That's the point, one way to evaluate has been withdrawn.

AAIB did not have to prove anything, they just mention that what is possible with a more conventional cockpit layout is not with an A320 concept.

The point the report is making is that what would be unwise to do in a conventional cockpit layout is especially unwise in an A320.

If dozy knows better than the AAIB ...

Because "following through" means just that "following through"
Who said "controlling" ?

Anyway, are you really in the best position to comment on what's happening or should be done or not on a flight deck ... ?
Some guys around have Experience you know ... they can make such statements if they wish.

It was designed because it is cheaper than conventional controls, and because they wanted a unique selling point. All of the "protections" are there as a simple requirement to overcome the drawbacks in the no-feedback concept - not included to help pilots.

you have just to accept that your stick and the stick of your buddy may perform differently.

I didn't say that... But on the other hand I think we're reading different things in the report. I can see the brief mention of conventional controls, but I can't see the AAIB mentioning anywhere that conventional controls are better - you're reading your own position into it.

Of course they can, and I'd never presume to tell them not to. However, they must also accept that their position is not universally held, and the number of pilots who are happy with the Airbus layout is not just confined to the younger generation, but also top-drawer ex-mil veterans of the kind they refer to as "real" pilots.

That's a bit cynical. In fact, the R&D costs of developing the control system were pretty massive and had been ongoing since the '70s. They're easier to maintain and provide weight savings, certainly - but the main reason (as I've said before) was the potential for unprecedented flight deck commonality between types - in the sense that conversion across the range from the A320 series through the widebodies up to the A380 is massively simpler (and thus more straightforward and less expensive) than for any of their competitors.

The protections do far more than overcome the drawbacks, particularly in terms of offering safe manoeuvering authority across the full range of control inputs, useful in avoidance of terrain and mid-air obstacles.

Indeed - it's not about size, it's what you do with it that counts (or so my wife reassures me... :E)!

FBW and pilot controls
 

One mo' time

We have returned to many discussions from the AF447 thread.

Once the decision is made to move from a control stick/wheel that provides mechanical connections to the jet's control surfaces, we must consider the implementation of the controls.

We went thru this in the 50's when hydraulic valves at the base of the stick/wheel controlled pressure to actuators out on the wings or tail. Many commercial jets maintained actual mechanical linkages to the control surfaces, and maybe a few still do. By 1960, only a few military trainers and older designs did that. All we had were valves at the base of the stick/wheel to control pressure to the control surface actuators. Control feedback was extremely artificial, as using springs!!!! No aero force feedback whatsoever. One jet used a bellows to help with backstick pressures in order not to over-gee the plane. But that went away by the 70's.

Then we came to FBW.

Do we mechanically connect dual-control jet sticks/wheels? Do we implement some kinda feedback as to how much we pilots are commanding of the control surfaces? Do we use a stick that moves or simply use stick pressure? and the beat goes on.

And then we had the "protections" and the autopilot and autothrottle connections to our flight control systems. GASP!!! To this day I am amazed at all the modes and control laws of the systems in operation today. Even a techno-geek pilot such as myself gets confused. And make no mistake, I would appreciate a simple control law that closely resembled what we had in the "old days", like late 60's. Heh heh. This would still require electronic analogs to the old hydraulic valves and feedback springs and such that we had. Hence, we have "gains" that limit control surface rates and total movement depending upon dynamic pressure the jet is experiencing. So only a so-called "direct" control law would resemble the "old days", and even then such an implementation could have problems with over controlling.

From my point of view, mechanically connected flight controls in family models seem like a good idea most of the time. But we still have to engineer the electronic connections to the jet's flight control system. Failure modes and "who has control" issues come to mind. I can tellya that I "helped" many a nugget land during severe turbulence/crosswind landings in one of the "old" systems. Then did same in a FBW system with zero control feedback from the jet or the nugget in the other seat. Know what? No difference. The AF447 situtation was different, and I believe that control position feedback might have made a difference in the PNF reactions ( although he mentioned a few times "we're climbing", etc).

