Max fuel imbalance demonstration
 

G'day flight testers,

Just came up in the type rating as a discussion.
AFM shows max fuel imbalance 360 kg (some 800 lbs).

The question is: What if a leak empties one of my wings completely. Low wing aircraft engines underslung like E190/A320/737.

FAR 25.343 hints in the direction. But I was not able to find more specific demonstration requirements. Any of you able to enlighten me on how this is tested?

I would expect the aircraft to be flyable for a landing with maybe major trim changes on the aileron and rudder but still manageable for a landing.

Thanks for enlightenment

The maximum fuel imbalance is usually calculated by the Flight Sciences group as a reasonable limit (from offset lateral CG) and set in the EICAS/ECAM as a Caution message. The flight test is to show that the AFCS can cope with that offset during critical maneuvers such as during approach - for example fuel offset is usually one of the limit conditions flown during the 30 approaches to demonstrate that the Autopilot meets CAT II requirements. Don't forget that most systems return unused fuel continuously back to both wings - the amount held in either one is determined by the relative input & output flows. A fuel leak is flagged by the difference in fuel used (flow into the engines from flow meters) relative to the total fuel used (from the Quantity & Guaging system's level sensors) and is as likely to be by the difference in rates as actual quantities.

Something you won't find in the regs directly, but hinted at in the reply above, is keeping the imbalance to a "reasonable" value so as to not need excessive trim or (worse) control input. There are a number of reasons for this. One is crew workload in the normal circumstances, but also important is the implications for failure cases. If you're going to be considering various failures that require lateral control inputs, you can't afford to be giving up control authority to account for theoretical imbalances. And all forms of trim input - whatever the cause - are hated by the performance people, as they represent more drag, so again it makes sense to keep them small if you can without too much pain.

To the original point of a leak draining one wing, it all depends on the circumstances. If you know that range isn't an issue, you may even transfer into the leaking tank. If range is critical, you'll be burning the leaker as fast as you can and trying to save the good tank, so creating a bigger imbalance. Only when you start feeling a control issue would you stop that (and that's a lot more imbalance than any messaging) - and trying things like flying faster to gain more control power would be worth trying in those circumstances.

Great Guys.

That's exactly what I thought it was. It's kind of the same thing as with "demonstrated" values. Up to the caution imbalance one is within reasonable limits. Once outside the normal imbalance that would be an emergency case with all the implications as stated above.

From the flying point of view I was just interested if there is a requirement to demonstrate in flight the maximum imbalance.

Obviously on an operational side one would try to fly to a place offering the best runway/wind combination, land with the least possible flap setting and the highest possible speed for controllability.

I have personally flown an aircraft with 60 % over the AFM limit imbalance. 1'000 kg instead of 680 kg, at some point 1'200 kg in flight. As it was an anhedral wing the moment arm of the imbalance gets shorter vs. the actual aircraft which is conventional dehedral where the moment arm increases with imbalance.
The actual case required some 10% of possible aileron trim, so I guess that landing that aircraft with one wing tank empty would be possible within reasonable limits.

Someone with experience about the limits. e.g. for a 737 one empty, one full?
From flight testing?

737 with one wing full and the other empty, at least in the sim, is a non event and not really noticeable for landing. Haven't tried it for real but assume the simulators are accurately modelled.

In the flight crew training manual Boeing have a page or so of text which points out that the limitation is for long term structural protection, and caution crews not to be to hasty to start opening cross feed valves as you may well make it worse than it already is.

Agree with the above, the 737 flies perfectly normally with one wing full, one empty and approaches/lands with about a third deflection of the yoke in roll. A non-event.

arcdu, Wageslave, take care in relating what a simulator might do and a flight assessment based on previous knowledge that an imbalance exists.
I recall a 737 incident many years ago (UK AAIB ? – looking for the report), where fuel ‘mismanagement’ – fuel transfer remained open, resulted in a gross fuel asymmetry. The autopilot was in use with FMS input; the crew noticed poor track keeping / failure to turn at a waypoint. On disconnecting the AP the aircraft rolled 120 deg before recovery.
One aspect of the surprise was that the crew attempted to zero the roll stick positon / force vs the opposing force to achieve wings level. Few crews are familiar with flying ‘out of trim’, particularly in roll, where although the sick positon / force and control demands are relatively low / easy to handle, the situation is so unfamiliar it conflicts with the norm – ‘cognitive dissidence’.

I used that example as guidance in flight tests, together with a 4 sec pilot intervention recovery delay (cruise) from an inadvertent AP disengagement – a double failure, but its guidance. My aircraft also had an AP roll demand limit light, whereas I recall that some versions of the 737 only had ‘torque’ indicators or nothing at all.