Another Blow to Airbus

On Aug 12, 1:42*pm, a wrote:
The good news is, it's rare to the best of my
knowledge that moving the controls to the extremes,. either fast or
slow, will uncover a problem, or that they fail in flight.


Err, I don't think you mean that.

Cheers

On Aug 13, 12:18*am, "vaughn" wrote:
"a" wrote in message

...
*Did see someone in a 680 Commander

get out to take off a rudder clamp, that would have been found because
the excursions were not 'free'.

I once saw a renter pilot *return* from an apparently normal flight in a 172
with the rudder lock still installed. *The fellow apparently never noticed!


Never noticed? Wow, that is scary. While I admit to a tendency for
lazy feet, I usually wake them up on final.

Cheers.

Flaps_50! writes:

Because the computers don't know actually know the relationship
between yaw, airspeed and allowable rudder input/structural load and
they are not required to.

They can be programmed to know this. And much of the reputation of Airbus
rests upon its implicit and explicit claims that their heavy computerization
of their flight decks somehow makes aircraft safer. But if the computer isn't
even programmed well enough to prevent something like this, how can it be
making the airplane safer?

Neither do most pilots when they step on the rudder pedals. Think
about it...

It would be tough for a pilot, but not for a computer.

There's a fundamental contradiction between claiming on the one hand that
computer-enforced limitations on control movements can prevent structural
damage, and then claiming on the other hand that computers should not be held
responsible for that enforcement. Either they protect the airplane or they
don't. If they are only creating the illusion of protection, then they need to
go.

On Aug 15, 3:19*am, "Flaps_50!" wrote:
On Aug 9, 1:49*am, Mxsmanic wrote:



a writes:
A couple of days ago the NTSB found the 320 series to have too
sensitive a rudder, it can be torn off with peddle pressures. What's
especially of interest is the problem seems to persist even when crews
are given special training about the problem.

There are some details here.

http://content.usatoday.com/communit.../2010/08/ntsb-....

Hmm. The whole purpose of having computers that fly the airplane, and ignore
the pilots' inputs if they find them contrary to what French engineers have
decided, is to prevent exactly this sort of incident. Why don't the
all-knowing, all-wise computers prevent any rudder movement that might
endanger structural integrity?

Because the computers don't know actually know the relationship
between yaw, airspeed and allowable rudder input/structural load and
they are not required to.

You jest, surely?? If I understand what you say, most flights in
Autopilot through moderate turbulence would result in splintered
aluminum tubes raining down.

Any pilot would tell you that humans are incapable of matching
computers' sophistication in precision flying. Why else would most
airline SOPs actually bar pilots from hand-flying above 1,000 feet?


Neither do most pilots when they step on the
rudder pedals. Think about it...


Well, if the manufacturer intends to convey a limit that can be
reached where the airplane's structure is threatened, it should either
automatically limit the pilot's motion to a point before such
threshold is reached. Proper pilot training is a sine qua non, however
not a substitute for the automatic limitation.

I still can't believe the ultra-sophisticated Airbuses allow rudders
to move so much that the empennage can actually sever from the rest of
the fuselage. As omissions go, that must take the biscuit!

Cheers,

Ramapriya

D Ramapriya writes:

Any pilot would tell you that humans are incapable of matching
computers' sophistication in precision flying. Why else would most
airline SOPs actually bar pilots from hand-flying above 1,000 feet?

Above 1000 feet? Did you miss a zero there?

I know that RVSM requires autopilot and some airlines have policies that
require autopilot for normal operations under certain conditions, but
requiring that autopilot be used above 1000 feet is hard to believe.

Which airlines require this, and why?

I still can't believe the ultra-sophisticated Airbuses allow rudders
to move so much that the empennage can actually sever from the rest of
the fuselage. As omissions go, that must take the biscuit!

Having been the victim of French engineering on multiple occasions in the
past, I have no trouble believing that French engineers overlooked this. Their
objective is not to maximize safety, but to show the world how clever they are
(a rather tall order, given that they aren't actually very clever).