As with Asiana, et al, it's up to the "pilot in command" to harp on the nugget and/or take control himself when things aren't going right. Regardless of the control implementation, it's up the aircraft commander to do as much as possible to ensure a safe recovery from an unusual situation or a sloppy approach or...... This requires knowing what the jet is doing versus what it's supposed to be doing, no matter what the "other guy" is doing. There should be a clear chain of command, and if the aircraft commander decides to take action, then it is mandatory to do so regardless of the rank or experience of the other guy. Any objections to that point of view?

Oh gums, if only that were true. This particular back-and-forth has been going on since long before I joined this forum. I suspect, sadly, that it will continue for some time to come - as the subject acquired a political dimension very early on.

Absolutely none, and I couldn't have put it better myself.

History tells us that being too rigid here can bite us badly - I'm thinking of examples such as Tenerife in 1977, Birgenair 301, Palm 90... to name a few off the top of my head. I'd agree that the aircraft commander carries ultimate responsibility, and as such makes the final call - but only in split-second life-or-death scenarios should that call be unilateral, because your colleague may have spotted something you haven't. The book on QF32 shows Aussie-style CRM at its finest, and Capt. de Crespigny is not ashamed to admit that his first ideas were not always the best.

EDIT:

You do bring up some interesting points though, so let's have a look-see.

To the best of my knowledge the only airliner types still being used in significant numbers to retain a cable-to-surface connection (albeit for backup/manual reversion only) are the various B737 models, and the DC9/MDx0 derivatives. The design trend since the '70s seems to have been more in favour of engineering greater failsafes into the hydraulic system than cable-based reversion. This was unavoidable for widebodies, as the forces required to move flight surfaces of the required size are way beyond human muscle power. As to why this translated to narrowbodies, I can only speculate - but I do remember reading an interview with the Aloha 737 Captain who stated that flying the 737 in manual reversion almost completely exhausted him physically.

So it would seem that one benefit of beefing up the hydraulics means that the controls would behave in a manner to which the pilot is more accustomed. This is also the reason for the existence of Alternate Law. On the AF447 thread, people seemed to get bogged down in the minutae and making things seem more complicated than they are. There are three control laws in the Airbus system - Normal (which very few Airbus pilots will ever leave outside of a sim check), Alternate and Direct. There is only one Alternate Law, but it has various submodes depending on the type of failure that caused the degradation - the specific submode is largely academic, because the purpose of Alternate Law is ultimately very simple - and that is to work around the failures to provide as close to Normal Law behaviour as possible. Direct Law only engages in the air in the event of multiple serious systems failures - going back to the example of QF32, they were down to 50% of their electrical systems and 25% of the hydraulics with significant damage to the fuel system and holes in the starboard wing - even with all that, the control law they degraded to was still Alternate.

Not cynical in the slightest. It was stated as fact in a nice airbus animated presentation at the beginning of the endorsement.

... and if you think about it - building aeroplanes is a business, and that's all it is. The airbus system is cheaper and saves operating costs. That's why they did it, and that's how they sell it to buyers.

... the problems start when someone starts "selling" the concept to pilots with rubbish about how safe it is and how you can't stall an airbus.

The jet is supposed to be doing what is commanded by the PF but when the PNF is not aware of the PF inputs ... something is missing, something nice is lost.

Because the mention is not such. The mention is linked flight control commands allow things that unlinked FCC don't. That's something you don't have the experience to appreciate and you don't accept it to be mentioned.

I have to agree with Conf.

If you are forced to fly a "crew" jet with dual controls, then having the visual and mechanical duplication of control stick/wheel is one addition to "awareness".

The implementation of such systems will be more complex than what the 'bus has, or the Shuttle or the Viper that I flew. Nevertheless, I can appreciate the added awareness factor.

