Fly by Wire questio0n

[punkalouver]

quote from an article....

"In modern airliners, the primary advantage of FBW is that it allows engineers to use lighter wing and tail structures"

Why?

[Milt]

One big advantage of FBW is that it enables the use of artificial stability with the computers, usually triplicated, working like mad things to give us a smooth progress.

Instead of the usual down load on the tail to produce natural stability in pitch the designer can now program the computers to provide a significant up load from the tail and at the same time reduce its size and weight. This also leads to lower weight for the wing's structure as they have less of a lifting task. Result - a big increase in efficiency. Stability in roll and yaw is simplified and smoothed.

Another big advantage is that the designer and the TPs are able to introduce automatic g limiting. stall inhibiting etc to prevent below average operators from coming to grief.

Artificial feel has to be an integral element of the system as humans all manually fly and drive by the feel of the controls. The F-16 fighter has a Side Stick Controller force per g of 3. The pilot is thus pulling 27 pounds with his right wrist to reach an automatic limit of +9g. Meanwhile the tail is waggling around smoothing out tiny accelerations and preventing a runaway pitch up with a fairly hefty Up load.

These are just a few basic features of FBW flight controls.

There is already some spin off of aviation's FBW into ground vehicles.

[Chris Scott]

Quote from Milt:
One big advantage of FBW is that it enables the use of artificial stability with the computers, usually triplicated, working like mad things to give us a smooth progress. Instead of the usual down load on the tail to produce natural stability in pitch the designer can now program the computers to provide a significant up load from the tail and at the same time reduce its size and weight. This also leads to lower weight for the wing's structure as they have less of a lifting task. Result - a big increase in efficiency. Stability in roll and yaw is simplified and smoothed.


My knowlwedge on FBW is very limited, but what Milt has said is in need of some qualification in relation to current airliners. As no one else has yet done so, I'll do my best. The design of airliners has not yet been allowed to advance as far as he is suggesting.

My understanding is that well-established military FBW aircraft, the F-16 being an old example, have either neutral static stability or even static instability - whether it is one or the other doesn't affect what I am here for.

One of the advantages of a canard layout - where the horizontal stabiliser is forward of the main-plane, is that a canard is inherently stable in pitch but, nevertheless, produces positive lift from its stabiliser, thus reducing the lift requirement of the main-plane. In a 'conventional' layout, however - with the horizontal stabiliser aft of the main-plane, static stability in pitch is only achieved by the stabiliser generating negative lift, which has to be offset by extra lift from the mailnplane (as Milt has said).

So an aeroplane with a rear-mounted horizontal stabiliser (let's just call it a tailplane, so much less of a mouthful) - like all current airliners since the grounding of Concorde - can only be stable if its tailplane produces negative lift throughout the flight envelope. [U]And they all do.

That all FBW airliners still have negative-incidence tailplanes and positive static stability is best illustrated, I think, by the fact that it is still possible for the pilot to fly them with a complete failure of FBW. [This is usually done using the stabiliser trim for pitch and the rudder for yaw/roll.] But they do tend to be trickier to handle than the pre-FBW airliners were.

Returning to punkalouver's original question:
"In modern airliners, the primary advantage of FBW is that it allows engineers to use lighter wing and tail structures." Why?

Milt is right, I think, as to why FBW has produced weight and efficiency savings already.
1) The tailplanes can be and are smaller, because FBW reduces the pitch stability required (hence the trickier handling with total FBW failure).
2) On Airbus, FBW limits vertical acceleration to +2.5g, the certification limit for all airliners.
3) Certain gust-alleviation features are sometimes available, like the LAF system on the A320, which alleviates the wing loads in turbulence by rapid movements of the spoilers and ailerons.

In summary, airliners are well behind the military in exploiting the possible (cost) benefits of FBW. This is because the regulatory authorities, thank goodness, take a conservative approach to airliner stability. But it's exactly 20 years since digital full FBW first flew passengers (A320), and a lot longer since analogue FBW was pioneered by "the pointy one". Civil digital FBW has largely confounded the doom-merchants, and policies are likely to move on.