On Aug 15, 6:41*am, Mxsmanic wrote:
D Ramapriya writes:
Any pilot would tell you that humans are incapable of matching
computers' sophistication in precision flying. Why else would most
airline SOPs actually bar pilots from hand-flying above 1,000 feet?

Above 1000 feet? Did you miss a zero there?

Nope. 1000-2000 are the typical figures for most airlines in the
Middle East. I could get you figures from Qatar Air and Etihaad (two
of the leading lights, not just of the region) in a few days.


I know that RVSM requires autopilot and some airlines have policies that
require autopilot for normal operations under certain conditions, but
requiring that autopilot be used above 1000 feet is hard to believe.


Ditto here, but that's the way it is. "Passenger comfort and safety"
is what's apparently at the root of this requirement.


Which airlines require this, and why?

I still can't believe the ultra-sophisticated Airbuses allow rudders
to move so much that the empennage can actually sever from the rest of
the fuselage. As omissions go, that must take the biscuit!

Having been the victim of French engineering on multiple occasions in the
past, I have no trouble believing that French engineers overlooked this. Their
objective is not to maximize safety, but to show the world how clever they are
(a rather tall order, given that they aren't actually very clever).


I beg to differ, mate. Apart from one A320 crash - a runway overrun in
Warsaw? - where the computers misread aquaplaning and didn't allow
braking, I struggle to think of an incident where computers and/or
automation caused a crash. On the other hand, I know a few instances
where the automation forfended accidents by thwarting ill-judged
premature takeoff attempts, which were an upshot of wrong loading
figures having been input, etc. There have been at least two incidents
involving Emirates A340 aircraft and one Virgin A330.

Not being a pilot, I'm utterly unqualified to enter Boeing-Airbus
debates but it does strike me that Boeing does have more friends in
the press, with its glitches getting downplayed. The dicky RA that
contributed to the Turkish crash at Schipol and the near-disaster with
the BA 747 @ Jo'burg caused by a faulty slat sensor are good examples.
If you analyze Airbus crashes, nearly every one of them has been
because of pilot error, including the Aeroflot A310 where they risibly
ended up blaming the kid on the Cap'n's seat when what really happened
was that the 3 other qualified pilots looking on within the cabin
failed for a very long time to detect that the AP had disconnected.
Most Airbus crash reports would tell you that they could've been
prevented had pilots acted correctly.

I admire the 747s and 777s and think the A340 a clunker, yet would
wager my life on Airbus's sophistication any day. It could be just me
but that's the way it is

Ramapriya

On Aug 15, 1:31*pm, D Ramapriya wrote:
On Aug 15, 3:19*am, "Flaps_50!" wrote:





On Aug 9, 1:49*am, Mxsmanic wrote:

a writes:
A couple of days ago the NTSB found the 320 series to have too
sensitive a rudder, it can be torn off with peddle pressures. What's
especially of interest is the problem seems to persist even when crews
are given special training about the problem.

There are some details here.

http://content.usatoday.com/communit.../2010/08/ntsb-...

Hmm. The whole purpose of having computers that fly the airplane, and ignore
the pilots' inputs if they find them contrary to what French engineers have
decided, is to prevent exactly this sort of incident. Why don't the
all-knowing, all-wise computers prevent any rudder movement that might
endanger structural integrity?

Because the computers don't know actually know the relationship
between yaw, airspeed and allowable rudder input/structural load and
they are not required to.

You jest, surely?? If I understand what you say, most flights in
Autopilot through moderate turbulence would result in splintered
aluminum tubes raining down.


I don't know if autopilots ever put in full rudder deflection during
yaw -do they?

