You never even have to ask because they are mechanically linked, and they are also in close visual proximity. If the two small joysticks on an airbus were placed side-by-side on the center console in plain view of both pilots, and they were mechanically linked (easy here because of their proximity), they would be just as natural and intuitive as the big twin yokes.
There’s an interesting book called The Design of Everyday Things[sup]*[/sup] that argues otherwise. It’s about how some things are easy and intuitive to use, and some things aren’t, even if we aren’t aware of why. You should never need the words “push” or “pull” on a door, for example; design it right and people will automatically use it right. He does briefly discuss the sidestick vs. yoke issue; not only the tactile feedback of the other pilot’s control input, but that the most important control (which core_dump might dispute) should be the most prominent.
- It may not be that specific book, but another one by the same author.
Indeed. My own personal preference is for a joystick. I think that when flying with one hand on the controls, it feels more natural to push the stick in the direction you want the wings to bank. If you get a model aeroplane, attach a vertical stick through the centre of gravity, hold the top of the stick and move it the way you would a joystick, the model moves in the same way an aeroplane with joystick controls moves in both pitch and roll. Push the top of the stick forward and the nose goes down, push it left and the left wing drops. Compare that to holding a yoke with one hand. If you’re flying from the left seat your left hand will be on the left side of the yoke and your right hand will be on the thrust levers. Now if you want to bank left you move your left hand down, to bank right you move your hand up. To me that’s not as intuitive as a stick.
Its a fine point and I doubt anyone thinks too much about it, particulary in a large aeroplane where your time spent hand flying is small, but if I was given a choice between identical aeroplanes with a yoke or a stick, I’d choose the stick. Put the stick on the side and it’s both natural and comfortable. The question then is whether to link it to the other side or not. Airbus chose not to. Instead there is a warning light and an aural warning saying “DUAL INPUT”. It should be hard to miss, but when under stress the brain can become overly focused to the point where important warnings are subconsciously discarded. There are YouTube videos of pilots landing light aircraft with the wheels up. All through the approach the gear unsafe horn is blaring and the pilot never seems to question what that noise might mean.
When the smelly stuff hits the fan, most humans revert to their oldest or strongest habit. Even pilots.
We have older pilots with lots of practice since they were very young leaning in cable rigged airplanes… They learned standard operation procedures that worked in all airplanes they ever flew.
Now you stick them in an airplane that has not standard stuff, they fly it for 2-3n thousand hours which is not a big a % of their total time… See the problem here where engineers do the deciding ???
Their are a great many things good about side sticks & fly by wire.
Have you noticed that acrobatic airplanes usually have a great big stick between the pilots legs.
IIRC, most naval fighter pilots do not fly piston powered aircraft except for any civilian training they had before being selected for flight training. Some never even see a propeller. After training and some flight experience they are 25 years old with maybe 2000 hours total time. Would you prefer this to the airline captains we have now? Will you accept the ticket cost to pay taking young people & training them in only a few similar type airplanes ( really expensive for airlines not funded by governments ) and have them be your captain? Want to pay him enough to keep him ??? Want him to gain all his experience & see if he will survive the bad experience he will probably see getting to say 5000 hours?
I have never heard of an aircraft carrier being commanded by a 25 year old under 99% of all conditions. Your information may know of it but I don’t.
The human / machine interface is seldom about what makes the most engineering sense but about the human capabilities.
Using robots eliminates this but in aircraft with public passengers, this is a real had sell.
Many a pilot have made a big mess the first time they tried flying a Beech Barron, engine controls are not fully standard in layout. When the first Cessna 175 came out, the carburetor heat & the mixture control were reversed. After the resultant mess they put them back were 99.99% of all pilots in the US expected them to be.
Old strong habits are really hard to break.
Yeah, I’m a bit old school, but I am a bit old. But I am an alive pilot with a few hours flown so I feel that counts a bit.
If it ain’t Boeing, I ain’t going.
Never fly the “A” model of anything.
Old & Bold are most incompatible.
Ernest K. Gann knew what to do with engineers. He took them along.
There are a lot of sayings… Some were founded from much bad experience.
::: aside::::
Why was not the captain informed when things got dicey, like with a storm? As a captain, I never allowed others concern ( even gov or employer guidelines, they were not up there ) for my sleep to override informing me of bad Wx, unusual happenings or even minor emergencies.
I am an old pilot, not a dead one.
Just because they do not or can not teach outside the box, there is no known law that the airplane can not get outside the box. Then inside the box solutions aren’t. Need a pilot that can & will go outside the box… Really hard to achieve within the airlines industry.
I am not advocating cowboys, but engineering for the sake of engineering is also not always the answer.
Your flight hours may vary…
I like what Richard Pearse has just said.
But … ( You knew that was coming didn’t ya )
I can much better fly with a big stick in either hand than a smaller one on the side.
But…
I have very little experience with those.
But …
In fighters where even an heavy bias one handiness is present, IIRC the big stick is better than the side stick.
Can the airlines not accept pilots like this as there are so many qualified that it is a disqualification condition? Not talking helpless, just not as smooth when hand flying enough that is is noticeable.
