AF 447 Search to resume

[Machinbird]

As long as we are talking about subsonic flight, for a particular wing configuration, the stall is always at the same AOA, even during accelerated flight (pulling g) which is what makes AOA so valuable. Drop flaps or slats, contaminate the wing with ice, and you have a different configuration. If you are flopping around in a post stall gyration, AOA tells you which way you have to go to get the wing flying again. If there is ever a doubt about the validity of airspeed, AOA will let you know when you are getting near the wing's limits. There is a learning curve to use it, but it isn't very steep.
It boggles my mind to think that you have the probes mounted on the aircraft, and glass cockpits, and yet no one wants to do the little bit of software work to put it on the PFD.
Machinbird

[Chris Scott]

The "Bird" and AoA

I agree with Machinbird. It used to be possible to see AoA by consulting AIDS via the MCDU2 on the A320, but - whether or not it still is - that's hardly practicable during a jet upset.

As CONF iture knows well, a good ball-park indication, without any actual numbers, is on the PFD in FPA mode (preferably with the FD switched off). The Pitch attitude angle minus (algebraically) the flight-path angle approximates the AoA. For non-Airbus pilots, the latter is indicated by the "bird" symbol. The snag, for this purpose, is that it also shows any drift in azimuth. So, when there is a lot of drift, it is not directly under the little black box* which represents the nose of the aeroplane.


* [black, but yellow inside]

[HazelNuts39]

RE: CONF iture #514
Thanks CONF iture, my estimates where 186 t; ISA+10C
regards,
HN39

[HazelNuts39]

AoA on the PFD

Just a personal thought on this issue. The average airline pilot will never be in a situation where he has any use for AOA. In recent posts one pilot puts the stall at 12 deg, another at 18 or even more. If a pilot finds himself suddenly in a complex situation where AoA could be really useful, how quickly would his mental processes adapt to using an indication that he has never used before?
regards,
HN39

[Backoffice]

Off topic for aviators...but I'm sure someone will know

Can I just pose a new challenging question (while we wait).:-

As the 2nd or is it 3rd stage of the search is about to begin. Many posters here have postulated the theory that the aircraft may have broken into fairly sizeable chunks before sinking to the bottom.

The perfect solution of course for flight recorder recovery, would have been for it to be torn apart, that way, somewhere on the seabed would be the recorders laying to be discovered, as seems to have been the case in recent accidents.

However, we may well have a near intact rear section of fuselage, probably somewhat compacted due to the pressure depth in which these vital recorders are bolted.

Is it at all likely that an ROV could cut and recover the recorders in this situation or would the whole piece of wreckage have to be recovered to the surface.

This rear section may well have a mass of 10 tonnes and a surface area of 30 sq.m. +
Could it be lifted, how and, what is the physics involved.

Bit of surfing has revealed they lifted a 17 ton piece of the Titanic, however they dropped it 2.5 miles back to the bottom the first time due to a storm upsetting the cables.

[PJ2]

Originally Posted by bearfoil

Asked to give an approximation of Vs, his post was, with all due respect, "Duh". An experienced Airbus 330 jock, PJ2 doesn't do the a/c's reputation any good, notwithstanding his patience and consummate skill.

Well, I tried, bear, where others didn't. There is a great deal in knowing and flying the airplane which cannot be conveyed easily within this kind of discourse. If you didn't pick up the sense for example that the Captain/crew would know that the airplane was in serious diffculty long before indications of an actual stall then perhaps it is my writing that has failed you and, as you claim, "the readership"; if I have contributed to a lesser understanding of the A330 when precisely the opposite was my goal, I sincerely regret that.

You are correct in the exercise of patience because that and a certain forbearance on the part of the readership because print is difficult to convey some notions in, is what it takes to understand the airplane. One cannot convey a sense of the airplane in a few paragraphs so it becomes all to easy to dismiss the airplane out of hand without ever setting foot in it.

To those who fly it, including, clearly, the thirteen crews who successfully dealt with a momentary loss of airspeed data, the airplane is not nearly as antagonistic or obscure as defined in your response.

Anyway, I guess with explanations garnering a "duh" I'm done here and someone else who is willing to spend the time and who has a better comprehension of both engineering aerodynamics and flies the airplane can sit in. Good day.

[Mr Optimistic]

Bearfoil

a tad graceless don't you think ?

[bearfoil]

Utterly graceless, and I see my failure of communication clearly. The word has taken on a meaning that is perhaps not as disrespectful as some would think it. This may sound like an excuse, and disrespect should deserve the outrage it provokes. I accept it, and offer everyone an apology, a sincere one.

bear

[infrequentflyer789]

Originally Posted by bearfoil

A pilot would seldom need an AH, or BUSS, or AoA 'alert' (other than Stall); except there are those 13 uA/S incidents in which they would have been welcomed, surprise or not.

