Blogs and Stories

Clive Irving

The Secrets of Flight 447

BS Top - Irving Flight 447 Mystery AP Photo Brazil's air force discovered passenger remains from Air France Flight 447 Saturday. As they continue the search, Clive Irving explains why the black box might be a moot point—and why Airbus is worried, as French investigators uncover a shocking 24 error messages sent from airplane’s onboard monitoring system.

French investigators looking at the disappearance of Air France Flight 447 have seen something that they do not like. On Saturday, France’s chief crash investigator, Paul-Louis Arslanian, revealed that there were 24 error messages sent from airplane’s onboard monitoring system. And, tellingly, he said that the airplane had had a problem calculating its speed. “We have seen a certain number of these faults on the A330,” he admitted.

On Friday Airbus sent out an urgent notice to airlines operating not just A330s but any Airbus model “reminding” them how to handle an airplane if the instruments recording air speed seem to be in trouble. This directive could only have come as a result of what has been revealed in data received from the doomed airliner by its Aircraft Condition Monitoring System, or ACMS.

ACMS has never before been depended on for clues to a crash. But in the absence of the airplane’s black box, the data is significant but frustratingly incomplete.

What could have happened to Flight 447 to make this specific problem of gauging air speed so critical?

Autopilots don’t know when they are getting bum information.

Indications are that the A330’s nerve ends that measure air speed—two independent “collectors” from the slipstream passing over the airplane—were confused. They thought the A330 was flying faster than it actually was. That, in turn, would cause the automatic pilot, then in control, to slow down until it believed it was reaching the normal speed, of around 500 miles per hour. Consequently, if the airplane is slowing down when, in fact, it does not need to, it can fall to a speed at which it becomes unstable and even stalls. That means that although the instruments are telling the autopilot that it’s flying at 500mph, it could be several hundred miles an hour less than that. At which point it becomes very hard to fly. In the storm there were likely to be vortexes as well as powerful updrafts and downdrafts that could seriously disrupt the airplane’s monitoring of its speed.

Becoming unstable in the kind of violent weather that Flight 447 was encountering over the southern Atlantic would be a recipe for disaster.

Saturday’s statement did not make any clearer at what stage the pilots seized control from the autopilot—or even if they ever did. Chief inspector Arslanian said it was impossible to tell from the ACMS data whether the autopilot had been disengaged or was not working. Timing is crucial. He was talking about what could be divined from the final burst of data. By then, the A330 was suffering a series of terminal failures. In effect, the airplane was becoming brain dead. The pilots should have taken control much earlier than that.

Autopilots don’t know when they are getting bum information. Unlike pilots, they have no ears, eyes, or instincts. A very relevant example occurred in thick clouds over Amsterdam on February 25. A Turkish Airlines Boeing 737 stalled and fell out of the sky. Its altimeters—measures of its altitude—were giving false readings. The autopilot thought the airplane was on final approach. In fact, it was still at 2,000 feet. The two pilots were chatting to a third in a jump seat. They failed to hear the engines power down to idle, they failed to see that they were stalling. The three in the cockpit died, along with six passengers.

Now that the French authorities seem pretty sure that the autopilot on Flight 447 was getting false information on air speed, the acuity of the pilots remains a huge question.

Xtra Insight: Before 447: Seven Other Aviation Mysteries

Xtra Insight: Clive Irving asks Who Was Flying Flight 447?

Clive Irving is senior consulting editor at Condé Nast Traveler, specializing in aviation.


Back to Top
June 6, 2009 | 6:56am
Comments ()
robodog

You say "Why Airbus is worried." All aircraft manufacturers are worried after a crash - for good reason. They are going to spend months and months in court defending their equipment. This is nothing new but this article lends a sensational speculative air to the (lack of) information.

In fact, the article is pure speculation by an alleged expert who uses unvalidated information to support a scenario that is possible but has not been supported by anywhere near enough facts to be considered even remotely valid.

Clive the expert says: "Saturday's statement did not make any clearer at what stage the pilots seized control from the autopilot - or even if they ever did. M. Arslanian said it was impossible to tell from the ACMS data whether the autopilot had been disengaged or was not working." So it is supposedly impossible to tell if it was working or not. But two sentences later, Clive the expert says: "The pilots should have taken control much earlier than that." What kind of BS is this? You don't know what you are talking about.

Furthermore, you draw the analogy to the Turkish accident in Amsterdam. Again, the accident report is not complete on that accident, but I will grant you that it appears that the pilots were not "in the game" when they had a serious data malfunction. But AF 447 is entirely different. At cruise altitude in the middle of the night at cruise speed, apparently with multiple electrical problems and possibly getting tossed around so violently that they couldn't even read the instruments [yes, that can happen], Clive the expert knows precisely what they should have done. All without the benefit of the flight data recorder and cockpit voice recorder.

