Emirates emergency landing in JNB

[411A]

Ah well, last I looked, JNB was at about 5558 or thereabouts...not 8000 feet.

[SecurID]

not a cover up operation as they did 5 years ago

???

Please elaborate

[SecurID]

Hate to say it 411A but CAP56 is right. Density altitude on the day in question was around 8000ft, something like ISA+20C.

CAP56, you did not answer my previous question. When was there a cover up?

[skyslave]

Cap 56, please check your PMs

[JABAL]

Befor you all jump on the bandwagon and attack me this is only a RUMOUR heard it down route!!AND NO, I am not starting a crusade or Jihad against EK.
Because of MFF(mixed fleet flying) there has been strong emphasis on rotation and tailstrike. The rumour is the PF was a TRE,using flex t/off rwy 21R.the rotation was slow, the a/c drags along the rwy, PF applies TOGA the a/c pitches up with instant forward stick input from PF sidestick to counteract the pitchup tendency. once airborne part of the damage from the landing gear strikes the wing trailingedge causing the flaps to cease. Capt decides to jettison to max landing wt, and returns to land. On landing most tyres bust and a/c comes to stop at end or rwy.

[4HolerPoler]

The only thing that changes on an A340-300 when you go from FLEX to TOGA is the length of your arm. Especially in a place like Joburg. There's no "pitching" moment. Otherwise a good story JABAL. Can't wait to see what Flight has to say; normally good gen.

4HP

[DouglasDigby]

Flight International article

Looks like JABAL was very close, if not spot on. Hope this doesn't run into copyright problems, from Flight International (11th - 17th May 2004), but perhaps it makes a few things clearer & will stop some of the arguments:

Operational procedures and take-off calculations remain in the frame as accident investigators continue probe.

Centre-of-gravity problems and aircraft technical faults have been ruled out as causes of last month's Emirates Airbus A340-300 take-off incident at Johannesburg airport, South Africa, according to sources close to the investigation (Flight International, 20-26 April).

However, South African Civil Aviation Authority chief accident investigator, Dr Andre de Kock, declines to comment, and the aircraft manufacturer will say only that it has not been instructed - nor had cause - to issue any technical or operational advice to A340 operators as a result of the event.

The aircraft, bound for Dubai with 216 passengers and 14 crew on 9 April, began its take-off roll on Johannesburg's 4,420m (14,490ft) long runway 21R, but did not get airborne until it had passed beyond the end of the runway, blowing three main gear tyres and damaging the flaps, which subsequently locked in a partly deployed position. It then returned to land having dumped fuel.

Flight International has obtained details of the take-off run from sources close to the investigation: the aircraft rotated to 6º nose up at a calculated rotate speed (VR) of 150kt (278km/h); the pitch was then reduced to 3.5º before increasing again as the crew selected take-off/go-around (TOGA) thrust; Unstick occurred at 175kt, 17s after initial rotation with 9º nose up. When the aircraft returned for landing four more of its tyres burst on touchdown, but the landing roll was completed safely.

The A340-300 flightcrew operating manual (FCOM) advice for take-off is, at VR, to begin rotation at a steady rate of 2.5-3º/s toward 12.5º nose-up attitude. Providing VR is correct, according to the FCOM, the aircraft will have left the ground by the time a 12.5º pitch angle has been attained. Tailscrape angle for the A340-300 with oleos not compressed is given as 14º in the manual.

Although 150kt was the VR the crew used, sources close to the investigation cannot say whether or not this was the correct VR for the prevailing weight, altitude and temperature. The sources add that they are not aware of the crew having attempted to use anything less than the whole runway length for the take-off.

Emirates declines to comment on the information.

[ReginaldSpotter]

411A,
Next time you look at a map for the height you might notice that Joburg is in the fairly hot bit of the map, (towards the middle)which does affect the density altitude a bit.

[411A]

All well and good, Reginald, but on all jet aircraft that I have flown, the takeoff performance calculations started with pressure altitude, corrected for temperature.

Unless Airboos aircraft are different (more than likely...) perhaps the takeoff thrust data was incorrect.
If derived from the FMS, and not checked with the AFM performance tables....the result may not be as advertised.
All speculation of course, will be interesting to find out the true cause, if it comes out at all.

[White Knight]

Not from the FMS, but good old fashioned manuals.

[Invictus]

Focus of Concern ?

Hello all... I am NOT a flying type..

Is it possible that a pilot might "resist" trained techniques if there has been an overemphasis on "Tail Strikes" during training.