BEA 2nd report lists almost 40 uA/S (due to pitot) incidents on 330/340. All (apart from 447) involved loss of automatics and reversion to alternate law... and not crashing.

Across this thread there have been pilots stating that uA/S / pitot failure is a "non-event", even on this type. Equally importantly, there are a few pilots who have found it not to be - they aren't here to post. This isn't a type (or manufacturer) specifc issue - a number of the latter group were flying boeings.

One is satisfied with a one hundred year old instrument sensor, (and why not be),

Acutally I am not (satisfied). It looks like a weak point in current aircraft design and deserves further investigation to see if we can improve on something we've taken for granted for so long. Sticking multiple "redundant" probes (of the same type) out into the same environment provides no redundancy to environmental issues such as icing. The B2 allegedly has 20+ sensors...

All that said, a common factor in the pitot incidents that have gone bad is night-over-water - so maybe the right thing to be focussing on is jet upset and spatial disorientation in general, rather than just one instrument that can lead to it.

This type appears to have granted its auto pilot a rather long leash,

Again, this isn't type specific. In other recent incidents the automatics (and this includes the auto-trim discussion) have flown planes to the stall and the parts spread across the landscape have been B, not A.

There is a good general discussion to be had (fragmented across a few threads at present - and maybe deserves its own) on this issue. I do have a feeling that as the industry has given airlines and pilots better and better automatics the, unintended, consequence is that pilots are being barred, by airlines, from flying the plane (SOPs and "safety"). I don't think it can possibly enhance safety if the only time the pilot is handed the controls is when the aircraft is heading out of control - on any type.

a host of surprise degradations to handling when its auto flight quits, and a reversion to 'Direct Law', as though that's the ticket to save one's bacon.

A look through the other uA/S incidents shows AP/AThr disconnect and reversion to alternate law. This isn't (or shouldn't be) a "surprise" degradation, or a difficult one to handle, as is evidenced by all the previous incidents (in fact the only thing that "degrades" are the flight control protections which some airbus critics seem to think shouldn't be there anyway).

Reversion to direct law is clearly not a direct consequence of uA/S (see the other incidents). Something else happened (uA/S may have started the incident). Someone (I think it was the pilot union head quoted in Spiegel article) has stated that the pilots initiated the PRIM reset - although how they know this isn't made clear. Obviously it didn't save their bacon whether it was pilot or system initiated - but my hunch is that they were already irrecoverable at that point.

[PJ2]

Bearfoil;

Thanks.

With regard to what happened to AF447, whether the pitot/ADR/PRIM issues were coincidental with an unrelated series of handling issues possibly (or not) related to weather, or were the cause of the LOC, there remains much to discuss.

PJ2

[HazelNuts39]

CAS and AoA

RE: CONF iture #521

Originally posted by PJ2: for any specific weight there is a direct, corresponding relationship between CAS and AoA regardless of altitude

IMHO that is not exactly true, but maybe good enough for practical purposes below Mmo
regards,
HN39

Example
An A330, 205t, cg as QF72, 235 kt CAS:

@ FL 350; M 0.7; AoA = 5.2 deg
FL 200; M 0.52; AoA = 6.1 deg
sealevel, M 0,36; AoA = 6.5 deg

This assumes that the lift coefficient/AoA relation derived from the QF72 maneuver corrected for Mach-effect per Prandtl-Glauert does not change with Reynolds.

[mm43]

BEA - Press Release

Flight AF 447 on 1st june 2009
A330-203, registered F-GZCP

Press release, 19 March 2010

The « Anne Candies » is now expected in the port at Recife on 24 March.

As announced on 15 March, the BEA will hold a press conference before the departure of the ships to the search zone. This will take place in the presence of the teams of investigators and those involved in the sea and undersea search operation on Thursday 25 March from 14 h to 16 h in Recife in the port administrative buildings in the port area (main port entrance).

Journalists who wish to attend are requested to confirm their presence with Martine Del Bono at the latest by 23 March.

Following this press conference, a media pool will be able to go on board Phoenix International’s « Anne Candies » and the « Seabed Worker » to film the equipment of the US Navy and Geomar, Seabed and Woods Hole Oceanographic Institution (WHOI).
------------------------

mm43

[broadreach]

Anne Candies

The ship was about 300 miles off Belem this afternoon, at just over nine knots. With about a thousand miles to go to Recife she should be there by Tuesday evening. Allow a few days for victualling, bunkering, spares to arrive etc and she may be off to the search by the end of next week.

