Free Flight, Free Speech

I just reviewed all the accidents listed on Wikipedia, and made a list of what seems to be the causes:
Bear in mind this is only what is on wikipedia

A319: 0 fatalities, 5 incidents listed

• Two on ground collisions: one caused by DC9 pilots, one by MX personnel taxing the aircraft

• Two runway mishaps with broken gear: one caused by windshear

• One cowling separation: doors not latched by MX, bad walkaround

Primary cause: Human Factors, Weather
So, EXCELLENT safety record. 0.0034 accidents per airframe, 0.00 fatalities per airframe, and 0.22 accidents per year.

A321: 377 fatalities, 5 incidents listed

• Transasia 543: collision on ground with truck

• Airblue 202: CFIT. Arrogant captain, poor CRM, weather, not following SOP, ignoring GPWS (That one is really sad: the captain was stupid and killed 152)

• Lufthansa 1829: frozen AOA sensors confused the FBW system, forcing a pitch down which could only be recovered with constant aft input. -- this is the first one where I admit you are correct, KL-666 but so far, not bad

• Metrojet 9268: under investigation, probably terrorists bomb aircraft

• Dalla 159: suicide bomber kills himself, but noone else.

Primary cause: Terrorism, human factors, sensors
So, not quite so good as A319, but still only only 0.0032 accidents / airframe, and 0.24 fatalities per airframe, and 0.22 accidents / year.

Now, the big one:

A320: 27 incidents

• Air France 296: Too low and slow, poor preparation, disuputed: FBW error (teething trouble?)

• Indian 605: confusion over automation (poor training?), CFIT

• Air Inter 148: confusion, turbulence causes autopilot flaw (teething trouble)

• Lufthansa 2904: Weather, hydroplaning, design flaw in ground spoilers / reversers

• Phillipine 137: overrun, thrust left at CLIMB THRUST: pilot error

• Gulf 072: not following SOP or CRM: CFIW. Disorientation: "believe your instruments

• Iberia 1456: Windshear, design flaw

• Jetblue 292: nose gear jammed

• Armavia 967: fatigue, stress (captain getting angry that he can't land?), weather

• Tam 3054: same as Phillipine 137

• Phillipine 457: overrun, misjudged landing

• TACA 390: wet runway, fast landing

• XL888T: pilot error -> stall, AOA sensors frozen so could not prevent stall

• US 1549: BIRDS!

• Gulf 270: weather caused overrun

• Syrian 501: in-air collision

• Cebu 971: lack of night non-precision low-viz approach experience

• US 1702: RTO after rotation

• AirAsia 8501: stall, after captain disabled the FACS (removing stall protection)

• GermanWings 9525: belived to be murder/suicide

• Air Canada 624: pilot error

• Asiana 162: poor weather, possible windshear

• Turkish 1878: hard landing: collapsed gear

• Egyptair 181: hijack

• Egyptair 804: terrorism believed to be the cause

• Afriqyah 209: hijack

Primary Causes: Teething troubles up to the early 2000s, then mostly pilot error, weather, or terrorism

So really, it looks like there haven't really been that many accidents caused by the fly-by-wire: there have been incidents where the protections were disabled, which allowed the accident to occur.

So, after the teething troubles were sorted, it looks like the A320 is just as safe as any other aircraft, and not a Scarebus!

I think the main problem is when the protections are lost. Some airlines do not teach A320 pilots that the Airbus can be stalled, even in normal mode, especially in alternate mode. As far as I know, Qantas taught this from the very beginning.

Lego, that was some scary post. I'll still love flying thou

123apple wrote:it looks like there haven't really been that many accidents caused by the fly-by-wire

Not directly, that is correct. If normal law keeps working everything is fine.

123apple wrote:I think the main problem is when the protections are lost.

This is exactly the problem i mean with loosing confidence in FBW. It sometimes goes in a degraded law, and history shows that many pilots can not handle the plane then. Sure one can say the pilots must be trained for different laws. But why train for 3 different aircraft? And even when well trained, it will be hard for a pilot to switch mentally to another aircraft, especially because the law switch always happens at a high workload when the plane is already in trouble.