All that being said, I still have a problem with several accidents where basic airmanship played a larger role than mechanically connected controls.

Hey gums,

Agreed. However, looking dispassionately at recent civil aviation history it can be argued that said addition is of limited benefit in a two-crew airliner scenario.

CONF is right when he says that the non-linked controls do not allow for "follow-through" in the training scenario, and he's also right when he says this makes the FBW Airbus require a very strict interpretation of when a pilot transfers to line training in that he or she must be consistently competent in terms of handling the aircraft before being signed off for the line. However, if I understand modern airline ops, then that prerequisite is mandatory even with conventionally-controlled types. I suspect that if, say, a landing was performed in which a follow-through became necessary, then both the training Captain and trainee would be summoned for tea and biscuits, with the former being required to explain why they let the trainee proceed to that point.

As we're both painfully aware, a recurring theme in the AF447 discussion was the argument that if the controls were connected, the PNF would have been able to see the errors made by the PF and taken control. This is countered by the fact that the same scenario has cropped up several times on types with conventional controls - and the PNF either did not see what the yoke was doing or ignored it, resulting in a crash.

Well, the T7 and 787 have FBW implementations with software-driven force-feedback, and yes - both systems are way more complex than that of the Airbus FBW types. But there are already differences in how the systems in the two aircraft operate vis. commanding rate vs. deflection - such differences would be anathema to Airbus's original goal of having the whole family of aircraft behave as close to identically handling-wise as possible.

I suspect that all of this would have been a non-issue had the subject not acquired a political dimension early on. The idea that the transition to sidesticks was a physical embodiment of the reduced role of the human pilot and increase in automation seems to be as widespread as it is inaccurate. The sidestick concept was being tested on a Concorde airframe years before the launch of the B757/767, which was when digital FMS became truly mainstream. We might have even seen sidesticks as an option on the T7 had United, the launch customer, not vetoed it.

Indeed - and truth be told there has not as yet been a single accident in which connected controls could definitively and provably made a difference.

Paraphrasing Gums I can say ..
All that being said, I still have a problem with several accidents where basic airmanship played a larger role than mechanically or not mechanically connected controls
And paraphrasing Dozy I can say ..
Indeed - and truth be told there has not as yet been a single accident in which connected or not connected controls could definitively and provably made a difference.
So that being said and accepted as irrefutable facts ! .. it is obvious that there is one who handles these commands that will be able to make the difference between a flight that ends normally or crash
Conclusion .. the problem is more than often the pilot(s) :}

No, jcj - not at all. It could be any number of things from insufficient or improper training to poor application of procedures, fatigue... the list is very long!

Fatigue = pilot problem .. not aircraft problem
Poor application of procedures = pilot problem .. not aircraft problem
Improper training = pilot problem (pilots are trained by pilots) .. not aircraft problem
Exactly what I said ... the pilot(s) :}

Or if you want ...
Fatigue .. you don't drive a car with fatigue as you will fell asleep and have accident
Procedures ...You must respect the traffic laws or you will have accident
Training .. You must go to driving school and have a driving license .. or you will have problems
So ... the driver is the problem .. not the car ...

Not at all - if the pilots have not been trained or briefed properly - or if they have been put in a position where safety is compromised and have been coerced into flying anyway, then at least some of the responsibility lies with the airline.

A car is usually a personal mode of transport - most drivers are not employed to do so.

There are 2 aspects of this argument going on at once:

1. Asyncronous side sticks verses conected yokes based on crash statistics/reorts.

2. Asyncronous side sticks verses conected yokes based on pilot's evaluation/preference.

User feed back (pilots in this case) is very valuble when evaluating a concept.

