Go Back  PPRuNe Forums > Flight Deck Forums > Rumours & News
Reload this Page >

A340 Rotation Problems

Rumours & News Reporting Points that may affect our jobs or lives as professional pilots. Also, items that may be of interest to professional pilots.

A340 Rotation Problems

Old 10th Jul 2019, 14:48
  #1 (permalink)  
Thread Starter
 
Join Date: Nov 2015
Location: Paisley, Florida USA
Posts: 265
A340 Rotation Problems

I just saw the linked article in the Daily Digest of the Aviation Week Network.

The link:
https://aviationweek.com/commercial-...de9291345fa795


The problem seems to have been limited to overly shallow takeoff rotations from the high elevation environment of Bogota El Dorado Airport in Columbia. Could this problem be related to lack of side stick feedback feel while trying to avoid a tail strike? It would seem to me, given adequate airspeed etc., that it would be instinctive to apply whatever pitch was required to maximize climb and terrain/obstacle avoidance.

Cheers,
Grog
capngrog is offline  
Old 10th Jul 2019, 16:24
  #2 (permalink)  
 
Join Date: Jun 2019
Location: Tana
Posts: 0
It's about time Airbus does something about sidestick feedback. At least one flight, AF447, crashed because two experienced pilots didn't know the only rookie in the cabin was pulling on the sidestick. It's very simple logic - if pilots can't see each other's actions, they should feel them on their controls. Yet, even the newest A350 still goes without it. Airbus should work on this issue before it turns into their very own MCAS.

PS Did I read somewhere that the Russian MS-21 has "backfed" sidesticks?
UltraFan is offline  
Old 10th Jul 2019, 16:48
  #3 (permalink)  
 
Join Date: Dec 2002
Location: UK
Posts: 2,090
There is nothing in this which should be particular to side sticks.
The problem involves high altitude take off and would apply to any aircraft irrespective of control system.
A slower acceleration - longer takeoff roll, would require some compensation in rotation rate and / or limiting pitch attitude. If crews overcompensate - too slower rate, or don’t achieve the required takeoff attitude, where the maximum might not be usually seen in other operations, then the ground roll could be more than expected.

The factors reported were the need for faster and higher initial stick inputs and over awareness of the risk of tail strike.
Training refreshed, manuals updated, closing the loop on inservice monitoring; safety management lessons.

safetypee is offline  
Old 10th Jul 2019, 16:51
  #4 (permalink)  
 
Join Date: Jun 2002
Location: Geneva, Switzerland
Age: 55
Posts: 1,666
To be honest I'm always scared by a fully loaded 340 taking off from Bogota...
atakacs is offline  
Old 10th Jul 2019, 17:50
  #5 (permalink)  
 
Join Date: Jul 2013
Location: In a Pineapple Under the Sea
Age: 58
Posts: 104
BEA: Air France A340 abnormal takeoffs caused by pilot input

https://atwonline.com/safety/bea-air...ed-pilot-input

"In an investigation report that effectively highlights the value of flight data monitoring, the French Bureau d’Enquetes et d’Analyses (BEA) has determined that some abnormally long Airbus A340-300 takeoffs from Bogota El Dorado International Airport in Colombia were the result of improper pilot practices."
WillFlyForCheese is offline  
Old 10th Jul 2019, 18:03
  #6 (permalink)  
 
Join Date: May 2002
Location: mighty Blighty
Posts: 286
Originally Posted by UltraFan View Post
It's about time Airbus does something about sidestick feedback. At least one flight, AF447, crashed because two experienced pilots didn't know the only rookie in the cabin was pulling on the sidestick. It's very simple logic - if pilots can't see each other's actions, they should feel them on their controls. Yet, even the newest A350 still goes without it. Airbus should work on this issue before it turns into their very own MCAS.

PS Did I read somewhere that the Russian MS-21 has "backfed" sidesticks?
What a load of irrelevant tosh spouted by someone who doesn't understand Airbus flying....
If Boeing got rid of their moving throttles, it may have saved a couple of recent crashes too! Just as inane!

goeasy is offline  
Old 11th Jul 2019, 03:19
  #7 (permalink)  
fdr
 
Join Date: Jun 2001
Location: 3rd Rock, #29B
Posts: 812
Originally Posted by capngrog View Post
I just saw the linked article in the Daily Digest of the Aviation Week Network.

