View Full Version : A330 - Longitudinal Stability

Neptunus Rex
10th Jun 2009, 10:47
Trim Drag is the drag produced as a by-product of the extra lift necessary to compensate for the aerodynamic download on the tailplane (nothing to do with trim tabs protruding into the airflow.) In the A330 Trim Drag is reduced in the cruise by pumping fuel aft into the tailplane tank, thus reducing the required downward lift on the tailplane. The aft transfer takes place automatically as the aircraft climbs through 25,000 feet and, if any fuel remains in the tailplane tank during descent, it is pumped back into the mainplane tanks as the aircraft descends through 25,000 feet. The resulting decrease in Total Drag should give up to 5% reduction of fuel burn in the cruise.

However, you don't get it for nothing. The result is a reduction in longitudinal stability, which is why it is restored prior to landing.

Hand flying an A330 at altitude in Normal or Alternate Law in benign conditions demands a high degree of skill and concentration, considerably more so in Direct Law. In moderate or greater turbulence it becomes progressively more difficult. Should consideration be given to inhibiting or reversing the aft fuel transfer when severe turbulence is forecast for the planned route and so give the pilots more positive longitudinal stability to help them handle the steed?

Perhaps a Test Pilot could comment on the resultant difference in handling qualities.


10th Jun 2009, 12:39
Neptunus, what is it that makes a 330 so hard to fly at high altitude in normal or alternate law, compared to others? In a situation such as Air France, maintaining an exact altitude would not have been their priority. Maintaining a mach range would have, obviously. Other than the aft CG, is it really any harder to hand fly in these degraded modes than say an old DC8, or newer 747-400?
Anyone tried in the sim?
5% fuel reduction sound HUGE

Woodpecker, I think past posts have pretty well put to rest the myth of today's airliners in normal flight conditions having a tail that produces positive (ie upwards) lift.

10th Jun 2009, 17:28

An "Aft C of G" doesn't mean the tail has to produce force in the up direction for straight and level flight, unless you're loaded well outside allowable limits.
Assuming the aft C of G position is within the W and B envelope of the airliner, it just means that the tail is required to produce less down force than for a forward C of G. Less down force = more fuel savings, among other things. The use of the term "aft" just means the C of G is towards the rear of the allowable range on the MAC. Similarly, "forward" means the C of G is towards the front of the allowable range on the MAC.

We're not talking F16s, canards etc

Brian Abraham
11th Jun 2009, 05:46
Woodpecker, you will need to get your aerodynamic books out and revisit stability. No airliner built has a tailplane producing lift in an upwards direction, irrespective of the C/G position. An aft C/G is desired to reduce trim drag, but the tailplane will always be producing -ve lift. The Airbus family have what is known as "relaxed stability", the computers providing for the shortfall in aerodynamics. On the 380 for example they were able to reduce the size of the tailplane area by 10% and reduce trim drag by .5%, which in turn lead to a reduction in weight of 1,500 pounds. With the loss of all computers the aircraft is able to be flown by use of the trim wheel, the tailplane providing stability. If the tailplane were producing +ve lift in this case the aircraft would be unstable (refer to the moments and couples in your aerodynamic notes).

11th Jun 2009, 07:17
The fuel penalty to dispatch with the trim tank transfer inop according to the 330 MEL is 1%...

11th Jun 2009, 10:33
Maybe this "relaxed" stability referred to by Brian is the reason the 330 is so difficult to hand fly at high altitudes, in direct and alternate law, as mentioned in the first post by Rex. Anyone ever try this (not in RVSM airspace)?

1% fuel penalty. Sounds more reasonable than the 3-5% mentioned in the link from Woodpecker. But....maybe the article meant the difference between full forward and full aft is in the 3 - 5% range

11th Jun 2009, 13:07
The turbulence procedure on the A300-600 was to transfer any remaining fuel from the trim tanks to the main tanks. As we know,the A300 was a conventional aircraft. As for the A330,we dont have any procedure such as this because in Normal and even in Alternate laws,the FCPC and FCSC logic (as best it can)take care of any turbulence (flight path excursions) making the FBW aircraft a stable platform.
I can only imagine what the aircraft behaviour will be like with system degradation,full trim tanks and moderate to severe chops at altitude.



11th Jun 2009, 13:33
Can you tell us briefly what FCPC and FCSC logic are? And in degraded mode, (is this Direct mode?) would the 330's handling be any different than non FBW aircraft with conventional controls? eg 767/747?

Green Cactus
11th Jun 2009, 15:37

exactly which sim are you flying?

I sincerely hope that is all the "flying" you do....


11th Jun 2009, 15:50
Good point Woodpecker. Note that the quote can be attributed to Brian Abraham, not myself. And I think you're right, the tail can in fact produce positive lift, ie in the up direction, with the appropriate trim/yoke position. Brian's comment was probably meant to include the phrase "in straight and level flight"

Not sure what Cactus means...lots of pilots believe the tail produces up force in flight....nothing unusual there....