Well, that's my tuppence-worth. I now await a rash of corrections from the cognoscenti...

[Mad (Flt) Scientist]

So an aeroplane with a rear-mounted horizontal stabiliser (let's just call it a tailplane, so much less of a mouthful) - like all current airliners since the grounding of Concorde - can only be stable if its tailplane produces negative lift throughout the flight envelope. And they all do.

Sorry, but that simply is not true.

The load on the tailplane is required to trim out the residual pitching moment from the wing; because most wings are significantly cambered, there is a definite nose-down pitching moment for typical cg positions in 1'g' flight. However, this is a static force balance issue, and has nothing to do with whether the aircraft is stable in the pitch axis.

Stability is to do with the tendency of the aircraft, if disturbed, to return to equilibrium. That depends on the pitching moment derivatives, not the absolute values of pitching moment.

For example:

take a stable aircraft, with a tail of 100 sq ft and a download on the tail of 10,000lbf (all arbitrary numbers)
now make the tail ten times the size, 1000 sq ft
for TRIM the tail download is identical - 10,000lbf. Because the rest of the aircraft is the same, so the force required to balance the wing/engine/etc is unchanged.
But an aircraft with a huge tail would be significantly more stable in the pitch axis.

do the same in the other direction. As the tail gets smaller you don't necessarily lose the ability to trim, but you become less stable.

Or:

take a REAL aircraft. Trim in 1 'g' flight. There's probably a small download on the tail, especially if we have a fairly aft cg.
now start to bunt. Push forward on the stick. The load on the tail reduces and eventually will be negative. The aircraft doesn't suddenly become unstable when the sign of the tail load changes.

Or:

Can a tandem wing aircraft only be stable with download on the second wing?
Can a canard only be stable with download on the wing?

Obviously the answer is no; those types can be stable with upload on the rear aerodynamic surface. Whether I call it a tailplane or a wing is irrelevant to the equations used to calculate stability.

[fantom]

Upload (lift) from the tail?

I don't think so.

Also, my mate (a test pilot on Eurofighter/Typhoon) told me that his aircraft was the only one he knew that did have upload from the back end.

Remember:

aft CG = less stable in flite therefore more difficult to fly manually BUT better fuel consumption because the tail is producing less downforce = reduces apparent all-up-weight;

fwd CG = more stability (think a dart at the pub) therefore easier to fly manually BUT worse fuel consumption because tail is producing more downforce to balance therefore increase in apparent all-up-weight.

f

[captain87]

"In modern airliners, the primary advantage of FBW is that it allows engineers to use lighter wing and tail structures"
Why?

My answer:

Simply because electrical wires weigh much less than mechanical cables ... that is the primary advantage ! ... but there are many other important features which allow FBW to be chosen by manufacturers (i.e. computers fit better with FBW than traditional flight controls and have a lower cost).

:ok: Best Regards,
captain87

[Mad (Flt) Scientist]

Also, my mate (a test pilot on Eurofighter/Typhoon) told me that his aircraft was the only one he knew that did have upload from the back end.

Really? Then perhaps you should ask him to ask someone what the effective lift coefficient must be on the Gripen canards, if they are all that provide upload on that aircraft? just to pick one example.

Aft cg saves fuel by reducing trim drag, not by making the aircraft less stable. The stability of a configuration is not dependent upon its trimmability.

[Brian Abraham]

One of the advantages of a canard layout - where the horizontal stabiliser is forward of the main-plane, is that a canard is inherently stable in pitch but, nevertheless, produces positive lift from its stabiliser, thus reducing the lift requirement of the main-plane.

Thats true, but the problem with a canard configuration which some times gets overlooked is that the mainplane can not operate close to its CLmax because of the requiremjent for the fore plane to stall first, hence the mainplane tends to be over sized. No free lunch.