[fdr]

AoA/Alt & D.P.Davies

Traditionally the industry has never used AoA because for any specific weight there is a direct, corresponding relationship between CAS and AoA regardless of altitude. "For every air speed - as indicated on the Air Speed Indicator - there is a corresponding angle of attack at which level flight can be maintained (provided the weight of the aeroplane does not change)" - Kermode, Mechanics of Flight, 1962.
That's the part in red I'm not sure about, but I don't have the knowledge so I'll keep it quiet, maybe someone would explain more ... ?
The best would be to flight test these lower speeds during a ferry at FL something, but I still need/like my job.

PJ2/CONF iture

within reason, (3 decimal places @ representative crz AoA, α ...) D.P.Davies was more or less correct in his statement.

Angle of attack is linearly correlated to section lift coefficient, Cl below separation angles of attack. The correlation of section lift coefficient, a, is dependent on Re, and also ~Mcrit.

Reynolds is dependent on density & viscosity...
Re=(ρ.V.L)/μ;
= (density x mean fluid velocity x characteristic linear dimension)/dynamic viscosity of the fluid.
Total lift is also dependent on density;
L=Cl.0.5ρ.V^2.S

Density is a numerator in both cases, so is not self cancelling with variations...

At low AoA, there is almost no difference in the Cl for Re from 1,000,000 to 8,000,000, but at high AoA, the higher Re results in higher Cl for a given AoA.

At high subsonic MNo, the compressibility effects alter drag (drag divergence) but also alter Cl, for a given AoA, but the outcome is dependent on geometry of the foil.

remaining at normal cruise AoA and below drag divergence, D.P.Davies statement is reasonable that altitude doesn't effect AoA for a given CAS (restated). Humidity does have an effect but is minimal for normal operational conditions.

.................

Ideal Gas Law:

ρ = p / (R * T)

where: ρ = density kg/m3
p = absolute pressure Pa, N/m2
R = individual gas constant J/kg K
T = absolute temperature K

D=((P/(Rd*T))*(1-0.378*Pv/P)

where: D = density, kg/m3
Pd = pressure of dry air (partial pressure), Pascals
Pv= pressure of water vapor (partial pressure), Pascals
P = Pd + Pv = total air pressure, Pascals ( multiply mb by 100 to get Pascals)
Rd = gas constant for dry air, J/(kg*degK) = 287.05 = R/Md
Rv = gas constant for water vapor, J/(kg*degK) = 461.495 = R/Mv
R = universal gas constant 8314.32 (in 1976 Standard Atmosphere)
Md = molecular weight of dry air 28.964 gm/mol
Mv = molecular weight of water vapor 18.016 gm/mol
T = temperature, deg K deg C + 273.15

Links:

http://www.efm.leeds.ac.uk/CIVE/CIVE3400/stvenant.pdf

Equations - Air Density and Density Altitude

luizmonteiro - Altimetry Calculations / E6B Emulator

JavaFoil

Dynamic, Absolute and Kinematic Viscosity



References:

Batchelor, G. (2000). Introduction to Fluid Mechanics

Clancy, L.J. (1975), Aerodynamics, Pitman Publishing Limited, London. ISBN 0 273 01120 0

Kundu, P.K., Cohen, I.M., & Hu, H.H. (2004), Fluid Mechanics, 3rd edition, Academic Press

Ockendon, H. & Ockendon J. R. (1995) Viscous Flow, Cambridge University Press. ISBN 0521458811

Reynolds, Osborne (1883). "An experimental investigation of the circumstances which determine whether the motion of water shall be direct or sinuous, and of the law of resistance in parallel channels". Philosophical Transactions of the Royal Society 174: 935–982

[HazelNuts39]

RE: fdr post 535

All very true and interesting, except 'three decimal places ...' - see my reedit of post 532
regards,
HN39

[PJ2]

Originally Posted by fdr

remaining at normal cruise AoA and below drag divergence, D.P.Davies statement is reasonable that altitude doesn't effect AoA for a given CAS (restated).

Which was the original point that I tried, although not well, to make. Hazelnuts39, I understand the point you're making; - while not entirely precise and knowing that supercritical wings do not stall uniformly and that atmospheric conditions will cause variations which must be taken into any precise account, nevertheless there is, for practical work using the kinds of displays available on the A330, B777 etc, essentially one CAS for a specific angle of attack. As Chris Scott pointed out earlier in the thread, the angle of attack is represented on the A330 PFD when using the FPV - Flight Path Vector. Hopefully the schematic below will assist in making this a bit more clear. I believe I have used the terminology for actual flight path, (gamma), pitch, (theta) and AoA, (alpha) appropriately - any corrections gratefully accepted. PJ2

[Machinbird]