Air France....set a power setting, set a pitch, and hope for the best (??)
Use aoa (if supplied)?

11th Jun 2009, 15:56
If the tailplane were producing +ve lift in this case the aircraft would be unstable - you'll need to elaborate on that, Brian. It doesn't matter which way the tailplane is working, a change in AoA produces a stabilising tail moment. Basic P of F?

Woodpecker - I think you are confused - you have merely proved that if you push the stick forward, the nose tends to go down (or 'tail goes up' if you prefer!):) The event you describe is achievable on most a/c - it was called 'wheelbarrowing' in my day, and without a nosewheel-----------

PS - great fun in a Harrier as one of the OCU 'frightener' videos showed. (Yes, I know, it was all done by puffers)

Will Fraser
11th Jun 2009, 16:15
Some are overreacting. 'Aft loading' means less downforce is needed from the tail, the AoA of the wings is reduced (by a very small amount), meaning less drag, better fuel specifics. It is that simple.

11th Jun 2009, 19:55
Hi Hawk37

This will answer Neptunus as well. The A330 will go thru a rough ride if the FCPC's are lost due to failures and having a full trim tank. The Turbulence Damping or Alleviation function use the PRIMS to compute a turbulence damping command,which is added to the normal law command for the elevator and yaw damper. The alleviation system on the airbus addresses the Aeroelasticity problems of bending and twisting due to pilot inputs or atmospheric disturbances.
The way I see why we get a fuel benefit by having an aft CG. Aft loading= less downforce on the tailplane=less apparent weight the aircraft experiences=less lift required to keep the A/C flying at a constant AOA=less thrust required to maintain level flight...which amounts to fuel savings or an increase in range.



Will Fraser
11th Jun 2009, 20:03
'less downforce on the tailplane......' sigh. The tail has downforce increase due to the added weight aft. This requires the tail to produce less downforce aft. The net effect is the tail provides 'lift' not required by the wings, reducing the wing AoA, hence less drag for the same fuel burn, greater range for equal fuel. K?

Brian Abraham
12th Jun 2009, 03:35
I dragged out my 50 year old course notes and also some current course notes and was surprised that they all talk about +ve tailplane lift. Talk about having to give technically the wrong answer in order to pass the exam, but thats nothing new, equal transit time for the production of lift is another.

What was missing in all the course notes was any discussion about the CofG and the neutral point*. Remember that the CofG lies ahead of the wings centre of pressure, or aerodynamic centre, requiring a tailplane download to balance the moments. For an aircraft to have +ve longitudinal stability the CofG must be ahead of the neutral point. As the CofG moves aft (closer to the neutral point) it means the tailplane must produce less downwards lift. If the CofG were to lie on the neutral point the aircraft has no +ve stability and will retain whatever attitude to which it is disturbed. With the CofG aft of the neutral point the aircraft is unstable. This feature has been made use of in the modern fighter to increase manoeuvrability, the F-16 being the first. Of course this means you need full time computers to control the thing and loss of those means an ejection, a feature that civil certification authorities are somewhat adverse to.

The degree of stability is referred to as the static margin. Some figures I found on the net.
Cessna 172 0.19
Learjet 35 0.13
Boeing 747 0.27
P-51 Mustang 0.05
F-106 0.07
F-16A -0.02 (subsonic) .23 (supersonic)
F-16C 0.01 (subsonic) .26 (supersonic) increased stability due to larger tailplane
X-29 -0.33

* The neutral point is that point on the aircraft at which all the lifting forces are considered to act ie lift produced by the wings, tailplane, fuselage, nacelles etc. In other words its the aerodynamic centre of the aircraft as a whole.

12th Jun 2009, 07:51
Some observations.

There is a fair bit of confusion regarding static stability (which is what most appear to be talking about). The usual pilot training concept of balls in tea cups doesn't help much, unless you "get" the idea of looking at the force required to hold the ball away from the bottom of the cup and the associated desire for the ball to go to the bottom if you let it go.

In essence the following pilot-view requirements exist (considering an essentially constant altitude scenario) -

(a) trim for speed .. ie set up the aircraft to fly hands off

(b) without retrimming, you MUST require a stick PULL force to fly SLOWER, with the force increasing as the speed decreases.

(c) without retrimming, you MUST require a stick PUSH force to fly FASTER, with the force increasing as the speed increases

For (b) and (c) the higher the stick force gradient, the more stable is the aircraft.

(d) there must be no reversal of the stick force gradient (which is where SAS becomes necessary to fool the pilot into thinking that the plane is stable. In its simplest form, SAS is a variable downspring providing a preload to the elevator circuit).

