Co-pilot error caused AA 587 crash

On Tue, 26 Oct 2004 at 18:12:20 in message
TBCfd.18911$SW3.16862@fed1read01, Jay Beckman
wrote:

Is that due to the crash at the Paris Airshow several years back?

Many people say Paris - it wasn't Paris it was Mulhouse, in the upper
Rhine Valley near the junction of the Swiss and German borders.

It was only a local air show and the aircraft was an almost new A320 (it
had been in service for 2 days). The Airport (if you could call it that)
has one main paved runway only 1000m long plus some grass strips for
gliders. Air France were invited to display an A320. It could not land
there.

Not only that but it was a charter flight with 130 passengers aboard -
how often does that happen at the Paris Air Show I wonder?

The crew were probably given an inadequate briefing on the airport. The
idea was to do a low slow pass in landing configuration at about 100ft.
(Often done in France although the air show regulations said 170 ft.)
They intended to reach the maximum allowable angle of attack in the low
pass. They meant they would inhibit the 'alpha floor' limit which would
automatically increase power at that point. The co-pilot was supposed to
control the power.

When they identified the airport they were close but they saw that the
crowd seemed to be along a grass strip and not along the chosen paved
runway 02. They realigned and at 100 ft deactivated the alpha floor
function. They sank to only 30 ft above the strip. They then suddenly
realised there were trees ahead at the same height or higher than the
aircraft. They then called for TO power but it was too late. Speed had
reduced to 122k and the engines now at flight idle responded as they
should. There was then nothing anyone or the aircraft could do. 4.5
seconds after power started increasing it began hitting the trees.

That is a very much abbreviated version but I believe substantially
correct.

IIRC, the pilot commanded a flight attitude in the landing config that the
software wouldn't allow and that led to the aircraft settling into the
trees.

Sorry that is wrong. And it did not settle into the trees; it flew
horizontally into them at an altitude of 24 ft and then sank!

This accident is very often badly reported. Although the system would
not permit main flight restrictions to be exceeded the performance at
those low limits was as limited as any conventional aircraft would have
been. It could not climb at flight idle at 122 knots and 15 degrees nose
up. That is not a surprise.

This case is a bad example but often used.

Ref: Air Disaster Volume 3 by Macarthur Job. Roughly 13 pages
--
David CL Francis

David CL Francis wrote:

That is a very much abbreviated version but I believe substantially
correct.

My memory has blurred during all those years, but yes, now I rememberm,
this is exactly how it was found in the report. Thanks for refreshing my
memory.

Summary: You can fly any airplane into the trees if you deliberately
wish to do so.

Stefan

David CL Francis wrote:

On Tue, 26 Oct 2004 at 18:12:20 in message
TBCfd.18911$SW3.16862@fed1read01, Jay Beckman
wrote:

Is that due to the crash at the Paris Airshow several years back?

Many people say Paris - it wasn't Paris it was Mulhouse, in the upper
Rhine Valley near the junction of the Swiss and German borders.

It was only a local air show and the aircraft was an almost new A320 (it
had been in service for 2 days). The Airport (if you could call it that)
has one main paved runway only 1000m long plus some grass strips for
gliders. Air France were invited to display an A320. It could not land
there.

Not only that but it was a charter flight with 130 passengers aboard -
how often does that happen at the Paris Air Show I wonder?

The crew were probably given an inadequate briefing on the airport.

Indeed they were.



The
idea was to do a low slow pass in landing configuration at about 100ft.
(Often done in France although the air show regulations said 170 ft.)
They intended to reach the maximum allowable angle of attack in the low
pass. They meant they would inhibit the 'alpha floor' limit which would
automatically increase power at that point. The co-pilot was supposed to
control the power.

When they identified the airport they were close but they saw that the
crowd seemed to be along a grass strip and not along the chosen paved
runway 02. They realigned and at 100 ft deactivated the alpha floor
function. They sank to only 30 ft above the strip.

Nope - they were using the analogue readout barometric altimeter not the more
accurate digital readout radalt ( for reasons of being easier to read whilst
rapidly changing in this instance ).

During the take off phase the radalt and baro altimeter somehow got
'misaligned' by 70 odd feet.

Hence they thought they were flying at 100ft when it was actually 30 ft.


They then suddenly
realised there were trees ahead at the same height or higher than the
aircraft. They then called for TO power but it was too late. Speed had
reduced to 122k and the engines now at flight idle responded as they
should. There was then nothing anyone or the aircraft could do. 4.5
seconds after power started increasing it began hitting the trees.

Actually, increased power was called for as planned. They were indeed unaware
of the trees due to the poor briefing material.

Exactly how and why the engines responded as they did has been a matter of
considerable debate.

That version of the CFM 56 ? was subsequently altered and units in service
'upgraded' following a compressor stall incident on another early A320 at
Geneva ? IIRC.


That is a very much abbreviated version but I believe substantially
correct.

IIRC, the pilot commanded a flight attitude in the landing config that the
software wouldn't allow and that led to the aircraft settling into the
trees.