Cheers

On Aug 15, 1:45*pm, "Flaps_50!" wrote:
On Aug 15, 1:31*pm, D Ramapriya wrote:



On Aug 15, 3:19*am, "Flaps_50!" wrote:

On Aug 9, 1:49*am, Mxsmanic wrote:

a writes:
A couple of days ago the NTSB found the 320 series to have too
sensitive a rudder, it can be torn off with peddle pressures. What's
especially of interest is the problem seems to persist even when crews
are given special training about the problem.

There are some details here.

http://content.usatoday.com/communit.../2010/08/ntsb-...

Hmm. The whole purpose of having computers that fly the airplane, and ignore
the pilots' inputs if they find them contrary to what French engineers have
decided, is to prevent exactly this sort of incident. Why don't the
all-knowing, all-wise computers prevent any rudder movement that might
endanger structural integrity?

Because the computers don't know actually know the relationship
between yaw, airspeed and allowable rudder input/structural load and
they are not required to.

You jest, surely?? If I understand what you say, most flights in
Autopilot through moderate turbulence would result in splintered
aluminum tubes raining down.

I don't know if autopilots ever put in full rudder deflection during
yaw -do they?


My point is that flight automation would forestall a situation where
the airframe is imperiled, including not deflecting the rudder beyond
safe limits.

Ramapriya

D Ramapriya writes:

Nope. 1000-2000 are the typical figures for most airlines in the
Middle East. I could get you figures from Qatar Air and Etihaad (two
of the leading lights, not just of the region) in a few days.

I had a sneaking suspicion that "most airlines" actually meant "most
Third-World airlines." That's the kind of rule I'd expect from them. I can
think of several reasons for such a rule ... and all of them are bad.

I don't think Southwest or British Airways are forbidding their pilots to fly
by hand above 1000 feet. You're not even clear of obstacles at that height.

Ditto here, but that's the way it is. "Passenger comfort and safety"
is what's apparently at the root of this requirement.

A serious misunderstanding of how safety works is probably at play as well.

I beg to differ, mate. Apart from one A320 crash - a runway overrun in
Warsaw? - where the computers misread aquaplaning and didn't allow
braking, I struggle to think of an incident where computers and/or
automation caused a crash.

Well, there's Habsheim ... but we cannot be sure, since Airbus modified and
removed data on the flight data recorders in order to hide something (and I
don't think it was pilot incompetence).

On the other hand, I know a few instances
where the automation forfended accidents by thwarting ill-judged
premature takeoff attempts, which were an upshot of wrong loading
figures having been input, etc. There have been at least two incidents
involving Emirates A340 aircraft and one Virgin A330.

Maybe if the pilots were more competent and actually flew hands-on a bit more,
those problems wouldn't arise.

It's not the computers' job to compensate for incompetent crews.

Not being a pilot, I'm utterly unqualified to enter Boeing-Airbus
debates but it does strike me that Boeing does have more friends in
the press, with its glitches getting downplayed.

It has more friends among pilots and mechanics, that's for sure. Boeing
designs airplanes that help a pilot do his job. Airbus designs airplanes that
try to eliminate the pilot's job.

The dicky RA that contributed to the Turkish crash at Schipol ...

If it's the one I'm thinking of, the pilots were the weak spot, not the RA.

If you analyze Airbus crashes, nearly every one of them has been
because of pilot error, including the Aeroflot A310 where they risibly
ended up blaming the kid on the Cap'n's seat when what really happened
was that the 3 other qualified pilots looking on within the cabin
failed for a very long time to detect that the AP had disconnected.
Most Airbus crash reports would tell you that they could've been
prevented had pilots acted correctly.

This is true for all crashes, not just Airbus crashes.

The problem is that the Airbus design philosophy encourages the employment of
less competent pilots, since the computers will take care of everything (in
theory).

I admire the 747s and 777s and think the A340 a clunker, yet would
wager my life on Airbus's sophistication any day. It could be just me
but that's the way it is

If it's not Boeing, I'm not going.

D Ramapriya writes:

My point is that flight automation would forestall a situation where
the airframe is imperiled ...

So would good pilots.