Long experience has shown that the big stick worked well & there are not as many years nor the need to qualify as many pilots as there were in, say the early days of WW2
If all pilots who are flying side stick could have any of the three with no mocking of other pilots or company balking, as a system were the pilot could bring his own & just stick it in on which ever side he sits on, how many would still pick the little flyby wire side stick?
Might be an uninteresting talking point while waiting for the union meetings to start…
Is this really a problem? If you want to turn left in an airplane you bank left by turning the yoke CCW. If the same pilot is in his car driving home to turn left, he turns the wheel CCW.
My question is this: most people have a dominant hand. If a pilot gets very good as a co-pilot he is using his right hand. If he then gets promoted to Capt he will be flying with his left hand. Will he be as good with his left hand as he was with his right?
There is the gorilla arm (no relation) factor to consider. If you hold your arm out for a while, it gets heavy. You can grip a yoke so that your hand supports the weight. With a sidestick, I assume there’s someplace to rest your elbow or forearm. Sticks on the center console need to be comfortable to use, but not block you from getting out of the seat when you need to.
Since you never need two pilots to provide extra muscle, I wonder if they considered a single stick that could be repositioned for either pilot, like the throw-over yoke on some Beechcrafts.
I can understand the preference for a stick, but do you think the controls should be mechanically linked?
It’s not a problem, it’s just the reason why I have a slight preference for a stick. I haven’t flown an Airbus and maybe I wouldn’t like it, all I know is that I preferred it when flying light aircraft. The only thing I didn’t like about it was that your legs were more likely to interfere with full control movement.
The answer to your question is yes, but it probably takes time. How long it takes I don’t know because there are a heap of other things being learned at the same time and any slight loss in flying skill goes unnoticed amongst the mass of other things going on. Although flying does require some finesse, it is a relatively coarse motor skill compared to something like writing or hitting a ball with a bat.
[QUOTE=Robot Arm]
I can understand the preference for a stick, but do you think the controls should be mechanically linked?
[/QUOTE]
Not necessarily mechanically because there might be technical limitations to doing that with side sticks on opposite sides of the cockpit without having excessive play in the system, but I do think that the controls should provide feedback and should move together. Airbus made a similar decision with their auto-thrust system. On a Boeing the auto-thrust moves the thrust levers so you can tell by feel or at a quick glance roughly what power setting you have. The Airbus system doesn’t move the thrust levers so the only indication you have of what the engines are doing is on the engine instruments. I doubt it makes any difference in normal operations, but when things are going wrong it is one less thing the pilot has to form their mental picture of what is going on.
Agreed, there are ergonomic and cockpit-layout considerations that pretty much dictate the joysticks be located at the sides of the cockpit rather than the center; my point was just that there’s nothing uniquely “natural and intuitive” about big twin yokes that isn’t also true about small joysticks.
Less stuff to break. Mechanically linked yokes add another set of failure modes (ironically, one problem is that in conflicting-input situation, you’re matching the left-hand strength of the most experienced pilot with the right-hand strength of the less experienced), and regrettably, there’s no lack of documented incidents of pilots (some with, some without linked yokes) stalling their aircraft in response to bad/conflicting airspeed indications. The BEA report on the AF447 crash lists a handful (page 159 and forward):
An operator doing the exact wrong thing after the automated system has handed full control off to him, then doing so for minutes, is a pretty hard scenario to design for. Direct readouts of the AoA indicator might have helped to interpret the situation correctly, but it appears as if the confusion in the cockpit was close on total. They weren’t noticing the “Dual Input” warning, for one thing.
It is a sad that about 1 minute into the incident, the ice had again cleared from the pitot tubes and the pilots were manning a perfectly healthy aeroplane. If the PF had simply taken his hand off the stick (let alone reacted as the loss of IAS data checklist proscribed), this event would never have made headlines.
"Do pilots in modern jets have failsafe instruments to fall back on? " If by “fail safe” you mean some mechanical/electronic thing that will never fail, then the answer is “no”.
In this particular case with a 40 degree nose up attitude I would have thought the Perrier pouring into their laps would have been a clue.
There are a few things to clear up here:
One: I was responding to using flight simulators to model this flight to try and understand what happened. While the circumstances can be input into the system to play out in a certain order, the pilot inputs in a simulator will change the outcome (obviously) and some insight can be gleaned from that, but it all comes with the caveat that what happens outside of certain parameters is merely extrapolated guesswork and isn’t really enough to say aha, the plane should be designed this way instead! There are some design things that can be determined - mainly cabin/human factors issues, but things about how, perhaps, an aileron was behaving aren’t necessarily true.
Two: XPLANE isn’t a full-flight simulator. I’m sure it uses logic similar to some of the big industrial sims, and may be a faithful representation of some of the behaviour of aircraft under normal or typical failure circumstances (these things are “true” representations in the full flight sims), XPLANE would still be merely guessing probable behaviours once you go beyond what is being simulated based on actual data. This is a full-flight simulator. It is used for pilot training, not for aircraft design. XPLANE works on a desktop computer. A FFS uses a dozen computers to run (IIRC - I don’t work for CAE, but I’ve toured them!)
