Airbus controls are unsafe

[ChrisCooper]

People often forget that modern Boeings are also fly by wire, the 777, 787 and 747-8 all are, and the 747-400 has FBW throttle and trim. The actual set up is different, Boeing use a yoke and direct control (i.e control surfaces move proportional to the pilots input) with soft envelope protection (it can be overridden by extream force on the controls), Airbus use sidesticks which select pitch and roll rate (normal law anyway) with hard envelope protection (again normal law). Envelope protection prevents the pilots doing anything that is likely to damage the plane or cause it to stall. Airbus in Alternative or Direct Law are much more conventional though, i.e control surfaces move in direct proportion to sidestick movement.

Both systems have some flaws, and for ever Airbus accident where a Boeing might have survived or Boeing survival that might have killed an Airbus, there is a Boeing accident where an Airbus might have survived or Airbus survials where a Boeing would likely have crashed. The classic view is that if a Boeing gets into trouble the pilot is more likely to be able to get it out of trouble than an Airbus, but an Airbus is less likely to get into trouble in the first place compared to a Boeing.

The thinking behind the Airbus system really is that in the event of an incident where the pilot needs to make an extream maneuver, do you want the pilot to be limited by the fly by wire to keep just within the limits of the plane, or do you want them free to either overstress the plane and cause it to break up or not fly close to the extream to avoid overstressing it and be unable to carry out the maneuver required in time? The Boeing philosophy only really works in cases where the plane can only maneuver quick enough or recover by exceeding the limits but not by enough to critially damage the plane (or cause the crew to black out).

 

[Old Timer]

CC
Fair points to a degree, but the problem with the pitot heads icing up was well known and not acted upon on the A330 series despite these being the very a/c that flew in the sort of conditions where this was likely !

There have been a number of incidents involving the Airbus FMCS doing "strange" things.

Taking manual control of the a/c requires the pilots to take command in precisely the situation that the Airbus FMCS does not allow / like in order to get the a/c to a condition where it can be flown manually whilst the FMCS is reset, now standard procedure since the AF447 incident I am told where there is any doubt as to how the FMCS is managing.

I prefer to take my chances with the Boeing system and the flight deck crew rather than a system that is responding either to inaccurate inputs (which defeats the purpose) or has decided that black is white - memories of one incident where the Airbus FMCS decided the a/c was flying backwards when it encountered strong headwinds, albeit many years back.

 

[futureA]

This debate is one that often comes up on the internet. Although I belive that the allegation that the airbus FBW system is 'unsafe' is false.

There are two things to remember. All modern planes use computerised fly by wire in some capacity. Everything from private jets, to boeing planes such as the 777, 787 and the 474-8. The difference between the airbus system and they systems used by other planes is the flight envolope protection, as explained by ChrisCooper. The airbus flight envolope protection will stop the pilot from making any manovres that will injure the passengers or cause the plane to be damaged. The difference on boeing fbw planes is that the envolope protection can be overridden (with a great amount of force).

It is also worth noting that before this incident, the a330 had not had any passenger fatalities in its 16 year history and considering that over 650 have been delivered (with another 400 on order) that is a prety good saftey record.

It is also worth remembering that the control system used in the boeing 737 is not infalable. Last Febuary, a turkish airlines plane crashed in Amsterdam on aproach to the runway.

A preliminary investigation found that the crash was caused primarily by the engine's automated reaction which was triggered by a faulty radio altimeter, which had failed twice in the previous 25 hours. This caused the autothrottle to decrease the engine power to idle during approach and the crew noticed this too late to take appropriate action to increase the throttle and recover the aircraft before it stalled and crashed. Boeing has since issued a bulletin to remind pilots of all 737 series and BBJ aircraft of the importance of monitoring airspeed and altitude, advising against the use of autopilot or autothrottle while landing in cases of radio altimeter discrepancies.

Source

Regarding the air france incident and this tv programme, the only criticism of the airbus control system was that the throttles did not move automaticaly (although the information regarding engine power was displayed on a monitor) and it was also suggested that the main reason this crash happend is because this incident would have been difficult for any comercial pilot.

Since airbus began in the 1970's they have been known for introducing many inovative technologies that are standard accross the industry nowadays. Today airbus sells (more or less) half of the worlds comercial airoplanes and the A320, A330 and A340 programes have over 2000 planes flying around the world every day. Whilst they obviously a major player, i doubt that they have the influence or money to bribe the worlds aviation authorities into allowing them to continue to sell planes with unsafe control systems.

I am perfectly happy to fly on any airbus aircraft (and do often) and i believe they are a quality product and safe.

 

[Old Timer]

futureA
An interesting response from an obvious Airbus fan and I presume in answer to the original post, which was somewhat sensationalist.

I notice however that you never addressed the issue of the Airbus FMCS system about which there have been many reports of issues with the which have resulted in incidents rather than fatalities.

I do not think the AMS crash is totally comparable as the actual cause was the failure of the crew to notice the falling airspeed through the engine power reduction whilst on finals. There is certainly sufficient existing evidence to make a strong suggestion that AF447 lost control through the pitot heads icing up, an issue (known about) for which there was an alternative, unfortunately not French.

As far as the throttle controls are concerned, they were certainly a factor (amongst a number of others) involved in the loss of a TAM A320 at Congonhus airport in Sao Paulo a few years back.

 

[futureA]

futureA
I notice however that you never addressed the issue of the Airbus FMCS system about which there have been many reports of issues with the which have resulted in incidents rather than fatalities.

I accept that any incident where something is not working as it should is worrying. However, I do not believe that this makes airbus aircraft unsafe, or even less safe than other manufacturer’s aircraft. It is also worth remembering that similar incidents have happened on planes made by other manufacturers.

futureA
I do not think the AMS crash is totally comparable as the actual cause was the failure of the crew to notice the falling airspeed through the engine power reduction whilst on finals.