[Chris Scott]

Thanks for your point on the inherent limitation of canard main-plane lift (CLMAX), Brian Abraham. I guess it doesn't matter at cruise speeds, e.g., VMIN-DRAG or above, where main-plane alpha would be only 2 - 5 degrees?

fantom has identified at least one point that I don't follow in Mad (Flt) Scientist's admonition of my pitch-stability argument. M(F)S seems to be overlooking the fundamental fact that, in both canards and conventional aeroplanes (forgetting hybrids, like the Gripen), the C of G is normally FORWARD of the centre of lift of the mainplane, assuming natural pitch stability is required.

So on a conventional aeroplane at 1G (one with what I have called a tailplane), a CofG near the aft limit means the download on the tailplane is lower, the mainplane lift for 1G flight is (consequently) also lower, the fuel consumption is lower, BUT the A/C is less stable in pitch – as fantom points out.

Continuing with the 1G case; as the CofG goes forward, the tailplane download increases until, ultimately, it will stall and the aeroplane will depart in bunt. This would happen sooner with a small, efficient tailplane.

On a canard, the aeroplane presumably also becomes more stable as the CofG goes to the forward end of the envelope; the upload on the foreplane being higher. So in both types of configuration, I propose: maximum pitch stability is achieved when the stabiliser load is high; minimum stability when it's low; instability when its load (lift direction) is reversed.

STICKING WITH CONVENTIONAL AIRLINERS

Quote from M(F)S:
[I]take a REAL aircraft. Trim in 1 'g' flight. There's probably a small download on the tail, especially if we have a fairly aft cg... [Unquote]

An aft CofG REDUCES the download that the tailplane needs to generate for 1G flight.

Quote from M(F)S:
...now start to bunt. Push forward on the stick. The load on the tail reduces and eventually will be negative. The aircraft doesn't suddenly become unstable when the sign of the tail load changes. [Unquote]

It's normally negative, so I presume you meant positive? It may reverse and become positive, but only because of the down elevator. As soon as the elevator is neutralised, the tailplane will produce negative lift (download) again, restoring pitch stability. If you maintain down elevator and existing stab trim, the aircraft will obviously continue to fly at less than 1G (i.e., bunt).

Broadly speaking: for a given IAS, slat/flap/gear configuration, thrust setting (i.e., thrust pitching-moment), and CofG position, there is only one tailplane (stab trim) setting for 1G flight, unless you are happy with a constant elevator input. [The trick is to find it.]

[PantLoad]

Gentlemen:

I personally love the FBW system of the SA Airbus. Coincidentally, there is an active thread going on regarding simulator suggestions for the Bus. One says not to touch the stick. Generally, this is good advice for sim problems, but it should not be construed to mean that hand flying the Bus is a problem. In fact, the SA Airbus is a real joy to hand fly...especially the 320. (I fly the 319 and 321, as well.)

The replies from others in this thread are correct. FBW saves weight, it's more reliable, it offers protection (through computers), it offers stability, etc. Actually, none of the limitations (invoked by the flight computers) is an issue during normal regimes of flight. (Normally, you wouldn't want to do rolls in a transport category aircraft. In the Bus, the computers won't let you. I don't find this constraint overly restrictive.) :O

Of course, in many modern fighter aircraft, the computer flight control system is necessary for stability. In the Bus, it's not. The FBW system merely makes the plane easier and more enjoyable to fly. And, I find my flying is more precise with FBW...thanks to the Airbus system...certainly not due to flying proficiency on my part....:{

Anyway, that's my two cents on the subject....


PantLoad

[Milt]

PantLoad

Not often do we see something like a good handling assessment. I've been asked to produce one for P-51 Mustang. The last I partially did was for an RAF Comet 2C - and was it ever a dog having powered flight controls with large breakouts into spring feel.

Twice flew an A300 out of Toulouse in R seat and was introduced to the 'standard performance take off' with runway temp and other relevant factors taken into account.

As a TP with minimal experience in handling side stick FBW some comments on the feel of the Airbus side stick controllers will be appreciated particularly as my experience as a TP dictated that the SF/g for a heavy military could not be less than 11 pnds/g when using a two handed yoke. So for your controller

How far does it move from full forward to full aft and what is the force required to input full aft.? Is there any feel of IAS?