Give me a break.

|
|
Reply
9:35 am, Jun 6, 2009
dwurry

It remains to be seen if Airbus can conduct a valid investiation and acknowledge its problems. Boeing went through this in the early 70's with the DC-10. If Airbus fails--both to conduct a valid investigation or to address the nightmare public relations issues it will be a significant problem for Airbus.

|
|
Reply
|
10:43 am, Jun 6, 2009
Joe206

The DC-10 was a McDonald Douglas aircraft. They (McD-Douglas) did not merge with Boeing until 1997.

|
|
Reply
11:35 am, Jun 6, 2009
hibou2121

HERE"S and interesting and radical idea for you.....WHAT IF....some faction of terrorist (say perhaps even aliens) used multidimensional technology to zap the Air France flight out of the sky....If you check U TUBE for declassified Area 51 footage, one can see "clouds". Apparently under Truman our government negotiated a secret contract which allowed us to back engineer alien craft. Any of our stealth crafts at any military show will display anti gravity type technology. It sounds radical but we are officially in 3d but ask any physicist who studies string theory if that is really true.....WAKE UP PEOPLE.....it could happen. Look at England's crop circles. Not ALL of those could be man made. Just a wild theory and argument. All is Seen and Unseen by the Divine...The TRUTH will eventually come to light. That's Instant Karma my friends.

|
|
Reply
7:09 pm, Jun 7, 2009
Hawnzz

Every place model, with their extreme complexity will have engineering issues. (Hopefully not this severe) But this particular model has an amazing success rate. Planes do not have accidents because of one failure. They fail because of many unique factors happening all at once. They fail for reasons that an engineer cannot think of.

Before you throw Airbus under the bus... lets find out what happened first. Boeing has had just as many issues in it's long life. This isn't the first time and won't be the last for such things to happen. But as we know, they are EXTREMELY rare.

|
|
Reply
|
11:59 am, Jun 6, 2009
Hawnzz

place = plane

|
|
Reply
12:16 pm, Jun 6, 2009
jimors

Relative to the airline industry, AIrbus is the new kid on the block. They employ a new technology termed "Fly-By-Wire". What this means is that rather than utilizing redundant hydraulics with the associated hoses and such, they use use servos. The servos are controlled by wires, hence, the name. A fear that the industry had was that the Airbus relied way too much on the computer controls. In fact, the initial platforms actually overruled the pilots decisions. This caused a crash landing at a trade show because the pilots control was overruled. The industry was assured and new revisions showed that this redesigned. I am hoping that this is nothing to do with those preliminary flaws.

Please do not accept my input here as gospel but I am sure that with a little research you will find full articles regarding these issues.

|
|
Reply
|
3:10 pm, Jun 6, 2009
Hawnzz

I would bet, that many of Boeings planes are also fly by wire. Anyone know where we could look it up?

|
|
Reply
|
5:52 pm, Jun 6, 2009
exploora

Here are some links.
[#
Unlike Airbus, Boeing lets aviator override fly-by-wire technology
The only Boeing plane with fly-by-wire technology is the 777. Cashman, who was chief pilot for the 777 program, said Boeing could have designed the 777 with ...
www.seattlepi.com/business/boe202.shtml - Cached - Similar pages
#
Aircraft flight control systems - Wikipedia, the free encyclopedia
1 Jun 2006 ... Boeing followed with their 777 and later designs. Electronic fly-by-wire systems can respond flexibly to changing aerodynamic conditions, ...
en.wikipedia.org/wiki/Aircraft_flight_control_systems - Cached - Similar pages
#
Re: FLY-BY-WIRE (AIRBUS vs. BOEING)
In the case of Boeing, it would >SEEM that the pilot can do what he wants with just a warning on a screen if he >exceeds the aircrafts' capabilities (what ...
www.kls2.com/cgi-bin/arcfetch?db=sci.aeronautics.airliners&id=%3Cairline rs.1995.862@ohare.chicago.com%3E - Cached - Similar pages
#
Re: FLY-BY-WIRE (AIRBUS vs. BOEING)
Re: FLY-BY-WIRE (AIRBUS vs. BOEING). From: mpreuss@electra.synapse.net (Merlin R. Preuss) Organization: Transport Canada Aviation Date: 01 Jun 95 05:00:54 ...
www.kls2.com/cgi-bin/arcfetch?db=sci.aeronautics.airliners&id=%3Cairline rs.1995.702@ohare.chicago.com%3E - Cached - Similar pages
#
How does the fly-by-wire in Boeing 777 differs from the Airbus system?
3 posts - Last post: 4 days ago
Airbus fly by wire. I flew the Boeing - 767, retired quite a while ago, but believe that the windshear recovery maneuver was max power, ...
www.askcaptainlim.com/index.php?option=com_content&view=article&id=631:h ow...the-fly-by-wire...boeing... - Cached - Similar pages
#
Boeing 777 Twin-Aisle Twinjet Airliner - Aerospace Technology
The aircraft has Boeing's fly-by-wire controls with flight envelope protection and automatic pilot and stabilisation functions. The fly-by-wire system sends ...
www.aerospace-technology.com/projects/boeing777/ - Cached - Similar pages
#
Aircraft Ebook: Boeing B-777: Fly-By-Wire Flight Controls - Free ...
Boeing B-777: Fly-By-Wire Flight Controls. Fly-By-Wire (FBW) Primary Flight Controls have been been used in military applications such as fighter airplanes ...
www.pdfee.com/boeing-b-777-fly-by-wire-flight-controls.html - Cached - Similar pages
#
Design Considerations in Boeing 777 Fly-By-Wire Computers
The new technologies in flight control avionics systems selected for the Boeing 777 airplane program consist of the following: Fly-By-Wire (FBW), ARINC 629 ...
ieeexplore.ieee.org/iel4/5939/15811/00731596.pdf - Similar pages
by YCB Yeh - Cited by 27 - Related articles - All 6 versions]