Sorry that is wrong. And it did not settle into the trees; it flew
horizontally into them at an altitude of 24 ft and then sank!

This accident is very often badly reported. Although the system would
not permit main flight restrictions to be exceeded the performance at
those low limits was as limited as any conventional aircraft would have
been. It could not climb at flight idle at 122 knots and 15 degrees nose
up. That is not a surprise.

The response of the engines was a surprise to the pilots.


Graham

nobody wrote:

Pete wrote:
I thought that was one of the main advantages of fly-by-wire systems,
to eliminate truly stupid actions of pilots. Sounds like Airbus shares
a lot of blame for the crash.

The A300-600 is not fly by wire. It is a 1970s plane updated to some extent
in the 1980s.

And I have been told that because rudders are so rarely used in
flight.............

You're not a friend of John Tarver are you ? He insisted that rudders on big
jets were *purely* yaw dampers.


that Airbus didn't actually make it "smart" with software to restrict movement

depending on airplane's speed etc on its FBW planes.

Note that similar rudder use on Boeing planes would also cause the tail to
break off.

After the accident, I hear that Boeing issued a similary advisory to Airbus
regarding use of rudder.


Graham

Pooh Bear wrote:
You're not a friend of John Tarver are you ? He insisted that rudders on big
jets were *purely* yaw dampers.

Isn't he the one who was certain planes have slaps, a combination of slats and
flaps ?

:-) :-)

Howdy!

In article , nobody wrote:
Pooh Bear wrote:
You're not a friend of John Tarver are you ? He insisted that rudders on big
jets were *purely* yaw dampers.

Isn't he the one who was certain planes have slaps, a combination of slats and
flaps ?

"splaps"... and the P1T0Tube that has nothing to do with measuring dynamic air
pressure...

Just an ID ten T problem...

:-) :-)

yours,
Michael
--
Michael and MJ Houghton | Herveus d'Ormonde and Megan O'Donnelly
| White Wolf and the Phoenix
Bowie, MD, USA | Tablet and Inkle bands, and other stuff
| http://www.radix.net/~herveus/wwap/

"Peter" wrote

But if there's a clear rule for what 'shouldn't be done' then it
would seem prudent to build it into the firmware for the fly-by-wire
system so that it can't be done.

BINGO

Seems to me that Airbus is, if not criminally responsible, morally and
legally responsible.
--
Jim in NC


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.782 / Virus Database: 528 - Release Date: 10/23/2004

On Tue, 26 Oct 2004 18:05:42 -0400, Morgans wrote:


"Peter" wrote

But if there's a clear rule for what 'shouldn't be done' then it
would seem prudent to build it into the firmware for the fly-by-wire
system so that it can't be done.

BINGO

Seems to me that Airbus is, if not criminally responsible, morally and
legally responsible.


????

From what I hear (1) the US certification standards *do not* require the
rudder to be able to withstand the sort of forces the exercise in question
resulted in, and no plane, whether Boeing or Airbus, builds rudders that
would. This is presumably public knowledge, and presumably open
information available to American Airline; incidentally, the same scenario
would have led to a similar accident with a Boeing plane. (2) apparently,
Airbus had repeatedly warned AA about the flaws in AA's training
procedures, which recommended excessive rudder use, even in situations
that were patently unsafe.

So, it does seem to me that the biggest share of the blame should be with
AA. AA doe claim that the warnings from Airbus were not clear enough or
not strong enough. My problem with that is that AA was recommending the
same procedure with Boeing planes too. So presumably Boeing's warnings
were not strong enough either?

"devil" wrote in message
news
On Tue, 26 Oct 2004 18:05:42 -0400, Morgans wrote:


"Peter" wrote

But if there's a clear rule for what 'shouldn't be done' then it
would seem prudent to build it into the firmware for the fly-by-wire
system so that it can't be done.

BINGO

Seems to me that Airbus is, if not criminally responsible, morally and
legally responsible.


????

From what I hear (1) the US certification standards *do not* require the
rudder to be able to withstand the sort of forces the exercise in question
resulted in, and no plane, whether Boeing or Airbus, builds rudders that
would. This is presumably public knowledge, and presumably open
information available to American Airline; incidentally, the same scenario
would have led to a similar accident with a Boeing plane.
*****************************

My point was that a FBW aircraft that did not have limiting software, is
wrong. I now see that the plane in question was not FBW.

"Nevermind! "g


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.782 / Virus Database: 528 - Release Date: 10/23/2004

My point was that a FBW aircraft that did not have limiting software,
is wrong. I now see that the plane in question was not FBW.
"Nevermind! "g

Yeah, me too. I assumed all Airbus aircraft employed FBW. Mea
culpa.

But to start another flame war, maybe AA has a culture problem
of ignoring manufacturers' advice. Remember that it was an AA
DC-10 that lost an engine at ORD, and AA's maintenance practice
of removing engines with a forklift was the culprit, contrary to
McDonnell Douglas' advice.


Pete