Presumably the pitch feel is well harmonized with roll but don't you need differing roll forces to be compatible with your differing wrist strength left and right.

With controller free does the system freeze the attitude?

Is your trim controller a 4 way switch which you operate with your thumb and/or do you have auto trim?

Presumably also all of your auto stabs and limitations are auto but maybe with some variations available.

How can you cope with electrical power failures and computer malfunctions?

Was a simulator conversion to side stick adequate or did you need some air time?

Smooth landings

[PantLoad]

Milt, thanks for the response.

My only Bus experience is with the 319/320/321. Don't know anything about the earlier models (300, etc.).

I don't have numbers regarding forces required to move the stick....probably could dig into the FCOMs to find this...not much value, really. But, in my humble opinion, there is not much 'feel' vis-a'-vis airspeed, loads, etc.

Trim is automatic...and it doesn't matter what you're doing...it trims. For example, you want 20 degrees of left bank, and ten degrees nose up...well you 'put it there' then sit back and sip your coffee. That's pretty much where it stays. The exceptions are unusual bank angles like, I think the bank limit is 32 or 33 degrees...beyond which, you have to continually hold the stick to keep it at bank angles beyond the 32/33 degree figure. (Can't remember things too well, anymore.) There is no trim switch...except for the rudder. You have a stabilizer trim wheel...you move it by hand, while in direct law. Never been in direct law (or alternate law) during normal line operations, so, other than in the sim, I never used the trim wheel.

Regarding electrical failures, you can fly the plane....it takes several massive, momumental failures to get into a situation where you have no electrics. It'll fly.

But, hydraulics are a different matter. You really need at least one hydraulic system to keep the shiny side up.

To be blunt, despite all the crap they do to you in the sim, the Bus is quite reliable, and the systems are quite redundant. Had, on only one or two occasions, a computer failure. Each time, it was a non-event. (To me a 'non-event' means you pretty much ignore it and just keep on flying like everything's normal. Write it up when you get on the ground.)

My conversion course (type rating, as we call it in the States) involved a sim that did everything. In other words, the first time I flew the actual airplane was with passengers on board and a line instructor in the right seat. It was no problem. (I would never have made a P.A. announcement stating such...but that was how it was!)

The long and short of it...in my opinion, the Bus is a joy to hand fly. Once you get used to the stick and the handling, it's a wonderful piece of engineering and art. I find it to be a particularly 'honest' airplane...and easy to land (especially the 321).

PantLoad

[Chris Scott]

Milt,
I'm probably suffering senile dementure, but still can't work out what PantLoad means by "SA" Airbus - there used to be a Toulouse-Blagnac (Caravelle) manufacturer called Sud Aviation, a predecessor of Aerospatiale, but I don't suppose that's what you've got in mind, PantLoad?

Anyway, following on from PantLoad's excellent description of the A320 family (the 330/340 are supposed to be similar), you can find a bit more on handling matters at the following link on PPRuNe:
http://www.pprune.org/forums/showthread.php?t=308926&page=3
Being biased, I suggest #56, #65 - 68, and #71.

Re the sidestick, it's worth mentioning that there is a special armrest, adjustable in height and rake, for each pilot. The idea is to rest the flat of your forearm on it, wrist slightly raised. You do not touch the stick until you need to make an input.

For simple roll-rate demands, you can use your thumb and 1 - 3 fingers. For up demands, use your fingers; for down demands, your thumb. To do these different tasks, your wrist will twist to the necessary angle, but will not move up or down, or significantly sideways. Inputs are normally the most successful when small, short-duration, and frequent. The obvious exceptions are rotation on take-off, and a simple roll from one side to the other.