If you are a person who considers himself a responsible professional and intellectually honest .. you never accept or do what you explain above
However, if you agree ... you do not need excuses or loopholes and must agree to be designated as responsible if anything happens because of you

In my professional life I have been sometime (exceptional) forced to do things contrary to or not in keeping with the rules .. and always by verbal way (as always happens)
I always said I wanted it to be postponed by a written order signed by a senior or responsible person ...
It has always been told that they were going to study another way to do the job and so what they had asked was canceled ...
I've never had to suffer because of this trouble

Doze, have you had the opportunity to jumpseat at all?

I think if you were to jumpseat on a few line training flights for example or a base training session (simulator even) it would really help you understand what a lot of people are trying to get at.

(Sadly, I'm not in a position to offer such a thing or I would)

The snag with side sticks is training.

With the explosion in new 200 hours cadets straight onto an A320, it would make things a huge amount easier if there were moving side sticks so you could feel what forces were being put on at the earliest possible stage.

When you have more experienced crews, I think it is much less of an issue and the additional complexity and more to go wrong argument probably wins. The problem has been that our training is 200 hours then in the RHS, and for the first few hundred hours it's much more difficult for the RHS and LHS.

Guess we have to have a few here that have flown the mechanically-connected sticks/yokes as well as the system implemented in the 'bus and the last jet I flew, where there was no visual or force feedback as to what the nugget in the other seat was trying to do.

I fully appreciate the views of those who make the point that "feedback" between the dual controls can make a difference. OTOH, with many, many circuits flying close chase in single-seaters, and then in the family model of the Viper, I can tellya that seeing and feeling what the jet is doing compared to what it is supposed to be doing counts a lot more than the connected flight controls.

The engineers will say that everything was working as designed, so why the crash?

I maintain that basic airmanship counts more than all the "protections" and AP modes and auto-whatever. The recent SFO crash showed how a seemingly perfect system, operating as designed, cannot protect a careless or whatever crew. I also see a trend in the AP/FMS/FLCS connections that do not help the credw when various sensors or malfunctions come into play.

We must have some common core flight control logic and autopilot logic and such to depend upon when things go wrong. Made my case on the AF447 threads, so 'nuff said. Asiana sems to confirm my beliefs.

If the law would allow ( too old to do it legally now, heh heh), and we had two planes with good comm between us, I would take any pilot up and fly chase, do patterns, and land successfully.

I've said it many times before but I'll guess it's not the last time I will repeat: if you are not familiar with basic flying, electronics and airline ops, your chances of correctly understanding the Airbus FBW are close to nil. If despite unfamiliarity with aforementioned you still have the urge "to know" about the uncoupled sticks it is very easy to satisfy it by believing folks claiming to be experts while airing theories that only show they either have never got into transport cockpit or enjoy misleading others.

Uncoupling the sticks is not an issue because when it comes to transports only one pilot can be in control at any time, be it left, right or auto. Grabbing the wheel from your cockpit significant other or "helping" him/her is not an usual, approved procedure (there is good reason for it) and can be used only in extremis and for very limited period of time. Also algebraic sum of sticks nicely replaces the rigid coupling; as simple spring makes displacement proportional to force the effect is the same as if two pilots were fighting over coupled controls.

Both AAIB and BEA noted it is not possible for assisting pilot to know what inputs the handling pilot makes with coupled sticks. Neither made any conclusion or recommendation regarding it; most of the interpretations of their position you could read on this thread are pure conjecture. It might come as news to some but while you are flying stable and not particularly maneuverable aeroplane with relatively large weight and C.G. range, it's not just amplitude of your colleague's across the cockpit inputs of concern to you, neither is your own when you are PF. How much pull? Just as much as it is needed and feedback is not through wheel/stick, it is from aeroplane's attitude as observed on AH or through windshield.

Artificial feel is not there to help you feel speed error, it is designed to a) prevent you from ripping your aeroplane apart through overload or killing passengers at high speed b) have reasonable forces and maneuverability at low speed maneuvering e.g. landing. If aeroplane gets mistrimmed, pilot trying to get the aeroplane to proper flying speed will have to fight the stick forces while retrimming. No need for a lot of imagination to understand what happens to pilot firmly believing he can feel speed deviation through wheel. Thomsonfly at Bournemouth might provide some clue.