The link:
https://aviationweek.com/commercial-...de9291345fa795


The problem seems to have been limited to overly shallow takeoff rotations from the high elevation environment of Bogota El Dorado Airport in Columbia. Could this problem be related to lack of side stick feedback feel while trying to avoid a tail strike? It would seem to me, given adequate airspeed etc., that it would be instinctive to apply whatever pitch was required to maximize climb and terrain/obstacle avoidance.

Cheers,
Grog


The report has a great use for cleaning buttocks post bowel movements.

The aircraft discussed in the report on the BEA website, which was delegated by the GRIAA to the country of manufacture, the BEA.... states that the aircraft achieved a threshold crossing height (DER) of 6', end of CWY at 20' and cleared the LLZ antenna obstacle thereafter by 12'. It specifically states a speed of V2+9 at that point.

OK. do the maths.

If the aircraft was compliant with the regulations by any stretch of the imagination, then with 3 engines it is required to achieve 35' for the end of the CWY, on 3 blenders, having suffered a complete power loss 1 second prior to V1. In the event data, there is no power loss. The target speed in flight is to achieve V2 by 35' [25.107(c)(1); 25.111(c)(2)]. In normal operations the flight crew are targeting V2 to V2+10 on all engines to ensure that at least V2 is achieved.

The reported event gets to achieve V2+9, and 12', 20' for the DER and CWY crossing in turn. If the rotate is at a low rate, then the speed will be high, and crossing heights are reduced, fair enough, but, the aircraft is within its normal speed range, has not had an engine failure and gets to... a dangerously low crossing height of an obstacle some 800' beyond the end of the CWY. The aircraft is imparted with a certain amount of thrust, and drag. The thrust is not linear, but is near enough for a rough assessment, and the drag is a function of rolling resistance and air flow resistance, which is reasonably easy to consider. Putting both thrust and drag together, and accounting for the rotate, the resultant acceleration curve starts off with a peak at relatively low speed, below 80KIAS, and which then decays at a reducing rate of reduction until rotate occurs, at which point the acceleration reduces towards but remains above 0 [due TAS change for constant KIAS target]. If you under rotate, and the energy was there in the first place, the result is a high speed trading off the altitude. So... how much altitude are we talking about? The aircraft had to be able to make 35' wheel height on all 4 engines (100%/115%), or 87% of the TODA, at V2. That means that the aircraft has to be higher without a failure when it gets to the end of TODA, by the time it takes to cross 13% of the TODA, at a rate of climb that occurs at a steady speed of V2, from 35' in the second segment. The rules give a mandatory requirement for the OEI case for gradient, and of course the aircraft actually has 4/3 or 133% of the thrust available to meet that gradient, so the rate of climb is considerably better in the all engine case [disregard the drag from an wind milling engine, rudder input, spoiler rise, aileron input etc... which all occur in the OEI case, reducing the all engine climb rate ]. The aircraft is endowed with a certain specific excess power [dh/dt], and if it is not used to achieve a climb, then that equates to acceleration, if it really exists.



So, we are to assume then that on all 4 engines, given that we are in fact at a lower drag point at V2+9 than at V2, the additional 9 KCAS that is within the normal tolerances of the aircraft. NOTE: SRS [the manufacturers very own system design....] commands V2+10.... when on all engines, which is of course 1 KCAS higher than the speed that the unfortunate crew achieved and which are tacitly being admonished for by the BEA, and being professionally maligned for being the cause of the low crossing heights..

So... the white wash of the lack of performance attributes the causation to the flight crew, who are fully compliant with the design of the aircraft as certified, yet, the aircraft doesn't meet anywhere near the required clearance for safe flight. The clearance of the LLZ is about 3 diameters of a main wheel...

There are myriad reasons why an aircraft does not meet the certification requirement, density- humidity, actual air mass temperature at engine intake over the heat soaked runway.... wind variations, weight variations, and yes, pilot control technique. Of all of those, you will only see a low crossing and excessively high speed for the pilot technique case, and V2+9 at 12' RA, is not indicative of that case, the aircraft is about 25KCAS too slow at that point for this to have been due to the pilots under rotate.