Also, with the commonality of flight decks that Airbus are known for, is there a risk that Airlines are not putting as much time and emphasis on actual flight training on new type conversions ?


iNVICTUS

[Cerberus]

Invictus,

I think you may find that your comments have caused a nail to have a very sore head.

Cap,

Have you been to the Sir Humphrey School of Letter writing? That last diatribe was almost unintelligible; say what you mean man!!!

My opinion,

(Remembering that opinions are like ar**holes, we've all got one and they usually stink)

If there was nothing wrong with the aircraft or loading which appears likely then an inappropriate configuration, inappropriate training or an erroneous rotation technique are all that are left. I guarantee the pilots concerned were not planning to screw up!!

Cap, you are probably correct that CCQ across too many aircraft types can be a contributory factor. I am certain that emphasis on tailstrike avoidance rather than correct rotation technique to lift the aircraft off without hitting the tail is also a factor.

A lot of debate surrounds rotation training techniques and the use of the 'iron cross' prior to SRS guidance becoming available or put the stick in this position and wait. However the weight range available on the 340-300 and the differences between the -300 and -500 are likely to make the use of feel or control surface positioning pretty worthless. A T/O out of Jo'burg in a full -300 requires different control surface positions to say a T/O in an empty -500 out of Al Ain. In fact with a heavy aircraft, initially the control surfaces should probably be over deflected to overcome inertia and then reduced to provide the required rotation rate. Once the system goes to load factor demand from direct that is essentially what occurs.

Cap you are certainly right that stick and rudder skills have decayed over the years but that is a trade off for the safety improvements that better automation have brought. Perhaps though, the attempt to use feel rather than looking at the picture and ensuring that the actual rotation rate meets the required rate is more important. Maybe a lot of experienced pilots fall back into trying to use feel when the Bus' flight control system is almost devoid of it. Better to ensure that the nose of the aircraft is doing what you want!

So overall, this will probably end up like most other incidents in that the crew did not rotate the aircraft correctly. BUT.....the contributarory factors were:

1. The airline was expanding so rapidly that the crew had very limited experience of the jet they were flying.

2. The risks of hitting the tail on take-off were over-emphasised at the expense of the correct lift off technique.

3. The trainers provided some useful hints and tips from what they had seen which although ok in the Al Ain circuit or sim probably didn't help out on a dark night, at high altitude in Africa

So enough of my rampant speculation, I must stress that I do not have any priviliged knowledge and have only gleaned my information from what has already been written and by chatting with mates. I am glad it wasn't me on the flight deck and I am delighted that no one was hurt and trust the guys concerned will be not unduly disadvantaged. I hope that the right lessons will be learn't from the incident and that it results in improved safety for all in the future.

Cerberus

[autoflight]

Normally the load sheet is accepted as fairly close to the truth. Checking is for calculation or procedure errors. It is expected that crews will exercise discretion about physical checks.
Some things are obvious. If you expect the flight to have a full pax load, or a specified freight, and the figures don't show this, its grounds for at least a question. 300 pax and 30 bags also shouldn't compute. Too many doubts could finally be grounds for a physical check. Excess physical checks of payload & distribution will result in questions from Chief Pilot, and in many airlines, demotion or loss of job.
Crews should take all reasonable steps to determine the load before checking the load sheet. Mostly they might be quite rough estimates from observations during or after loading. We all need to be streetwise in this area.
Usually we have to have a measure of faith in our loaders.

[bonajet]

The interim findings must be out. Any information available?

[druckmefunk]

Well, for one the SVP Flight Operations was sacked two days ago.

[Plastique]

Revised FOT

Airbus has just revised the FOT with some more details. It appears that the PF used the commanded pitch angle data on the PFD as a reference during rotation. PF tried to keep the tick for commanded pitch angle steady, however as the angle of attack increased the scale on the PFD changed, so the achieved pitch angle was less than the PF desired.
The FOT goes on to recommend use of an external visual reference during rotation - i.e. fly the aircraft like an aircraft and not a video game...

[Plastique]

I think people are missing the point, the real story here is the botched take-off. The comparitively uneventful landing (given the condition of the aircraft) was simply a consequence of all the damage from the trip through the scenery on the way up.

In my opinion, fortune shone brightly on every man, woman and child on that aircraft. It is a tribute to the design and robustness of the aircraft that they are all still alive.

[CDRW]

Plastique - thought that your sentence should read:

"It is a tribute to the design of the aircraft that the accident happened at all"

Is there a conncection between this and the SVP getting the boot?