(e) if you release the stick force, the aircraft must return to a speed somewhere near the original trim speed.

(f) the design rules impose numerical boundaries for these requirements but the line pilot can presume that the certification process takes care of acceptability (ie "trust me, I'm from the government ..").

A statically unstable aircraft can be flown by hand if you know what the situation is and know what you are doing but it is difficult, demanding, tiring and unacceptable for certification. For most pilots, a statically unstable aircraft is an aircraft looking for a place to crash.

Dynamic stability looks at making sure that the short period oscillation characteristics are heavily damped and that the phugoid characteristics are benign.

A dynamically unstable aircraft is not amenable to hand flying by a human pilot.

As a general rule, moving the cg forward increases stability while moving aft decreases stability.

There are numerous reference books on the subject and numerous web references .. all containing lots of mind numbing mathematics .. but you can find out what static margin and a bunch of other buzz words mean.

Alternatively, download AC 25-7A from the FAA website and search for the buzzwords. The text describes what the flight test folk do to establish stability measures during the developmental and certification programs.

Now, if you have some gee-whizz computers between the pilot and the aeroplane, you can look at reducing the inherent stability ("relaxed" stability) by virtue of the fact that the computer can sample and react far faster than the pilot ..ie the computer can fly an aeroplane which is considerably less stable than would be flyable by the human pilot.

However ... if those computers go "tilt", the military fast jet pilot has a panic handle option to live to fight another day ..... the civil pilot just has an enormous increase in workload and probably not much of a chance of winning a medal.

12th Jun 2009, 08:17
Apologies, Brian, I thought you were focussed merely on the pitching moment from the tailplane when you referred to 'unstable' (and obviously for that to occur the C of G would have to be aft of the tail.........). All agreed.

I believe this began as an AF447 topic, and apart from 'thesilver' this query has not really been addressed. IF the 330 design allows significantly reduced longitudinal stability RELYING on the FBW to assist the pilot ....answers on a postcard please, as JT says.

Now, anyone for canards.........................?:)

Brian Abraham
12th Jun 2009, 09:26
This may go some way towards answering Neptunus Rex's original enquiry. The article refers to the A320 pre production. The reference to flying manually in turbulence, or when other things have gone wrong, is interesting.

http://www.flightglobal.com/pdfarchive/view/1981/1981 - 1770.html

Traditionally, fail-safe pitch stability has been assured by keeping the aircraft's e.g. forward of the centre of lift, the required nose-up moment provided by negative lift at the tailplane. Canard stabilisers are the traditional answer to the illogical idea of intentionally producing negative lift anywhere on the aircraft. The idea now is to position the resolved lift and weigh centres at the same point; but this makes the aircraft harder to fly manually—it becomes "twitchy"—and it would no longer have a natural tendency to pitch nose down at the stall, thus recovering.

But who will fly these new-generation aeroplanes manually, an overconfident electronics engineer may ask? The pilots will say that they want to be able to, and if the potential passenger were to hear that only a machine stood between his arrival or non-arrival at his destination, Airbus would have a job marketing its product.

But the relaxed stability aeroplane will come, and the A320 will be as close to relaxed stability as technology and safety allow. Britain's Boyal Aircraft Establishment has done quite a lot of work on the idea by moving sandbags about in the cabin of a BAe One-Eleven, and it says that flying the machine manually is not the problem it is made out to be. There remains, however, the problem of what it is like to fly manually in turbulence, or when other things have gone wrong and crew stress is high. What Airbus will probably do is leave the e.g. slightly ahead of the centre of lift and use a smaller, lighter composite tailplane. In fact the final tailplane, unlike the one shown in this feature, may look slightly odd to traditional aviators' eyes, because it will be "out of scale" from the rest of the machine.

There is an alternative to sandbags for moving the e.g. to the required position—fuel. But Airbus will not move fuel about in the A320. It costs too much to fit the systems. Though the engineers are there with the ideas, the accountants still hold sway, and Airbus is proving very astute at getting the cost/design compromise right for the market.

12th Jun 2009, 09:42
The OP talked about Hand flying an A330 at altitude in Normal or Alternate Law in benign conditions demands a high degree of skill and concentration, considerably more so in Direct Law. In moderate or greater turbulence it becomes progressively more difficult. required to hand fly the aircraft at high altitude..........

It would be easy to confuse "difficulty" to handle the aircraft accurately at high altitude and............ stability, and a whole host of other issues.......

Fundamentally, All aircraft are more difficult to fly at high altitude because of some pretty basic physics. Even those with a high degree of positive stability.

Think, Effects of Controls. High TAS, low IAS at altitude and momentum.

All this happens irrespective of CG, dynamic stability, computers, FBW etc etc...

Throw the above into the mix and it all gets incredibly complicated..........