Under rotate equates to excessive speed, so where is that speed when the crew are in fact 1 kt slower than the commanded speed of the SRS system, and are around 150-170' low on where they should have been to have complied with the rules. there is about 6 seconds of level flight gear up acceleration missing in the figures, and that shows that the specific excess thrust did not exist.

OEMs quote rotation rates that are not met in the real world, and which are alarming if undertaken. The A340 indicates the performance is based on achieving a pitch rate of over 3 degrees a second within 1.5 seconds of rotate commencement. The average in operation is much lower than that, achieving 1.75 degrees at about 2 seconds into the rotate. Any delay will increase the time to liftoff, but will have a commensurate increase in IAS at that point, and that is not evident in the performance of the aircraft; rotates are slow, but the speed is also not building up. It is necessary to get the attitude and speed outcome to achieve liftoff, the report appears to be quite silent on the speed that should have occurred given the slow rotate rate. The crew are targeting attitude rate and speed to achieve a safe liftoff, so it is not surprising that a lack of speed may give a slow rotate rate, and in this case, an inadequate rotate rate would have resulted in excessive speed, well above that recorded.

Stop bitching at pilots who are suffering from the lack of honesty in the performance figures, and start giving the SLF the performance margins that they are paying for and for which the industry has passively sat back and accepted the extent of BULL SH!T that is put into the PPMS.

I assure you that getting to 2' at the end of the runway when 427 people on your aircraft have paid to get to 185' is no pleasure, without a failure of an engine. The collective industry self denial at all levels, from the regulator, safety investigators, and airlines is demeaning.

Wake up and start smelling the colombian coffee.

For the pilot, don't expect any support from any level of your organisation, from the regulator, safety investigation body, airline or pilot union. Do learn your rights for single party taping, and comply completely with that and maintain evidence of your actions. Any party rocking the boat in a large corporation can reasonably expect to be victimised, the only question is how and when, not if. One example was an event where the crew assured data integrity, and completed independent stat decs during the flight, and took records of the resultant performance of the aircraft, which spoke directly as to causation. The data came back, and confirmed the high level of accuracy of the crews observations. The company suggested that the loading had been erroneous, yet the related cruise fuel burn penalty did not occur. The crew then asked for the company to look at other similar departures and get the data from those, and crew reports. That occurred some time later, and showed that out of a double handful of samples, none, (as in zero...) achieved better than the engine out performance. The crew in this case were asked to excuse a couple of samples as there was questions as to whether the aircraft had indeed got airborne on the grass or the concrete. Of the remainders, the majority were of a similar crossing height of the clearway as the reported event. At the end of the companies presentation of samples, none being safe or compliant, the crew was asked to agree that the samples showed no problem existed. This was in front of witnesses, with tape recordings, and copies of the samples provided. Good stuff. The national safety authority and the safety section of the regulator had all of the information provided under confidentiality, and oddly enough, the safety authority thereafter denied having any information on the matter, and then whitewashed the 15 odd events that they had information on. The deputy director of the safety authority while denying that they had any evidence of the event or events did unfortunately forward the data in a sealed envelope of their organisation, that remains sealed to this day, rather making the report of the authority, and the deputy directors formal statement to the govt ministry look shabby... The tape recordings of the meetings, and the data etc of these flights remain under lock and key.

Public safety is not enhanced by the usual response of organisations to staff that are fulfilling their reporting obligations under the law, and which may cause some inconvenience to assure the SLF of a level of safety. JUST CULTURE is a hollow industry buzz word, that is not evident in the abuse of power that occurs on a routine basis to the crews. SLF pay for a standard of safety, and companies, regulators and oversight entities that abuse the messenger are paying a disservice to the industry.





Last edited by fdr; 11th Jul 2019 at 04:45.
fdr is offline  
Old 11th Jul 2019, 05:17
  #8 (permalink)  
 
Join Date: Nov 2003
Location: Germany
Posts: 50
Hi,
great math and thoughts. Slow rotation should get you higher speed and lower altitudes....not lower speed and lower altitudes.