From the July Interim BEA report:
"FLAG ON CAPT PFD FD and FLAG ON F/O PFD FD (2 h 10)
Symptoms: Disappearance of the Flight Director on the PFDs, Captain and
First Officer sides, and display of the red FD flag.
Meaning: This message indicates the Flight Director function is selected and
unavailable."
Followed by:
"FLAG ON CAPT PFD FPV and FLAG ON F/O PFD FPV (2 h 11)
Symptoms: Disappearance of the FPV (bird) on the PFDs, Captain and First
Officer sides, and display of the red FPV flag.
Meaning: This message indicates that the FPV function is selected and
unavailable."
We don't really know how ACARS prioritized the transmittal of messages, but they could even have been virtually coincident.
Even if you could use this method during routine flight to determine AOA, it wasn't available to AF447 when the going got interesting.
Machinbird

[PJ2]

Machinbird;

Even if you could use this method during routine flight to determine AOA, it wasn't available to AF447 when the going got interesting.

Yes, I think that would be correct. Other than possibly establishing a stronger situational awareness I would not expect an A330 pilot to use the 'bird' for primary purposes. The schematic is merely intended to describe the FPV presentation and how it works in response to questions on same, earlier in the thread and by PM. PJ2

[Machinbird]

PJ2
It was only my intention to point out that when you really might need such information as AOA, your customary sources of inferring it would not be available. Once the ADIRU outputs are disqualified, the velocity vector information they provide is removed and the aircraft systems have to make do with what is left (which isn't much).

Much earlier on the other main AF447 thread, there was a posting about an available Airbus system (BUSS-BackUp Speed Scale),(an available option in purchasing the aircraft) that inferred airspeed using AOA and a computer performance model of the aircraft (Post 4041). I'm not sure that even that approach is adequate. Suppose a hail encounter "modified the aircraft" a bit, e.g. removed the radome. Would your AOA sensors then be valid?

If AOA is ever presented in an airliner, I hope there is not too much computer mediation of the display. You don't want a damaged or defective probe or vane causing an accident ( e.g. A310 in Africa), so redundant sources are desirable, but computed AOA involving airspeed derived validation scares me. I've flown a lot with single source AOA indicators and have found them to be reliable and simple to interpret. You do have to pre-flight the probes/vanes but it is just a visual check. The info could be available on your PFD with minimal work, and a few flights/simulator sessions would give you the big picture of how it does its job.

Has anyone ever worked out how many damaged/destroyed airliners might have potentially been saved by providing the crew better AOA awareness? Is that stall warning after takeoff real or bogus? Might be an interesting study.
P.S. Didn't the L1011 have AOA indicators? How did that work out?
Machinbird

[PJ2]

Machinbird;

It was only my intention to point out that when you really might need such information as AOA, your customary sources of inferring it would not be available. Once the ADIRU outputs are disqualified, the velocity vector information they provide is removed and the aircraft systems have to make do with what is left (which isn't much).

Thank you for your reply. Understand your intention, but while other remarkable developments have occurred the use of AoA in airliners as a primary indication doesn't appear to be widespread perhaps because it just hasn't been a problem. Regarding a survey of accidents, off the top of my head I can recall six accidents in which a stall was involved in the accident sequence; Colgan Q400, Turkish B737, One-Two-Go MD82, Spanair MD82, Empire (FedEx)ATR42, USAirways B737 (Pittsburgh) and possibly AF447. This doesn't include several fatal Bombardier RJ stall accidents caused by contaminated wing surfaces. Of course, no AoA indication is going to assist the crew in the contaminated wing case.

I would like to know what I'm missing about AoA. In my opinion, and I hope HN39 and others will comment, on the surface and except for possibly one accident I don't see the direct availability of AoA to the crew making a difference in these accidents listed above even assuming, obviously, that crews would have been trained to use the indication.

The question and argument here I think is, what does AoA offer that CAS does not. There is the obvious case where 'g'-loading is involved but in considering the accidents described herein the low altitude (around 400') of the Turkish B737 when the secondary stall occurred (I believe due to pitch up from increasing engine thrust) would have made any successful recovery doubtful. It might have made a difference in the USAirways B737 accident near Pittsburgh, providing sufficient information for the crew to "relax" on the elevator just enough to unstall the airplane. That may have been mentioned in the report, I don't know.

BTW the A330 has 3 AoA vanes - Capt., F/O, Stby, with inputs to the three ADIRUs. Not sure you were referring to the A330 present design or to the previous suggestion regarding "computed AOA involving airspeed derived validation scares me". I agree with you even for a backup system, but just to make clear that the A330 system is sensor-derived although the display of the 'bird' is of course DMC (Display Monitor Computer) driven.

PJ2