There have been a couple of fatal incidents involving Airbus aircraft, namely Indonesia AirAsia Flight 8501 and Air France Flight 447, that could have been prevented had the captains orders / input been the only accepted one. In both cases the captain was commanding push down to recover from a stall, while the FO was pulling up. Because of the dual-input system they cancelled each-other out and the stalls were not recovered from.

Has Airbus done anything to address this? It seems to me that if the aircraft is in alternate law, as it was in both cases, perhaps it should only accept input from the left seat.

  • 7
    $\begingroup$ You mean besides each side's priority switch, never mind CRM? For what it's worth, there have been many accidents where the captain has been in the wrong, too, so clearly "blindly trust the captain's input" wouldn't be the correct approach either. See Who has the final authority over sidesticks? and Why is the Priority Takeover button used? and probably some others. $\endgroup$ – user Sep 27 '18 at 8:57
  • $\begingroup$ @MichaelKjörling It was just an example. I don't know about the priority switch. $\endgroup$ – Cloud Sep 27 '18 at 9:11
  • 1
    $\begingroup$ "Only accept input form the left seat" is a very bad idea. What if the reason the airplane is in alternate law is an accident that not only damaged the airplane systems but also knocked the captain unconscious? Also, the assumption that the captain is also the pilot flying is on thin ice. Yes, in case of unforeseen circumstances, often the most experienced pilot will take the controls, but … who is to say that the captain is actually the most experienced pilot for this particular type of aircraft in these particular circumstances? Also, what if flying the aircraft isn't actually the most … $\endgroup$ – Jörg W Mittag Sep 27 '18 at 9:38
  • 1
    $\begingroup$ … demanding task at the moment, but rather figuring out what the hell happened? In that case, you want the most experienced person (which might be the captain) to work on figuring out the problem, and "the other guy" flying the plane. What if the left seat is unusable as a result of the accident? $\endgroup$ – Jörg W Mittag Sep 27 '18 at 9:39
  • $\begingroup$ Adding to @JörgWMittag's comments above, consider British Airways Flight 5390 in June 1990. If that airplane had had control stick input addition similar to that of FBW Airbuses, the copilot could just have pulled his stick full aft while the captain was wedged against the left side controls to maintain a roughly neutral pitch. Instead of a critical, unanticipated descent, it might have been more like the trim being stuck at a nose down attitude setting, even without priority override. $\endgroup$ – user Sep 27 '18 at 14:17

Nothing was changed in the stick design or logic.*

By checking both accident reports, neither list the side stick design/logic under the final recommendations (of the issues were the training and lack of CRM**). Recommendations toward the manufacturer, for example, are with regard to the barrage of warnings the crew of Indonesia AirAsia had:

The KNKT recommends that Airbus to consider in developing a means for flight crews to effectively manage multiple and repetitive Master Caution alarms to reduce distraction. (PK-AXC accident report)

But as far as I know, Airbus aircraft already have an 'EMER CANC' button to mute spurious alerts.

Coupled digital side sticks weren't there when Airbus first introduced the A320, as a technology they became feasible in the early 90s. And were first introduced in c. 2015 on the Gulfstream 500 and 600 business jets.

Imagine the following scenario, you are flying with a friend in a light plane with traditional yokes and suddenly you face a huge mountain, your options are to turn left or right. You decide left, and turn the yoke, but the yoke won't turn because your equally strong friend decided to turn right. What do you do? Communicate by simply stating, "I have control," and hearing your friend's confirmation, "you have control." This is as simple as it gets and how good pilots should be trained, training ingrains those basic responses.

An equally panicked and under-trained pilot in an active sticks cockpit might think their stick got jammed or the coupling motors are acting up. And seeing how long it took for the active sticks to make it into a business jet already hints at the snail pace (and cost) of aviation hardware certification and approval in general.

As for the captain's / left seat inputs being correct, as others have commented, there have been numerous accidents where the captain was in control and the first officer's recommendations were not heard or they kept silent, again bad CRM.

* As others have commented, there are already buttons to take priority, as well as aural and visual cues as to which seat currently has priority.

