Why are Airbus side sticks asynchronus?

December 4, 2015 12:37 PM

Does anyone know why Airbus designers decided not to have any tactile feedback between the pitch roll sidestick controllers?

After reading the AirAsia final crash report I was struck once again by one aspect of the flight control system that I find a inexplicable: the joystick/sidestick controllers move independently. On the face of it it seems like a poor design from the human factor side of things but I reserve judgement as I assume that the designers were very smart engineers and that there must have been a reason to have unlinked controls. All I can think of is that the threat of an incapacitated pilot slumped over the side stick or mechanical jam was seen as more likely than the break down in CRM and skill that occurred in both the AirFrance and AirAsia crashes. So my question is not about good vs bad but why?

posted by Pembquist to Technology (6 answers total)

This has actually been a common criticism of the Airbus cockpit design philosophy since they started doing it. If I remember correctly, part of the rationale is that the pilots should be doing very little hands-on flying. When they do put hands on controls, one pilot should be the one playing with the joystick (pilot flying) and the other should be controlling radios and the other electronics in the cockpit (pilot not flying).

(Again, I'm going off memory here) I think there's a switch in the cockpit that tells the Flight Management System who is pilot flying, so it prioritizes inputs from that stick. There is a certain failure mode that reverts the sticks back to direct control, but in such a way that the left seat controls the left-hand side of the plane (left wing aileron and left elevator) and the right seat controls the right-hand side of the plane.

I believe these decisions were made during the development of fly-by-wire - the flight controls don't need to be physically connected in that case, and the flight management computer doesn't need to do something like physically move the yoke to automatically prevent a stall as in older designs (after the stick shaker activates, the computer will attempt to recover from the stall condition on its own in certain cases). It was probably decided that training could solve a lot of these problems, and operating limitations prevent the pilots from deliberately putting the plane in attitudes that cause this kind of response to begin with.
posted by backseatpilot at 1:24 PM on December 4, 2015 [2 favorites]

Response by poster: backseatpilot, the sticks can be deflected in different directions simultaneously and the computer will sum the inputs. Both sticks have a side stick priority button which works like this:

Sidestick priority logic
‐ When only one pilot operates the sidestick, it sends his control signals to the
computers.
‐ When the pilots move both side sticks simultaneously in the same or opposite
direction and neither takes priority, the system adds the signals of both pilots
algebraically. The total is limited to the signal that would result from the
maximum deflection of a single sidestick.
Note: In the event of simultaneous input on both sidesticks (2° deflection off the
neutral position in any direction) the two green SIDE STICK PRIORITY lights on
the glare shield come on and “DUAL INPUT” voice message is activated.
A pilot can deactivate the other stick and take full control by pressing and keeping
pressed his priority takeover pushbutton.
For latching the priority condition, it is recommended to press the takeover push
button for more than 40 s.
This allows the pilot to release his takeover push button without losing priority.
However, a pilot can at any time reactivate a deactivated stick by momentarily
pressing the takeover push button on either stick.
If both pilots press their takeover pushbuttons, the pilot that presses last gets
priority.
Note: If an autopilot is engaged, any action on a takeover pushbutton disengages it.
In a priority situation
‐ A red light comes on in front of the pilot whose stick is deactivated.
‐ A green light comes on in front of the pilot who has taken control, if the other
stick is not in the neutral position (to indicate a potential and unwanted control
demand).
Note: If the aircraft is on the ground and commencing its takeoff run and one stick is
deactivated, this triggers the takeoff “CONFIG” warning.

The fly by wire has 4 modes called laws ranging from full flyby wire to "direct control" see:


(In preview its not showing the link above so this is it in manual: http://www.airbusdriver.net/airbus_fltlaws.htm )

The sticks operate in the same way in all laws though the response of the aircraft will be different.