The stick can be moved in any direction, of course. I believe you are right: the stick does take account of your hook being stronger than your slice. Although the pitch and roll-rate resistances are well harmonised, combinations of the two are the easiest to get wrong. The sticks do not move to reflect autopilot inputs. Normally the signal to the EFCS is the algebraic sum of the 2 sticks. But if one pilot wants to take control, he/she can press a red button on the stick (eliciting a 'gringo' announcement and warning lights). The red button is also used for AP disconnect.
PantLoad has answered many of your other questions.

With controller free does the system freeze the attitude?
Generally speaking, yes, but it will not return the aircraft to that attitude after it has been disturbed by a gust – you have to do it.

How can you cope with electrical power failures and computer malfunctions?
There are 5 flight control computers, plus 2 augmentation computers (FACs) that do a lot of useful calculations for the EFCS. of the first 5, there are 2 ELACs (elevator-aileron) and 3 SECs (spoiler-elevator). In the early days, we often used to have failures of a SEC, but rarely of an ELAC. Each has a PB on the overhead panel, but – if cycling that didn't work – when convenient we used to reboot by tripping the CB, waiting 10 seconds, and resetting it... I think they may have made many of the CBs inaccessible on the latest models! I had about 5 lightning strikes in 13 years and never lost a flight computer (or any EFIS).

Loss of more than one computer, I THINK, causes the FBW to degrade from Normal Law to Alternate Law. There is still auto-pitch-trim, but some of the protections are lost. [Forgotten, ask someone current.] The next degradation is to Direct Law, using manual pitch-trim. On the A320 family, I think Alternate Law always degrades to Direct Law when the L/G goes down. [Something to do with functions being switched to battery power, because the hydro-electric RAT can be shielded by the nose gear in sideslip]. Finally, if you are having a good day, you should be able to stuff it on the runway using stab trim and rudder only (hydraulics permitting), if you lose all the computers. But the idea is to get one or two computers restored first.

Presumably also all of your auto stabs and limitations are auto but maybe with some variations available?
Very few variations, I'd say. You can exceed the normal max bank (36 degrees?), if you hold the stick over and keep holding it, to the 67-degree limit (2.5G). You can NEVER exceed +2.5G or -1G. You can NEVER stall the aeroplane, but you can reach and hold alpha-max if you keep the stick back. [That's why someone was able to do that perfect wings-level crash at Habsheim in our early days. They had relied on the alpha-floor function of the A/Thr to apply TOGA thrust automatically, forgetting that it never does that below 100 ft.]

As PantLoad says, it's easy to land. The Normal Law modifies in pitch in (roughly) the last 100ft, to simulate the increasing elevator control load of a conventional aeroplane in the flare. After the flare, a slight de-rotation helps, as on so many types.

Just as I immediately dispensed with the conventional electro-mechanical ASI during my first simulator session on the A310 (1984), I found the sidestick was second nature within an hour on my first A320 session (1988). Wish I could say the same for the non-driven throttle levers... Bernard Ziegler (Airbus chief design engineer) wouldn't budge on that one. But they are a conventional joy in manual thrust, which I almost invariably used for manual landings. Trouble is, all the airlines are banning manual thrust now.

DISCLAIMER ! [Airbus pilots] Don't believe any of the above. It's 6 years since I last flew. [U]Read your FCOM. Good flying.

[PantLoad]

"SA" means single aisle

Yes, in direct law...heaven forbid you should ever experience this in real life (joking...)...the airplane flies like a conventional aircraft. There is nothing to worry about, no need to declare an emergency...in direct law, the Bus becomes a 737.

Correct...Alternate Law goes to Direct Law with gear extension...except in the case of "Abnormal Law"....not really a consideration...just one of those questions in the Pub for a free beer.

Unfortunately, many carriers now mandate the use of Auto-Thrust. I guess I can understand why, as the Bus' Auto-Thrust system is magic.
But, I still like to hand fly the plane...still like to control my own thrust...If you want to disconnect the Auto-Thrust for approach and landing...and your company's SOP doesn't dictate otherwise...be sure to do it before 1000 feet RA. (Airbus recommendation)

Overall, the Bus is a good, solid aircraft (in my opinion). Been on it for almost six years, now, and I love it.

PantLoad