My thinking is, that winds may have played a role. Didn´t find the report on which the Aviation Week article is based.
Flight Safety foundations´ Aviation Safety Network says:
SKBO 112300Z 35006KT 9000 VCSH SCT015CB SCT070 13/10 A3025 RETSRA RMK CB VCSH LTNG/NW
>> SKBO 120000Z 31004KT 9000 VCSH SCT015CB SCT070 13/10 A3029 RMK CB/VCSH/NW
Takeoff seems to have occurred around 00z .... winds from 310 deg., on rwy 13... makes it a tailwind, right ?
Now, its just a question of how much of that was around. My bet would be that the 04 kt may not have been the full winds at
the specific takeoff run time.
Tailwind and high altitude....oops, that gets you into negative margins fast.
Klauss is online now  
Old 11th Jul 2019, 06:07
  #9 (permalink)  
 
Join Date: Jun 2002
Location: Geneva, Switzerland
Age: 55
Posts: 1,666
fdr very interresting contribution. Is this a he general issue with the 340 or with that specific aircraft (I guess both)?
atakacs is offline  
Old 11th Jul 2019, 06:30
  #10 (permalink)  
 
Join Date: Jun 2019
Location: Tana
Posts: 0
Originally Posted by goeasy View Post
What a load of irrelevant tosh spouted by someone who doesn't understand Airbus flying....
If Boeing got rid of their moving throttles, it may have saved a couple of recent crashes too! Just as inane!
Thank you for your concise yet informative reply. Dare I ask you to elaborate? You strike me as a reasonable and unbiased person, and as a pilot flying an Airbus I find myself invigorated by your response and striving to improve my understanding of the process and my skills. I'm sure a professional as knowleadgeable and experienced as you can explain the numerous things I got so terribly wrong in my short 25-year career.

Also, since I've never had the pleasure of flying a Boeing aircraft, I would appreciate your input on the dangers of movable throttles therein.

Thank you.
UltraFan is offline  
Old 11th Jul 2019, 09:32
  #11 (permalink)  
fdr
 
Join Date: Jun 2001
Location: 3rd Rock, #29B
Posts: 812
Originally Posted by atakacs View Post
fdr very interresting contribution. Is this a he general issue with the 340 or with that specific aircraft (I guess both)?
The issue is agnostic. It is purely dependent on the veracity of the original testing, and the variations of the real world form the assumptions that exist in the performance analysis. The "babe" is a great example from brand A, examples exist from brand B... As I have mentioned previously, look at the Kalitta aircraft at Brussels National, check how far the aircraft was down the runway before the engine actually failed. The plane was already in extremis before the engine failed, yet the report did not comment on that issue, but questioned the decision to reject the TO after the engine failure.

4 engine aircraft give glaring examples of the performance, as the all engine case is the limiting condition in most instances. 2 holers get to have such a decrement in performance with OEI, that they look great when there is no failure, yet the margins can be compromised.

Personal annoyance on this subject arises from

1. The indifference of the manufacturers, regulators and safety oversight agencies to assurance of public safety.
2. The lack of comprehension of airline management to the operational matters that occur and what a head in the sand attitude does for entity risk (as well as the SLF's)
3. The frequent lack of knowledge of the basis of the performance that the crews exhibit. Watching a 744 sagging under a full bag of gas fluff around on the runway chewing up 1500' of runway is frustrating; being in 1st class of the same type at the same point, before the wizards decide they have aligned on the centerline etc, and watching the aircraft scrape off the end of the runway from a passenger seat is annoying. Watching a crew take a runway that has WIP reducing the runway below the length required, and not being able to intervene, quite alarming.
4. An industry that shoots messengers, hardly a way forward to excellence or achieving the buzz words of "safety first", and the personal costs to the messenger is devastating.
5. Being personally threatened by a party that is an accountable manager is a sobering experience, and removes any doubts about the ethical standards of the industry.

