CG out of AFT limit

[N.N.C]

Yesterday we received the final figures before take-off and we noticed a very strange cabin distribution. (Since we make our own loadsheets at dispatch, we assume average cabin distribution.)
We completed a manual loadsheet and it turned out that we were just out of our aft cg limit.
Off course we reseated pax to obtain good distribution and calculated a new mac and load&trim sheet.

We were discussing all possibilities had we not discovered this checkin error.

Tailstrike during rotation

Difficulties in attitude control (unable to control and reduce pitch)


I am wondering how dramatic things can turn out on those
a: aft cg (out of limits) and
b: trim set to average cabin distribution

Is there any (sim) experience?

[PaperTiger]

I am wondering how dramatic things can turn out Smoking hole :eek:
Have a look through: http://aviation-safety.net/events/cgc.shtml

[ETOPS]

N.N.C.

The controllability in pitch of an aircraft is dependent on the distance between the centre of gravity and the effective hinge line of the elevators. This forms a lever arm and the longer it is the more stable the aircraft. If you move the c of g too far aft it can reduce the effective lever arm to such an extent that it passes the stick-free neutral point. This is the point beyond which it is no longer possible to trim the aircraft for stable "hands off " level flight. Any displacement in pitch will become divergent and loss of control and/or structural damage follows faster than pilot reaction time.

There have been a number of such accidents over the years which reinforce the need to check your load and balance every time...........

[mutt]

If you were flying a Boeing, then the answer to this

We completed a manual loadsheet and it turned out that we were just out of our aft cg limit. ........ I am wondering how dramatic things can turn out on those

You wouldnt have noticed, the cg envelope has quite a numbers of buffers built in to it, you therefore wouldnt not have reached the actual aft CG limit.

However, if you had ended up in a court of law, things would be slightly embarrassing for you..

Mutt.

[john_tullamarine]

Further to Mutt's post, assessing the trimsheet calculation depends on

(a) the philosophy embodied in the trim sheet.

Usually the sheet designer builds in some fat within the apparent envelope to cover the error analysis which he/she ought to have done as part of the sheet workup. In this case Mutt's comment is correct ... if the calculation shows you just outside the sheet limits, the actual calculated CG might well be inside the AFM envelope.

Conversely, some designers don't bother and the trimsheet is a direct recast of the AFM/W&BM data. In this case, you don't have the comfort of some fudge factors ....

(b) the extent to which you use standard weights (masses) and actual in-service loading asymmetries.

It is usual to approximate various things in designing a sheet to make it manageable... things like using loading zones in the passenger and baggage areas, standard passenger and baggage weights, standard fuel SG, etc. If this is done, the calculation relies on the actual loading being similar to the presumed loading incorporated in the sheet design. In addition some allowance should be made in the error analysis to accommodate such problems ...

(c) the accuracy of your empty weight.

Mostly aircraft are weighed on jackpad cells, many of which are a bit suss in accuracy .. all are very critical with regard to jack axial misloading. I did a specific study into this many years ago and the extent of cell errors, in the face of careless weighing procedures, is quite frightening .. and you thought that your empty weight and CG is accurate ..... don't bet your life on it ...


From an engineering point of view, the aircraft won't suddenly become dangerous as you transit the published limits ... however, the limits are, in effect, lines in the sand within which things are considered acceptable and/or good, and outside of which, things progressively become unacceptable.

The legal situation is, however, somewhat more black and white. Really, do you need the pressure of having to account for slapdash loading at the enquiry ?

Speaking to the first post, it sounds like your company loading procedures generally need to be revisited .. the situation ought not to have occurred ....

[dolly737]

Hi NNC, here’s the experience:

My colleague had just finished his “V1”-callout as the nose violently started pitching up. We were instantly airborne with about 25° pitch, increasing. Pushing the yoke almost to its forward limit plus simultaneously starting trimming was just enough to regain adequate control...!!!

This happened on a 737-300, when 400kgs of cargo were loaded into the aft hold, but, on the loadsheet, trimmed in the forward hold (and this, btw, is NOT a sim-experience)! Recalculation revealed that we were only barely out ot limits!

You might also consider that on top of pure control-problems due to out of CG-limits, you also may have the aggravating effects of a wrong stab-trim setting.

@john_tullamarine:
And, yes, heads were rolling at this outstation and procedures were adopted!

[N.N.C]

Thanks for your replies.

@mutt: Would you happen to have any official documentation for the B737-800. E.g. certification or AFM cg-limitations?

@dolly737: which and in what way were your procedures adjusted?

[john_tullamarine]

Dolly737,

The 73 is a bit sensitive on gross stab setting errors ... with an aft CG and stab set for forward CG, the aircraft will rear up well below typical V1s and there is precious little you can do other than abandon the takeoff in a rather untidy manner ... a fairly interesting sim exercise to see if the guys are thinking ...

[dolly737]

john_tullamarine,

the exercise might become even more interesting, if the F/O is PF and, according to SOP only the captain can abort the takeoff, plus add, for whatever reason, a V1-reduction ;) . A lightweight 37 on a full thrust t/o will become airborne quite early (already when rotating thru around 5-6°), which will give you about 1 second to notice(!) that something is not right :eek: .

Would be a VERY untidy RTO indeed...

[john_tullamarine]

... first time up in the sim on endorsements .... invariably causes the operating pilot to say something along the lines of "oh golly ... gee .... gosh ...." ..... generally in a rather surprised tone of voice.