That Airbus is unstallable is early marketing hype turned into libel. No such claim can be found in FCOMs or derived from them, at least by the reasonable and knowledgeable pilot. Airbus FBW has very clear architecture, modes of operations, operating procedures and limitations. Those unable to cope with it are welcome to air their inability to understand it on PPRuNe but shouldn't be type rated. In ideal world, that is.

200 hours wonders made their way into multi engine heavies' cockpits since at least WW2. Except the hype, there is nothing out there to suggest that they were inadequate in their role or that those who couldn't cope with basic items, such as flaring to land at 10 hours could somehow be magically taught at 200, in heavy. Type rating course just shows you the difference between flaring 172 and 777. In the days of yore, it was done in 6 circuits. I don't see folks who made their first landing in type on line after ZFT bending airframes a lot, can anyone?

As there are a few folks who managed to keep the bus within protection limits and still wreck it and LoC accidents are fortunately rare even on unprotected aeroplanes, protected FBW was not made mandatory on passenger transports. As what you read about it on PPRuNe is mostly untrue, it is not banned or Airbus was asked to redesign it either.

"I say again, AF447 was not an "in extremis" accident." It was a "usual, approved" procedure so there was absolutely no need for the PNF to either see, "grab the wheel" or "help" the PF, who was in the normal process of crashing the aircraft. :cool:

What's all this Dozy and Clandestino codswallop about the horrors of grabbing the (real) stick off the FO if he can't cope?? So what if I take over? If I was flying with 200hr FOs, I would expect to occasionally.

Hi Clandestino,
Since the autopilot was engaged during that approach, why do you think the crew could "feel" the speed deviation through the wheel with the AP trimming the stab?

From AAIB http://www.aaib.gov.uk/cms_resources...9%20G-THOF.pdf" Within 1.5 seconds the stick-shaker (stall warning) activated and in the following two seconds the thrust levers were advanced to the full forward position. The autopilot mode changed from localiser and glideslope to Control Wheel Steer (CWS) pitch and CWS roll."

Do you mean "uncoupled sticks"?

Why do Airbus still fit coupled rudder pedals? Why not save weight and complexity with uncoupled rudder bars?

Absolutely un recommended procedure.
Should never take place.
Anything but nice.
Where have you been ?

Er - neither Dozy nor Clandestino argued against taking control from an FO out of their depth, or anyone else for that matter. The point is that whether in a conventional or Airbus FBW layout, such a transfer of control must be done correctly - with "I have control/My aircraft" etc., in conjunction with a "Go Around" call if necessary. Simply yanking on the primary flight control without warning is a very risky thing to do with any type of control layout - as I'm sure you well know - because it runs the risk of startling the pilot in the opposite seat which in turn risks an unpredictable outcome.

I suspect you know why - because the absolute worst-case scenario in terms of control reversion in a FBW Airbus is manual trim only in pitch and rudder. Tying the THS and rudder control into the digital flight control system exclusively would remove that last fail-safe.

As with conventional types, the coupling of the rudder and trim controls is a side-effect of the electro-mechanical implementation - and it's the one aspect of control in which combined muscle power might still have an effect in terms of expediting the required control input.

True, but it's also true in conventional-layout ops. Unannounced manipulation of the controls from the opposite side is risky, as I said above - and you must know that when going by the book, "follow-through" has no place in line ops, even line training. It should not be necessary in anything other than trainers and the simulator.

I never even implied that you did. Let me remind you what you said:

The "horrors" of it all! Tea and bikkies, explanations required! The reality is nothing like what you hypothesise.

It's a shame you don't fly RPT two-crew, Dozy. You'd realise it normally doesn't work like that in the scenario the discussion is alluding to.

And where did you drag that gem up from??