When the industry gets away without comment when attacking QA inspectors, safety officers, and crew reporting safety issues in accordance with their designated duties, in accordance with regulations, the disregard of the moral high ground is quite sickening.

fdr is offline  
Old 11th Jul 2019, 10:00
  #12 (permalink)  
 
Join Date: May 2002
Location: mighty Blighty
Posts: 286
Great analysis FDR. it does seem to be a lot of ‘looking the other way’ when some performance issues are analysed for certification. 35ft looks OK upon graphs or performance text books, but not so nice to see from outside aircraft at DER.
goeasy is offline  
Old 11th Jul 2019, 11:26
  #13 (permalink)  
 
Join Date: Dec 2001
Location: Leeds, UK
Posts: 281
could they extend the runway? If this is the only daily/weekly flight that needs it perhaps it's not worth it, but plenty of airports have extended runways so they can accommodate larger aircraft and open long haul markets..

G
groundbum is offline  
Old 11th Jul 2019, 12:03
  #14 (permalink)  
 
Join Date: Mar 2019
Location: Canada
Posts: 72
Originally Posted by groundbum View Post
could they extend the runway? If this is the only daily/weekly flight that needs it perhaps it's not worth it, but plenty of airports have extended runways so they can accommodate larger aircraft and open long haul markets..

G
Air France "solved" the problem by switching this route to the 787 instead.
futurama is offline  
Old 11th Jul 2019, 13:01
  #15 (permalink)  
 
Join Date: Feb 2017
Location: Isla Grande
Posts: 958
Thx fdr, great analysis.
Even I loved to fly the A340 (very quiet on the flight deck, stable flight path, etc.) the T/O performance caused some grey hairs...
I remember the takeoff performance calculations on the -200 were three times corrected (too optimistic) by AI resulting in PTOW over 10 tons less ...
gearlever is offline  
Old 12th Jul 2019, 14:24
  #16 (permalink)  
 
Join Date: Nov 2000
Location: Canada
Posts: 546

Artificial runway lengths avert long A340 take-off rolls

  • 12 July, 2019
  • SOURCE: Flight Dashboard
  • BY: David Kaminski-Morrow
  • London
Air France and Lufthansa introduced artificial reductions of runway length at Bogota as a precautionary measure after incidents involving prolonged take-off runs by Airbus A340s.

French investigators have disclosed the carriers' actions in an analysis of a serious departure incident from the Colombian capital in March 2017.

The A340-300 started rotating at 142kt some 2,760m from the threshold of runway 13R, which is 3,800m long.

But it did not lift off for another 11s at which point it was only 140m from the threshold of the opposite-direction runway 31L, crossing it at a height of just 6ft.

So low was the jet that the supplementary first officer asked the captain whether the wheels might have snagged the localiser antenna, which the aircraft had overflown at 12ft with "close to zero" vertical speed, says French investigation authority BEA.

The take-off distance was nearly 990m longer than that defined by performance models, and BEA says this exposed the aircraft to increased risk of excursion or obstacle collision.

"Simulations carried out by the manufacturer showed that the rotation rate…was significantly lower than that assumed for certified performances, and that this was the main contributing factor to the longer take-off distance," it adds.

Air France and Lufthansa, both of which served Bogota with A340-300s, conducted a joint analysis of rotation techniques in the aftermath of the incident.

The two airlines compared their flight data on Bogota departures – a total of 1,900 from runway 13R for Air France and 400, from all runways, for Lufthansa.

Lufthansa had previously looked into rotation issues after a take-off incident in Johannesburg in 2004 and long take-off runs in 2007, 2011 and 2012.

Its studies showed an average continuous rotation rate of 1.9°/s rather than the 3.1°/s required by the A340-300 performance model.

But BEA says that Lufthansa has resisted specific training to adapt take-off techniques, in order to avoid a possible increase in tail-strike risk.

For the combined Bogota study the two airlines showed similar average rotation rates of around 1.8°/s and average rotation times of around 7s.

These discrepancies between the theoretical performance model and the actual situation in operation effectively increase the take-off distance by 200m.

Air France's flight-crew techniques manual had specifically cautioned against low rotation rate. It pointed out that rates of 2-3°/s would have "minimal impact" on the take-off run, but rates "significantly" below 2°/s "should be avoided".