After a second go, both pilots are very aware of the problem .. if nothing else it engenders a healthy respect for misloading and mistrimming.

[Genghis the Engineer]

With the disclaimer that I know absolutely nothing about Boeings - I'm a little aeroplane expert, the effects of being out of aft...

- Overly light pitch controls. In a manually flown aeroplane, this increases pilot workload, in a computer flown aeroplane, this may muck up the control laws and cause a loss of control.

- Increased risk of a neutral or divergent SPO.

- Increased risk of a PIO due to probably more rapid pitch response.

- Inability to maintain airspeed (anybody who has flown a CH601 will be very familiar with this).

- If manually flown, increased pilot workload to maintain speed and height, which can cause reduced effort elsewhere for other tasks.

- I doubt you'd get a take-off tailstrike with a manually flown aeroplane because the pilots self-protection mechanism should stop him over-rotating beyond what he's used to seeing out the window. An AFCS may not be quite so clever. Of course- with a very aft CG the pilot may not have the chance if it over-controls too much, but I think that the inertia of a big jet should prevent this, which I'd consider a little aeroplane problem. A landing tailstrike is more likely because you tend towards full elevator in the flare which you don't tend to in take-off.

And anybody's sim experience will almost certainly be useless in terms of what really happens out of limits. Sims are produced using safe middle-of-the-envelope FT data, and virtually never with out of envelope data. The computer will try and extrapolate, but the odds are it'll get things wrong.

Incidentally, when we certify light aircraft we invariably (not written down anywhere, just good practice) put some fat on the CG limits - typically ½" at aft minimum. I don't think this practice is so common in the big aeroplane world due to a combination of a commercial need to make aircraft as flexible as possible, and the assumption that controls on how the aircraft are loaded and CG calculated are much tighter than in your average part 23 aeroplane.

G

[dolly737]

Genghis,

- I doubt you'd get a take-off tailstrike with a manually flown aeroplane because the pilots self-protection mechanism should stop him over-rotating...

Well, and still it happens, more often than you might think. And as john_tullamarine mentioned, some airplanes are quite(!) a bit sensitive on stab-trim setting errors..., combine that with an aft C.G. and you'll be surprised how violently the nose will come up before you even have thought of beginning your rotation.

...that the inertia of a big jet should prevent this,...

I rather tend to believe that it's just that inertia - assuming an already stabilized x°/sec rotation - which adds to the fact that you need a push on the yoke to stop or decrease the rate?!

...And anybody's sim experience will almost certainly be useless in terms of what really happens out of limits...

I totally agree, only what I was describing happened in REAL life! And having someone going through this in a sim will, of course, only be of any interest in regard to HOW the crew deals with the problem of being faced with an possibly uncontrollable aircraft right at V1 or even during rotation.

[Genghis the Engineer]

Yes, good question on inertia - does the relatively slow pitch rate give the pilot time to react, or does it mean by the time he (or she) reacts it's too late?

G

[dolly737]

Genghis,

Some facts first:
Boeing recommends a rotation rate of 3°/sec. On the 738, liftoff under normal conditions occurs at around 7° pitch. For optimum takeoff and initial climb performance, a continuous(!) rotation toward 15° is recommended. Aft fuselage contact occurs at 9,2° with struts compressed and 11° with struts extended.

Considering that before you react you have to realize that something’s not right makes the rotation of a tailstrike-prone aircraft a delicate manoeuvre. Under normal cirumstances, you have enough time to react - if for any reason the airplane is not airborne by, say, 8°, rotation is slowed down or even stopped until liftoff. You might see another 1 or 2°(overshoot), but at this time the struts are extending anyway, so no probem there, although from outside it does look close.

Now, if due to misloading and/or mistrimming the nose starts jumping up unexpectedly a couple of seconds before “scheduled” rotation with a rate twice or thrice as high, you can easily figure the outcome. And if you’re unlucky enough and have selected full thrust/bleeds off (even more pitch-up moment), it may be too late by the time you react.

[mutt]

@mutt: Would you happen to have any official documentation for the B737-800. E.g. certification or AFM cg-limitations?

Not sure if you can consider Boeing weight and balance course notes to be "official documentation"....... but I've sent them to you anyway.

Mutt.

[Genghis the Engineer]

Dolly makes a very clear case why overcontrol is a real risk, even without going out of aft !

G

[Basil]

Ref fore & aft trim, I used to experiment with the limits (whilst remaining within them) on the Aztec. As previous contributers have so knowledgeably explained in erudite detail, aft trim made it twitch in pitch and fwd trim made it easy to fly but slow to flare. So mid is best unless you're on an IR when a bit fwd may be a good idea :cool:

Oh!, and full aft may help if you're practising to fly PARC's 737 sim ;)

[GlueBall]

The tried and tested technique for recovery of severe over-rotation, whether it's because of aft CG or whether it's because of incorrect stab trim, is to immediately roll into a 30 to 45 degree bank while applying full forward elevator and full forward stab trim. Maintain configuration and throttle setting and accelerate in the turn.

[john_tullamarine]

... methinks that procedure, if adopted during the rotation flare, might be a little stressing ? ...

I guess, though, if the beast has got itself totally outside anything approximating a sensible attitude, anything is better than just waiting for the inevitable ?