[SortieIII]

"Heh heh" CDRW , I enjoyed that.

[jtstream]

This comes from Airbus addressed to all A330/340 operators:

ADDITIONAL ANALYSIS OF DFDR DATA, AND DISCUSSIONS WITH THE
OPERATORS, HAVE ENABLED THE IDENTIFICATION OF A POSSIBLE
SCENARIO INVOLVING, WHAT APPEARS TO BE, THE MAIN CONTRIBUTING
FACTOR LEADING TO THIS EVENT. EVEN IF INVESTIGATIONS, CONDUCTED
BY THE "AUTHORITIES OF THE COUNTRY OF OCCURRENCE", ARE STILL
ONGOING, WE FELT IT NECESSARY TO INFORM ALL A330 AND A340
OPERATORS ABOUT THESE RECENT FINDINGS. THESE FINDINGS ARE,
HOWEVER, PROVIDED WITH THE AGREEMENT OF THESE INVESTIGATION
AUTHORITIES AS WELL AS THE OPERATOR.

REVISED OR NEW PARAGRAPHS ARE INDICATED BY: **REV**.

SUBJECT: LIFT-OFF BEYOND RUNWAY END

1 - DESCRIPTION OF THE EVENT

THE A340-313 TOW WAS 233.000 KG AND THE TAKEOFF WAS PERFORMED
WITH MAXIMUM FLEX T/O THRUST AND CONF 1+F. DURING ROTATION, THE
AIRCRAFT REACHED THE RUNWAY END, THEN FINALLY LIFTED OFF AFTER
HAVING HIT SOME LIGHTS. THREE TIRES BLEW UP AND FLAPS MECHANISM
WAS DAMAGED, APPARENTLY BY TIRE DEBRIS. THIS DAMAGE JAMMED THE
FLAPS IN CONF 1+F.

THE CREW PERFORMED FUEL JETTISONING TO PREPARE FOR RETURN, AND
LANDING WITH FLAPS JAMMED. THE AIRCRAFT STOPPED ON THE RUNWAY,
WITH 4 ADDITIONAL TIRES BURST.

2 - EVENT ANALYSIS

2.1 - PRELIMINARY ANALYSIS

BASED ON A PRELIMINARY ANALYSIS OF DFDR DATA, THE FOLLOWING HAS
BEEN OBSERVED:
THE INITIAL TAKEOFF ROLL WAS NORMAL, UNTIL THE BEGINNING OF THE
ROTATION. NO AIRCRAFT MALFUNCTION CAN BE IDENTIFIED. AT THE
COMPUTED VR 150 KT, THE PITCH INITIALLY INCREASED TO ABOUT 6
DEGREES, BEFORE REDUCING TO ABOUT 3.5 DEGREES. FINALLY, PITCH
INCREASED AGAIN, LEADING TO LIFT-OFF 17 SECONDS AFTER ROTATION
INITIATION AT A CAS OF 175 KT. AT LIFT-OFF, THE PITCH ATTITUDE
WAS AT 9 DEGREES, INCREASING, AND TOGA THRUST WAS SET.

THIS FIRST ANALYSIS INDICATED THAT THE PITCH VARIATION DURING
ROTATION, THE ELEVATOR POSITION, AND THE SIDESTICK INPUTS, AS
RECORDED ON THE DFDR, WERE CONSISTENT. NO ABNORMAL SYSTEM
BEHAVIOUR WAS IDENTIFIED. THE AIRCRAFT RESPONSE IN PITCH WAS
COHERENT WITH THE PITCH ORDER.

2.2 - ANALYSIS OF THE ROTATION TECHNIQUE

**REV**

A REPRODUCTION OF THE PFD DISPLAY, BASED ON DFDR DATA, HAS SHOWN
THAT THE PILOT FLYING (PF) USED THE SIDESTICK POSITION SYMBOL
DISPLAYED ON THE PFD WHEN THE AIRCRAFT IS ON THE GROUND (THE
CROSS) TO PERFORM THE ROTATION DURING TAKEOFF.
AT VR, THE PF PULLED ON THE SIDESTICK TO SET THE CROSS AT A
POSITION CORRESPONDING TO ABOUT 9 DEGREES ON THE MOVING PITCH
ATTITUDE SCALE. AT THAT TIME, THE AIRCRAFT PITCH ATTITUDE WAS
CLOSE TO ZERO.
DURING THE ROTATION, THE PF ATTEMPTED TO KEEP THE CROSS ON THE
SAME POSITION OF THIS MOVING REFERENCE. HOWEVER, IN THIS PHASE,
THE AIRCRAFT PITCH ATTITUDE INCREASES, AND THEREFORE THE PITCH
ATTITUDE SCALE OF THE PFD MOVES DOWN. AS A CONSEQUENCE, THE FACT
OF MAINTAINING THE CROSS ON THE SAME MARK (ABOUT 9?) OF THE
MOVING SCALE LED THE PF TO PROGRESSIVELY REDUCE THE PITCH UP
SIDESTICK ORDER.