In both the AirAsia and the AirFrance cases there were times when both pilots were giving contradictory inputs and didn't seem to recognize this. I'm not sure if the AirAsia aircraft had departed before contradictory inputs were made but in the AirFrance case it appears that the pilots did not recognize who had control or that they were contradicting each other at a time when the aircraft could have been recovered. The tactile queue of "connected" side sticks would seem to be a better way to address this confusion then the aural alarm and lights that are supposed to.

Again I am not judging the choice of disconnected sticks just trying to understand what the rationale for implementing the sticks as they are was. Part of the perplexity is that I am unlikely to ever hear of a failure that having separate redundant joysticks addresses as their redundancy would keep them from being reported in the context of a spectacular crash so the separate/redundant idea holds much less force in my imagination than the image of hapless pilots pulling and pushing on a couple spring loaded joysticks in an extremely high stress situation.
posted by Pembquist at 2:40 PM on December 4, 2015

I have no source other than my own thinking on this, but some ideas:

1- With physically linked (or electronically-linked force feedback, same thing) controls, the pilots cannot discern whether the forces felt in the control are due to aerodynamic forces on the control surfaces versus due to the other pilot pushing their control. See Egypt Air 990 where according to the CVR data the pilots were fighting each other on the controls as well.

2- With a side-stick, the control forces are much less than in a yoke (hand/arm forces instead of arm/shoulder/torso as input device). A side-stick providing force feedback will necessarily have a smaller range of force that it "should show" to the pilot; of course in a small plane this is limited by the total aerodynamic forces on the control surfaces, in a bigger plane the computer is faking it all. But you could say the force information communicated to the pilot is less useful with a sidestick because it would have to be limited or "lower resolution".

Therefore I can imagine an engineering rationale which says do away with force feedback in the stick altogether. You could says it's counterproductive or confusing in some cases. Then you implement the red/green/priority system to handle the need for one pilot to take over and know where input to the computer is coming from.
posted by frontmn23 at 4:00 PM on December 4, 2015

Using the algebraic sum of the stick inputs is never the right thing to do. Taking the input from the pilot who touched the stick most recently would make more sense.

I suspect that Airbus realizes this by now, but can't implement any changes without admitting that the current implementation was a terrible idea, and opening itself up to lawsuits. It would take an FAA mandate to fix this.
posted by monotreme at 5:58 PM on December 4, 2015

It actually makes quite a bit of sense if you think about where they were coming from. Aircraft with conventional direct control have the yokes physically linked together - there's a large torque tube that runs underneath the cockpit floor to join them both to the elevator control, and a system of chains and pulleys to link them to the aileron control. If both pilots are touching the controls, you ultimately end up with a summation of the forces they both apply.

The goal when designing new stuff like this joystick control system is to keep things predictable, so if a pilot is transitioning from an older direct control system to newer fly by wire it should be as seamless as possible. A century of aircraft design tells the pilots when they both push on the controls, they efforts are summed, and it's a really bad idea to go around messing with that historic experience.

The question here is why there's no "force feedback" in that if the left pilot push the stick down then the right pilot's stick doesn't also deflect. That part I'm less clear on; it just seems like this particular situation (cross controlling in an extreme attitude emergency situation) was not encountered during whatever cockpit committees they had working this stuff out.
posted by backseatpilot at 8:20 AM on December 5, 2015

is this possible? i am a physicist, not a engineer, and i haven't worked a lot on control systems. but just trying to work out how to do it is making my head spin a little.

obviously you can have feedback, where the joystick shakes in warning. what i am questioning (and what i think is being suggested here) is that the controls (1) are both always in the same position as each other and (2) balance forces so that the two pilots can "push" against each other. in other words, that they behave as if mechanically connected.

i guess it must be possible, but i personally, after spending a few minutes thinking about it, can't work out how to separate the different forces in a meaningful way while also applying the forces necessary to make the positions identical.

anyway, i just wanted to raise the possibility that it may be technically challenging to actually implement.
posted by andrewcooke at 3:47 PM on December 5, 2015