|
6:11 pm, Jun 6, 2009
Hawnzz

Holy merde!

Are you trying to give me eye fatigue! lol

|
6:50 pm, Jun 6, 2009
exploora

:)

|
5:15 pm, Jun 7, 2009

This user is no longer registered.

|
|
Reply
3:35 pm, Jun 6, 2009
drmarkklein

Totally fly by wire airliners aren't airworthy.

|
|
Reply
|
4:55 pm, Jun 6, 2009
Hawnzz

Most of our military planes are fly by wire.

|
|
Reply
|
5:41 pm, Jun 6, 2009
Hawnzz

http://en.wikipedia.org/wiki/Aircraft_flight_control_systems

|
6:17 pm, Jun 6, 2009
C17Pilot

yes they are, and we are better off with fly-by-wire systems

|
8:22 pm, Jun 6, 2009
drmarkklein

Fighter pilots knowingly put their lives on the line each time they fly. Commercial airliner passengers expect to arrive alive at their destinations.

|
8:51 pm, Jun 6, 2009
misha1000

"Most of our military planes are fly by wire."

They are built to tighter standards than civilian aircraft.

Read about military product standards and commercial product ones.

|
10:19 pm, Jun 6, 2009
tnflyboy

you've said this before without providing any justification for it... again, what makes a fly by wire system less airworthy than a mechanical system? First off we need to define the mechanical system. It's similar to the power steering on your car. If the power steering goes out in a geo metro the it's no problem. If it goes out in a semi, I imagine things would get a little difficult. The same is true with a jetliner. The pilot is not going to be able to control the aircraft without the hydraulic system moving the control surfaces. If that hydaulic system, and all its redundants are severed, then you lost control of the aircraft even though you still have physical linkages to the control surfaces. This has happened multiple times in the past, see the DC-10 for an example.

|
|
Reply
2:19 pm, Jun 8, 2009
Blogger55

It seems scanalous that after almost 70 years of commercial avaiation, a modern jet could fall out of the sky due to upper level meteorological conditions with the needless loss of so many lives. All indications are that the crew became disorientated when their airspeed and other instruments failed and stalled or damaged the aircraft by applying the wrong power settings.
I have operated aircraft in in svere thunderstorms and I can tell you that maintaining the correct turbulence penetration speed is absolutely crucial to the safety of the aircraft. Air France and Airbus Industries have shown negligence in this accident.

|
|
Reply
4:21 am, Jun 7, 2009
Blogger55

It seems scandalous that after almost 70 years of commercial avaiation, a modern jet could fall out of the sky due to upper level meteorological conditions with the needless loss of so many lives. All indications are that the crew became disorientated when their airspeed and other instruments failed and they stalled or damaged the aircraft by applying the wrong power settings.
I have operated aircraft in in severe thunderstorms and I can tell you that maintaining the correct turbulence penetration speed is absolutely crucial to the safety of the aircraft. Air France and Airbus Industries have shown negligence in this accident and indeed have a lot to answer for.

|
|
Reply
|
4:42 am, Jun 7, 2009
nozferatu

Why...do you think we've conquered Nature?

|
|
Reply
1:32 pm, Jun 30, 2009
like-mind

It seems to me that slipstream sensors would get confused by violent turbulence of this mega-storm.

|
|
Reply
4:48 am, Jun 7, 2009
Margot62

I'd love to have this question answered:

In this age of computer technology, why do we still have black boxes? Their crucial role and critical importance in a plane crash underscores the need for a remake.