Some simulators are used for aircraft design - like the simulation technology being used in Bombardier’s CIASTA, but it isn’t dependent upon accident FDR results at all (mainly because the plane hasn’t even been built, and because accidents are due to one sequence of events and are not representative of the entire aircraft in general).
Does that help clarify things? Simulators are good - they are VERY good - at behaving the way the plane does. It’s just that they have a limit, and accidents like AF447 occurred outside of those limitations, for the most part. You can definitely learn something from them, but you can’t learn everything from them, if that makes sense?
ETA: whether the stall performance characteristics are realistic…yes (sims in general, I don’t know XPLANE), to get you into the stall. Once stalled, it’s pretty theoretical behaviour until various parameters return to a normal range. Thing is, that won’t tell you for sure if you can get out of the stall in the plane, it just tells you that the computer accepted that the plane returned to a normal attitude. You can pull a CRJ200 simulator out of a deep stall rather easily. The plane…not so much. It tends to hit the ground rather abruptly instead.
That’s not what failsafe means. Fail safe means that when something fails, it is safe instead of dangerous. For example, the signals in the NYC subway system are linked to hydraulic tripcocks beside the track that will trip the emergency brake of a train if it tries to go through a red signal. But the tripcock is designed in such a way that if the hydraulics fail, it gets stuck in the up position (a spring holds it up and the hydraulics push it down), preventing any traffic from moving instead of potentially allowing traffic to traverse a red signal.
When engineers talk about failsafe designs, they’re not saying “how do we make this never fail?” Because that’s a very difficult problem. They are saying, “if this thing does fail, how to we make it not fail catastrophically?”
40 degrees was the maximum angle-of-attack - the angle at which air was incident upon the wings. However, the aircraft was in a steep descent at this point; IIRC, the maximum nose-up attitude of the aircraft with respect to the horizon was 16 degrees, slightly less than you’d see during the intiial climbout just after takeoff.
You are ignoring some unique characteristics of X-Plane.
See How X-Plane Works | X-Plane
Now for various reasons I expect that this won’t be particularly effective in the stall or any other extreme event because of the complexity of air flows; but it is not correct to say that flying outside of the aircraft’s designed envelope will automatically break the flight model.
In an earlier discussion, one of the posters here talked about stalls in a very large aircraft. They are not pretty or smooth. The Perrier would not be flowing into their lap, it would be splashing onto the ceiling. (From what the experience pilot here said) the plane will rock violently from side to side will buffetted in every direction. Add to this the incipient cause was that hey flew into a thunderhead, apparently, and the moisture iced closed the pitot tube. I remember my flight instructor mentioned getting too close to a thunderstorm one day, and you could see the wings of a small plane bending visibly up and down.
All in all, this was likely a serious carnival ride combined with bucking bronco. The odds that their inner ear woud give them any clue as to their attitude (aircraft wise) and absent horizon or other cues, likely they were completely disoriented - which did not add to their ability to think rationally.
Either (a) copilot did not think rationally, or (b) he was trying to climb above what seemed to be excessive storm up/downdrafts that were beating the hell out of the plane and dragging it down (so it seemed) the logcal thing was to climb, climb, climb like hell. (I’m going iwth (C) both of the above…)
I’ve been in aircraft turning and maneuvering for approach, and looking up the cabin, your inner ear swears the plane is tilted up quite a ways; while looking outside shows that not olnly is the nose pointed down or level, but the whole plan is turning to one side. We assume out inner ear is telling us the truth, but we rely on eyeball clues a lot more than we realize.
Read about the JFK Jr. crash sometime. He basically twisted, turned and spiralled into the sea because (presumably) he failed to read instruments correctly and did not realize he was almost nose down - attitude is not obvious without a horizon in cases where you would think it would be.
While I don’t want to belittle Sculley’s job - he did an absolutely fantstic one; most pilots would have probably tried to exceed their glide range to reach a runway and then stalled into a built-up area - his smartest move and the best part of the whole scenario, was deciding he could not reach the airports, with plently of time for plan B; but he ahd peace and quiet and a smooth ride while he thought about it. The ride down was basically a nice smooth glide, with no distractons or engine noise. (And the landing, was, of course, brilliantly done).
I think the issue is that you can’t validate the x-plane model without flight testing. So it may be good enough or it may not be, we can’t know without testing and no one wants to test outside the flight envelope. Having said that, the NTSB are very interested in flight simulators having fidelity beyond the normal flight envelope. Maybe if enough pressure is applied to the right places someone will come up with a good solution.
The point of that was that the B747 is, either by design or accident, a very good, predictable aeroplane and the manufacturer truly believed it didn’t need a stick pusher and the FAA agreed.
The info is from a book called “Handling the Big Jets” by D. P. Davies who was one of the CAA certification test pilots for the B747. I don’t have the book with me at the moment so might have the details wrong.
Can we assume that is not possible if the aircraft is moving only horizonally, but since it had a rapid vertical component, the two numbers could differ? It seems like they normally would be the same.