** Example of such recommendation:

EASA ensure that operators reinforce CRM training to enable acquisition and maintenance of adequate behavioural automatic responses in unexpected and unusual situations with a highly charged emotional factor.

  • 4
    $\begingroup$ If I recall the AF447 report accurately, it noted explicitly that the left-seat pilot would not have had a way to discover that the right-seat pilot was pulling up on the side stick. In light of this, it is difficult to see what the relevance of your "the yoke won't turn" example is. Is your answer that Airbus addressed the issue by introducing feedback in the sidesticks? If so, it would be beneficial to state so explicitly. $\endgroup$ – hmakholm left over Monica Sep 28 '18 at 15:49
  • 2
    $\begingroup$ The question was not what was or wasn't recommended, but whether Airbus did anything irrespective of external recommendations. If there is no feedback in the sticks, then what is the point of your example about an aircraft that does have feedback? $\endgroup$ – hmakholm left over Monica Sep 28 '18 at 15:54
  • 2
    $\begingroup$ The big difference is your friend could "feel" you pull left and decide to let go of the yoke without an audible... Airbus cleverly decided to not mimic real mechanically connected flight controls in their infinite wisdom... Yes, well trained pilots should communicate, but as we've seen in the listed accidents, that does not always 100% without fail occur. What does happen 100% of the time with physically connected flight controls is receiving haptic feedback, possibly coinciding with audible communication. $\endgroup$ – SnakeDoc Sep 28 '18 at 17:06
  • 2
    $\begingroup$ @ymb1 you have a good answer, I'm just pointing out the mechanical connection is pretty valuable. Even if you're hands aren't on the yoke, you can visually see what the PIC is doing. The dead stick thing seems just plain wrong to me... but hey, Airbus sells a lot of planes. $\endgroup$ – SnakeDoc Sep 28 '18 at 17:17
  • 1
    $\begingroup$ @ymb1: your example designed to show that a linked yoke is not necessary is badly chosen because it actually demonstrates the opposite of what you want it to demonstrate, viz despite bad CRM the "2 strong friends facing a huge mountain" immediately realize the conflicting dual input, which is not possible with unlinked sidesticks in a dark cockpit as evidenced by AF447. This aspect was explicitly mentioned in the accident report. $\endgroup$ – summerrain Dec 9 '18 at 20:13

It's not clear that there is an issue.

Is the lack of coupling actually a lack?

The lack of coupling between the two side-sticks to provide feedback between the pilots is, intuitively, a lack in the interface.

We are used, in the physical world, to being able to feel what another person is doing through the things that we're manipulating, the machinery and tools we operate together.

However, flying a plane is not like carrying a ladder up a narrow staircase, or riding with someone on a tandem bike. That kind of implicit bodily feedback isn't part of the Airbus cockpit system.

Cockpit communication is supposed to be explicit

Instead, the communication is supposed to be explicit, so the pilots say what they are doing when operating the controls. Introducing physical feedback into that environment might seem like an enhancement, but in fact it would be the opposite: it would be the introduction of a source of doubt, that draws habit and practice away from the explicit communication around which the system has been designed, towards an implicit form of communication.

Even implicit communication requires verbal confirmation

It would then be necessary to introduce also a new system under which the pilots would decide whether their explicit verbal communication is now in operation ("I have the controls"), or the implicit physical communication (you feel the other pilot doing something).

Now there is a new burden! As well as deciding what they should be doing with the controls, the pilots must also decide whether they will be prioritising explicit/verbal or implicit/physical communications - two things to manage.

And working that out will require explicit verbal communication, so they may have well just stuck with that in the first place.

And if implicit physical communication is allowed to have weight, they still need explicit verbal confirmation of it, because that's what pilots always do. By the time I have asked my co-pilot "WTF are you doing with the stick and why?" it's already later and more stressful than if they had said in the first place that they were taking the controls to climb over an unexpected mountain.

It's not a lack, it's the removal of something unwanted

Implicit communication through the side-stick is not a free gift of additional useful communication. In the Airbus cockpit system, it's the introduction of something dangerous and unnecessary, that would require extra cockpit management and would harm communication in the cockpit.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.