Obviously not identical to flight 447 but the Turkish incident was one where the aircraft behaved unexpectedly and the pilots took incorrect/ insufficient action.

futureA
There is certainly sufficient existing evidence to make a strong suggestion that AF447 lost control through the pitot heads icing up, an issue (known about) for which there was an alternative, unfortunately not French.

The BBC programme suggested that the situation was recoverable, and the pilots probably regained some control.

futureA
As far as the throttle controls are concerned, they were certainly a factor (amongst a number of others) involved in the loss of a TAM A320 at Congonhus airport in Sao Paulo a few years back.

I think that there were a few much more serious factors in this incident. Maybe this aspect of the control system is one that could be improved, maybe it isn’t. But any system can always be improved. However, I am sure there is a reason why it was designed to behave in this manner.

Aviation is one of the most safety conscious industries on the planet. Everyone involved has to comply with numerous regulations. Regulators have the power to order redesigns/ replacements of components or systems and ground aircraft that are perceived to be a safety threat. They have exercised this right many times in the past.

I believe that if the authorities thought that there was a serious problem with the 6000 or so airbus FBW planes in operation then airbus would have been ordered to do something about it a long time ago. Let us also not forget the legal and financial implications of selling products with known lethal design flaws.

I hope I have answered your questions without sounding arrogant

 

[CosherB]

The Turkish Airlines crash at Amsterdam was pilot error, of a gross kind. Standard Operating Proceedures OR basic flying skills would have prevented it; neither were applied.

SOPs stated that with a faulty radalt, the auto throttle should be disengaged; it was not, and so was allowed to cause the aeroplane to attempt to land at 1000'. And also that if the aeroplane was not on a stabilised approach by 1000' above the runway, at the correct speed and configuration, and on glideslope and on localisor, then a go-around should be performed. It hadn't met any of those criterea, yet they continued the approach.

As it is, not only were SOPs ignored, but two experienced pilots and one rookie pilot watched the aeroplane pull off the power at 1000' and begin to flare (pitch up). They let it fly itself right into a full stall far too low to recover! Crickey, what were they smoking?

Here's an animation of that accident:

http://link.brightcove.com/services/player/bcpid18065281001?bclid=0&bctid=83635223001

My view, and it's only my view without knowing all the facts and therefore not a definitive statement, of the A330 accident is that the holes in cheese began to line up whan the wx radar didn't show the 'big storm' behind the little one. The big storm produced big turbulence, and then they lost the IAS. They delayed establishing 'power and pitch for speed' probably because that was not easy in the turbulence, and the throttles being already in the cruise power position (though the AT had selected a significantly lower turbulence-penetration power setting) probably added to the confusion. I think they stalled, maybe more than once, and impacted in a deeply-stalled condition.
--- old post above --- --- new post below ---
Having looked a bit further into this, apparently the AF Airbus had climbed to height early for max range to avoid a refuelling stop in the Azores. It was also very heavy because it was carrying a full fuel load. At that weight, the 'coffin corner' margin between stall and mach buffet was about 15 knots. That's easily handled by the aircrafts computer controls, but is marginal for the pilots to fly manually even in smooth air and good visibility.

At night, in severe turbulence, they had no chance.

The question that raises is why did AF flight-plan a flight where, if the automatics failed, the aeroplane would almost certainly crash, at least during the early part of the flight (as fuel burned off and the aircraft got lighter, 'coffin corner' would have got more manageble)?

 

[Old Timer]

CS

From what I picked up, at all up weight, and with the required fuel reserves they cannot make GIG - CDG, so therefore they file for Bordeaux (BOD) with CDG shown as an "alternate".

They then go for FL350 asap so as to reach the Coast of Africa with sufficient fuel to reach CDG within the minimum reserves, and request a diversion to CDG.

My view is that this partly explains why they went direct through the ITCZ with the weather as it was.

I know that the KLM crews take on more fuel and fly further north and around, rather than through, the ITCZ when the conditions are bad.

KLM appear to be more receptive to their Captains judgement and routing/fuel changes.

 

[cyclebytrain]

Before I say anything in this thread I should declare an interest -designing aircraft control systems is my job. However this means that I can actually say less than I could about many other topics for obvious legal reasons -there is an ongoing investigation and much of the knowledge that I have about Airbus systems is covered by a non expiring NDA (an Airbus employee would be able to tell you more ironically). Having got that out of the way, I want to make a couple of points:

The first point is that the more information you can provide a pilot and the more intuitively, the better. Moving throttles is an obvious one. There are others not so obvious such as 3d audio for warning indicators, etc. There are certainly Airbus engineers who are aware of the lack of feedback to the pilot relative to Boeing competitors.

The second point is that whilst AD 2009-0195 requires the replacement of Thales C16195AA outright and C16195AB from position 1 & 3 the reason for this is that there is more experience with the Goodrich 0851HL. It is not established that the Goodrich 0851HL is superior to C16195AB. Water ingestion with C16195AA had been observed on A320, rather than A330/340.

And whilst hindsight is a wonderful thing, Air France had started replacing the Thales AA about five weeks before the accident in question (I cannot remember if they were changing to the Thales AB or Goodrich). The impression of some airlines being better at safety than others is mostly a myth -all that are allowed to fly in European airspace are safe.

Having said that I don't think it would be wise for me to comment further on any specific working practices of any airline.

 

[Failed Unit]

This site has interesting information if people want to read it.

http://www.airsafe.com/plane-crash/air-france-flight-447-airbus-a330-atlantic-ocean.htm

Take from it what you want.

 

[flymo]

At least they could get the reg of the aircraft right. F-GZCP.