3 - DESIGN CONSIDERATIONS

**REV**

ON GROUND, THE SIDESTICK POSITION IS INDICATED BY THE POSITION
OF A CROSS, RELATIVE TO THE FOUR WHITE RIGHT-ANGLE MARKS THAT
REPRESENT THE MAXIMUM LATERAL AND LONGITUDINAL SIDESTICK
DEFLECTION. ON THE PFD, THE FOUR RIGHT-ANGLE MARKS ARE FIXED,
AND THE CROSS MOVES RELATIVELY TO THE FIXED SYMBOLS. THEY APPEAR
SUPERIMPOSED TO THE MOVING ATTITUDE SCALE, BUT THEY ARE NOT
REFERENCED TO IT.

THE SIDESTICK POSITION SYMBOL (THE CROSS) WAS DESIGNED TO BE
USED ON THE GROUND FOR TWO MONITORING PURPOSES:
- DURING THE FLIGHT CONTROLS CHECK (IN ASSOCIATION WITH THE
ECAM F/CTL PAGE),
- DURING THE INITIAL TAKEOFF ROLL IN ORDER TO ALLOW THE
PILOT NON-FLYING TO MONITOR THE PF SIDESTICK INPUTS.

THE SIDESTICK POSITION SYMBOL WAS NOT DESIGNED TO BE USED DURING
TAKEOFF ROTATION.


4 - OPERATIONAL RECOMMENDATIONS

**REV**

THE PF MUST PERFORM THE ROTATION MAINLY HEAD-UP USING OUTSIDE
VISUAL REFERENCES UNTIL AIRBORNE, OR AT LEAST UNTIL VISUAL CUES
ARE LOST, DEPENDING ON VISIBILITY CONDITIONS. THE PF MUST THEN
MONITOR THE PITCH ATTITUDE ON THE PFD.
PILOTS MUST BE TRAINED TO STRICTLY ADHERE TO THE ROTATION
TECHNIQUE PROVIDED IN THE FCOM SOP (3.03.12, PAGE 2), AND IN THE
FCOM SUPPLEMENTARY TECHNIQUES (3.04.27, PAGE 3).

THE KEY POINTS OF THIS ROTATION TECHNIQUE ARE OUTLINED BELOW:

- AT VR, INITIATE THE ROTATION WITH A POSITIVE SIDESTICK
INPUT TO ACHIEVE A CONTINUOUS ROTATION RATE OF ABOUT 3
DEGREES PER SECOND TOWARDS A PITCH ATTITUDE OF 12.5
DEGREES (A340-200/300), OR 15 DEGREES (A330 AND A340-
500/600).
- AS THE A340 AND A330 HAVE A LARGE INERTIA, IT IS IMPORTANT
TO INITIATE THE ROTATION WITH A POSITIVE BACKWARD STICK
INPUT (TYPICALLY 2/3 BACKSTICK). THE ROTATION RATE,
PRODUCED BY A GIVEN SIDESTICK INPUT, TAKES TIME TO BUILD
UP. ONCE IT HAS DEVELOPED, IT REMAINS RELATIVELY CONSTANT
FOR A GIVEN SIDESTICK POSITION. THE PILOT CONTINUES THE
ROTATION TOWARDS THE INITIAL PITCH TARGET OF 12.5 DEGREES
(A340-200/300), OR 15 DEGREES (A330 AND A340-500/600).
THEN, AFTER LIFT-OFF, THE PILOT FOLLOWS THE SRS ORDERS
WHEN THE FD PITCH BAR IS STABILIZED.
- RAPID VARIATION IN THE SIDESTICK POSITION WILL CAUSE SHARP
CHANGES IN THE RATE OF COCKPIT MOVEMENT, PARTICULARLY ON
LONG-FUSELAGE AIRCRAFT WHERE THE PILOT STATION IS WELL-
FORWARD OF THE MAIN WHEELS. SUCH VARIATIONS SHOULD BE
AVOIDED.