transmitted several messages via ACARS indicating multiple systems failures. [10] The first of these messages, at 2:10 UTC, reportedly indicated that the autopilot had disengaged and the fly-by-wire computers had switched to an alternate program used in the event of multiple system failures. Next, the aircraft transmitted several messages indicating failures of the Air Data Inertial Reference Unit, the Integrated Standby Instrument System (a backup system providing primary flight instruments), and two of the three flight control computers. The final message received, at 02:14 UTC, indicated a possible cabin depressurization at location 3.5777°N 30.3744°W.

I think two of the 3 primary flight control computers are meant. There are 5 flight control computers on the whole, 3 primary and 2 secondary, while one secondary FCC alone still can provide full aero surface control.

The loss of cabin pressure could also be explained by several systems faults by the way. The Air Data Inertial Reference Unit is connected to the cabin pressurization system for example. But I think a structural damage is the most likely scenario combined with the fact that the electronics and FCC's got "crazy".