Airbus last year revised its manual for the type, quantifying the effect of a 2-3°/s variability in rotation rate as translating into a 300m extension of the take-off run.

Both Air France and Lufthansa have introduced a restrictive measure for Bogota departures, an artificial reduction of the runway length for performance calculations in order to provide additional safety margins.

Lufthansa has implemented a 280m reduction in available length for A340-300s.

About a month after the Bogota incident, Air France put an initial 200m hypothetical runway restriction in place, and amended departure procedures from the airport to require full-thrust, including reaching 50% of thrust while on brakes.

BEA says the 200m reduction was calculated to take into account the most critical identified situation, and to ensure compliance with regulatory margins.

This limitation was subsequently increased to 380m in order to account for the possibility of engine failure. The carrier also extended the take-off thrust requirements for A340 departures to all airports served by the type.

Air France has set up specific training courses on A340 rotation techniques, intended to instil best practice set out in crew manuals, reduce the variability of rotations, and increase the awareness of crews to tail-strike or performance risks.

Initial results of this training effort, covering 600 A340 flights, showed that crews had been changing their rotation techniques, lifting the average rotation rate to 2.2°/s.

Air France's implementation of precautionary measures have enabled the carrier to reduce take-off distances – although BEA points out that they still remain higher than those of the theoretical model.
Longtimer is offline  
Old 12th Jul 2019, 17:24
  #17 (permalink)  
 
Join Date: Jun 2019
Location: Tana
Posts: 0
This article on Flight Global is very interesting. After all those incidents they are still taking off on less than full thrust? This is very strange. If you know that several crews had problems rotating from an airport, why not just give it full beans and hold the brakes longer? Why wait for the company procedure update to do the obvious thing which is also much safer? And if the company had those problems in 2007 why is the procedure only amended now, twelve years later? AF and LH are not some third-world ULCC, sure they must react sooner.
UltraFan is offline  
Old 12th Jul 2019, 17:30
  #18 (permalink)  
 
Join Date: Feb 2013
Location: 60 north
Age: 56
Posts: 19
I was once told by my brilliant boss Krazy Bill, that the Cessna 185 if it floated AND got airborn: It will climb!
( Ca 10 to 15% overloaded he figured, depending on floats, Edo or Caps)

AS opposed to the the Otter that was unsinkable and if You got it on the step overloaded would stay in ground effect and refuse to climb!

Key to this drift is ACTUAL weight , ground effect and the guaranteed effect of OEI.
Regards
Cpt B
PS
Gona tell my AME in august: I am 95 kilos including winter uniform and flight bag, no need to step on the scale!
DS
BluSdUp is offline  
Old 12th Jul 2019, 18:26
  #19 (permalink)  
 
Join Date: Jun 2019
Location: VA
Posts: 210
fdr, great analysis. There have been times Ive lifted off pretty close to end of runway and wondered what it would have looked like if I lost a motor. Pretty scary.
Tomaski is offline  
Old 12th Jul 2019, 19:52
  #20 (permalink)  
 
Join Date: Aug 2003
Location: FR
Posts: 222
For sure, the BEA report issued in July needs to be read more than once (which I have not done yet).
On page 46 they plot AF and LH pitch rate, always slower than assumed by Airbus. They do not explicitely show, in this particular plot, where the incident event is sitting.
Since this is really performance-limited takeoff, one would hope (expect) a table when all uncertainties would have their contribution estimated (for example: CG position ?)
I was not yet able to see in the report where exactly they put the exact data from the incident flight (pitch control) into the model, and they confirm the agreement betw performance model and actual performance?
The Report also mentions a software tailstrike preventer, introduced in 2008, (common to which other a/c type??) taking inputs from the pitch angle and from the radio altimeter (...) but not mentioned in the Flight Manual excellent !
The Report seems to conclude that pitch rate at rotation is main factor, but this was not known (not told) to pilots before this incident.
Personally (not a pilot!) I can imagine that if rotation is too slow you will lose some energy, no?
And also the Report shows the stick input during the incident ... with an unexplained reduction for about 2 seconds or less .. which cannot help, but the impact is not quantified exactly.


pax2908 is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service - Do Not Sell My Personal Information -

Copyright © 2018 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.