The mental differences between the laws can be huge. In normal law you get used to: pull is up and push is down. When getting in trouble, just tell the plane to get you out of there by pulling to the max. The plane makes sure you stay within the envelope. It will even descend a bit if it needs to pick up some speed. But in a degraded law you can not just go pulling to the max. You will have to fly yourself with very gentle control inputs. How often do pilots get that experience, except in a few training sessions?

No, to me that whole concept of switching laws is flawed. I think it is a lot more safe when a plane behaves in one way, whether there is trouble or not. Then the pilots can use their muscle memory for flying, and mentally concentrate on solving the problems.

Edit:
Having said all this, the ultimate consequence i have to accept for wanting planes always to fly the same, is that i have to reject the Airbus normal law, as it is too far away from normal flight (the other laws).

Kind regards, Vincent

IAHM-COL, this proves the A320 is relatively safe, particularly the A319... by statistics, so don't worry

Kl-666 thanks for the reply

123apple wrote:Kl-666 thanks for the reply

Well, thank you for asking the right questions, so that i find out about the hiatuses in my word choice, and get a chance to explain myself better. Being able to ask the right questions is a gift you have. I like that.

Edit:
Just elaborating a bit more. Why do we discuss at all on these forums? Is it to look smarter than the one we discuss with to others? I often have the impression that this is what we do. Or is it to really try to understand what someone tries to say? If it is this, then asking the right questions to get the best out of someone is a great asset.

Kind regards, Vincent

123apple wrote:IAHM-COL, this proves the A320 is relatively safe, particularly the A319... by statistics, so don't worry

I don t follow Lego.
Do you mean, the A3xx is safe, or the pilots flying it had somehow managed to keep themselves out of troubles?

I presume, the pilots had done a good job. I maybe wrong here.

Just elaborating a bit more. Why do we discuss at all on these forums? Is it to look smarter than the one we discuss with to others? I often have the impression that this is what we do. Or is it to really try to understand what someone tries to say? If it is this, then asking the right questions to get the best out of someone is a great asset.

Sorry, I don't follow you. What are you trying to say?


IAHM-COL, by statistics, the A320 is a safe plane But I agree that the main factor is human: the A320 when understood and operated properly is really quite easy to fly and easy to "keep out of trouble". If the pilots are trained properly ("you can stall any time by pulling aft stick"), you shouldn't have to worry about it.

123apple wrote:Sorry, I don't follow you. What are you trying to say?

I am even more sorry, because i do not understand which part you do not understand. Maybe you can give your interpretation for me to understand better?

I'll kick off with a general explanation: There are several motives to get into a discussion. Some motives i find not very productive, other motives i do find productive. With the productive motives, asking the right questions can even enhance productivity.

Kind regards, Vincent

The issue that most of us have with airbus in general and not FBW is the way it was implemented. Simply put, it's needlessly complicated. They only have systems in case the first one failed, and another in case THAT one failed. So its piling systems on top of systems making the needlessly complicated system even more needlessly complicated, and on top of which sometimes seem to completely forgo the human being part of the equation.

It's like car doors. All you need is a lever. Pull the lever the door opens. Some cars nowadays however, have electronic switches that you trigger to open the doors. So for example, in cases like when the car is moving, the door cannot open even if you trigger it. OK, "safety". So far so good. But what if you want it to open in while moving? What if you're being kidnapped? What if you need to jump from vehicle to vehicle?? And even if you're not being kidnapped or not a secret agent or have any valid reason to open doors while moving, what if the electrics fail and you're stuck inside?

Ahah, the manufacturers say. That's why they have a secondary 'fail safe' LEVER to open the door in an emergency.

Err, why not do that in the first place?? They just made things needlessly complicated. But meanwhile, people have died suffocated in their cars because they did not know about the failsafe to open the door, or they get into an accident and the electrics conked out, leaving them trapped when they need it most to get out. (In an emergency, last thing you have is a level headed mind to find the stupid hidden fail safe ).

And that's just as simple as a car door. We're talking about airplanes here, the list of whatifs just grew alot longer.

On the other side of the coin, I suspect as reliability improves and eletronics / computers / actuators /devices / understandings / AI improve, then it will come to a point where it may be as good as simpler mecahnical set up, but it's the in between that time that I don't want to be a lab rat for these things.