Why you keep making such statements on things you absolutely have no experience on is a mystery to me ...

Then hurry up to give a call to Airbus, rudders have now moved to the fully digital era ...

I don't. I took it just as an illustration of out of trim forces (which are usually but not necessarily introduced by autopilot), also significant was inability of the crew to immediately use manual trim to prevent nose from rising, which would not be issue with autotrim.

Yup, sorry, my mistake.

No savings there. Airbus was into preventing the pilots from going aerobatic since A300 - it is called flight augmentation. It is easier and cheaper to introduce limits into FBW than classic controls, that's why we have FBW since 320. Whether mechanical or electronic, rudder is just rudder, it's used for decrab and engine failures so no need to have sideslip protections as long as the pilot has some idea what he should be doing. If he keeps being clueless for prolonged period of time, even protections will be unable to save him.

Not necessarily. Depending on location (i.e. culture) it might end with note in the training log, air safety report or confidential report, a quiet word to chief pilot / head of training or training captain being reinforced in his opinion that all F/O trainees are indeed useless and should be administered beating until they improve.

It is worth noting that fight over controls does not leave a trace on QAR until it gets so extreme breakout mechanism gives up (if aeroplane is so equipped) and yokes go their separate ways while "Dual input" gets recorded and might get capt into trouble if it is discovered he made no attempt to elaborate the circumstances of it through ASR/CR. Darned Airbus. Darned. :E

Normally and overwhelmingly usually it does, resulting in major non-event. It is perfectly normal for PPRuNers to assume that the most extreme events are industrial norm, after all they are the only ones picked up by media.

I can only speculate whether his intention was to demonstrate that a) experience is helpful but not prerequisite for understanding b) personal remarks that can be reduced to "I know, you don't and I'm not telling more than that" are not particularly helpful in developing meaningful discussion but if it were, he made it.

Hi Clandestino,

It's a bit more complicated than that as DozyWannabe pointed out.
In the event of a total loss of electrical power to the flight control computers, the only way to keep the aircraft level will be by using slip. When was the last time anyone practiced that on Airbus?

Boeing 777 has cable connection to spoilers 4 and 11 to maintain wings level. 11.5.2
http://www.davi.ws/avionics/TheAvion...ook_Cap_11.pdf

"asynchronous"
 

Any time I am reading the Tech-log threads list, I am asking me if "asynchronous" is a false-friend from "asynchrone" in French..:} My dictionary says no.:{

In FBW, both controls use the same clock and the same timer ! They are everything but asynchronous, and specially during dual inputs when they do an addition using the same processor!

The words "coupled" or "uncoupled" [sticks] is not much clearer in the deep of our discussion, and Clandestino himself tied his feet in that carpet:O

Thanks to Rudderuderrrat and gums who maintain the heading along the thread toward the only goal of pilot piloting : put the plane on the wanted path:), and so never land before the threeshold :ok: (They don't fly Asiana" ,they both are automation addict, and, like Bubbers44 and all the old-timers and younger hand-flying, don't put their eyes in their pockets :ok:)

November 12, 2001.

Connected?
 

I guess we have to define terms.

The digital FBW systems have a basic time frame/sample rate. So dual controls that feed the inputs to HAL are sampled at "x" rate, then used by the system.

The pioneer system I flew back in 1979 was analog. The force inputs on our controls was used by four computers, and the most benign output was used to command actual surface movement/rate/etc.

In our family models, the control signals were "summed", so I could equal the force that the nugget in the front seat was exerting. The FBW used the resulting electrical signal to command control surfaces.

So I would prefer the term "mechanically-coupled" to discuss those planes with dual controls, regardless of whether they are FBW or the "conventional" implentation many of us grew up with. out,

@RRR:
 

Did they forgot to install that one on 787?

or is a total power failure to the flight control computers - up to HOT BAT BUSSES (Airbii)- such a remote event it will probably never happen?