Why can't black boxes transmit their information to a satelite-- a continuous feed of information---where it is stored at a different location? In the same way that subscribers use storage companies to back up their computers? If my computer crashes, my information is safely stored and is retrievable.

I'm sure such a system for black boxes has been considered before, but what prevents us from persuing this?

|
|
Reply
|
3:24 pm, Jun 7, 2009
exploora

My guess, it is the same reason why doctors continue recording medical records on paper. The records are easier for doctors/airlines to control and easier for doctors/airlines to exercise propriety rights when controlling documents, and possibly another reason could be less evidence available show the result of airline negligence if applicable.

|
|
Reply
5:14 pm, Jun 7, 2009
tnflyboy

Looks like Clive lifted your concern and made it his own on his current article...

|
|
Reply
2:20 pm, Jun 8, 2009
hibou2121

HAS ANYONE CONSIDERED MULTIDIMENSIONAL STRING THEORY AND BACKENGINEERED ALIEN TECHNOLOGY COUPLED WITH TERRORIST FACTIONS?

|
|
Reply
7:11 pm, Jun 7, 2009
robodog

If anybody wants to read a GOOD discussion (as opposed to Clive the aviation expert) about this accident, here's a link: http://www.timesonline.co.uk/tol/news/world/us_and_americas/article6446268. ece

|
|
Reply
9:30 pm, Jun 7, 2009
Blogger55

String theories and alien technologies aside, and based on relayed data, this seems to be a combination of pilot error and faulty aircraft design. Pilots are advised to give thunderstorm cells a very wide berth, something like 10 nautical miles. Equatorial cells are especially more violent than tropical ones, so even more reason to avoid them by a wider margin. Flying into one would be not unlike hitting a brick wall of hail, icing and violent draughts. The crew should have detoured or even turned back rather than risk entering a squall line of these killer storms. As for for aircraft design, no size of storm cell should render aircraft sensors useless. This should have been apparent in airworthiness trials before certification. It is well documented that the airbus had issues with its pitot sensors and that these were not rectified immediately is a damning indictment. There are a lot of question marks which need to be addressed as a matter of urgency.

|
|
Reply
|
2:20 am, Jun 8, 2009
C17Pilot

WE avoid thunderstorms by 20NM above 23,000' and 10NM below 23,000'.
For an extra safety margin, we increase the avoidance by 1NM for every knot of wind that is blowing in our direction...needless to say, we avoid them as much as possible.

|
|
Reply
|
3:11 am, Jun 16, 2009
DJM168

I agree with both of you gentlemen. There was a clear case to avoid flying through this area at that given moment, and the crew made a bad call to continue on course. It is also fairly well documented that the computer-controlled systems that control the A330/A340 families of aircraft are pretty seriously flawed. And even though this is known, replacement units have yet to be deployed. So human error on multiple levels abound! I will not discount the brutal forces of that storm system as a major contributor to the tragedy either. AF447 should not have been there. They should have taken evasive action. I just wonder how many more of these jets have to go down before the avionics issues are finally addressed...

|
5:22 pm, Jun 16, 2009
nozferatu

This issue with autopilot stuff is absolutely avoidable.

Why don't manufacturers install a simple GPS backup unit to measure ground speed based on satellite data? That way, regardless of what the air speed readouts are, the autopilot system has back up information it can rely on.

IF a $100 GPS unit is too much money for a cheap airline to handle, then program in a simple comparative algorithms in the autopilot software that compares trend data to what is happening instantaneously. That is to say...if the aircraft has been flying for the pasty 1-5 minutes at constant speed, constant altitude, constant thrust, etc....and suddenly encounters completely different readings and data....it will not act according to instantaneous data but rather trend data. The could be a certain amount of leeway give to control inputs but that should be it.

This would eliminate all sudden inputs from an autopilot.

Autopilots should not be relied upon for sudden control and thrust inputs. They should be used to keep an aircraft steady and true. If this basic function cannot be relied upon by autopilot systems such as in the airbus, then it should be ripped out of the aircraft and thrown away.

|
|
Reply
1:31 pm, Jun 30, 2009
Leave a Comment
Leave a comment

Thank you.
As a first time user, your comment has been submitted for review. It can take anywhere from a few hours to a day or two for your comment to be reviewed, depending on the time of week and the volume of comments we receive.

View Comments
Leave a comment

Please log in to leave comments.

The Secrets of Flight 447

by Clive Irving

Info
RSS
Clive Irving
Emails
|
print
text
-
+
Facebook
 | 
Twitter
 | 
Digg
 |