A320 the big flaw of the UNRELIABLE SPEED/ADR CHECK procedure
 

After studying the issue for years (not only the procedure but also the FCTM, other airbus publications, powerpoint presentations) and over two hours of LPC simulator dedicated to this issue, I have come to the conclusion that there is a big flaw on it:

How can we tell if the altitude information is affected or not???

The procedure will soon lead you to this. Inmediately if "safe conduct of the flight" is not affected or after the memory items if it is:

LEVEL OFF for trouble shooting

HOW ON EARTH CAN I DO THAT IF I DON'T RELY ON THE DAMNED ALTIMETEEEEEERS???

I don't have GPS in my fleet, right now. Nor I have that new gadget, the BUSS, which looks like a very good thing and apparently is fitted in the 380s.

Is there any method to determine if the static part of the air data is OK or not?

I mean: are there different modes of static air data failure? Or there is only one mode, where V/S remains zero no matter how fast you climb or descend and then is totally obvious that static data is un reliable?

By the way, I never have this problem in the sim because it cannot simulate problems in the static. only in the pitot.

You assume a pitch and power setting which should result in level flight, and you look out of the window. :\

I believe most a/c with IRUs use them for V/S instead of static.

Interestingly, a pneumatic IVSI, Instantaneous Vertical Speed Indicator, contains a small pendulum that starts tne needle in the correct direction to counter the delay inherent in the basic VSI.

GB

Most ADRs use a combination of baro and IRS information to resolve VS. And a VSI with the pendulum is called an ILVSI. IL standing for Inertial Lead.

My company's QRH doesn't have any procedure for Unreliable Airspeed in cruise, only during initial climb. Seems like another hole in the Swiss cheese.

scary comment above!
We have had some quite intensive training in this stuff over the last year, apart from regarding the basic pitch/power settings as memory items, we are advised to crosscheck alt with the GPS output.

Thanks for the correction, Dan. I went back and completed and clarified my post.

GB

the other flaw is

what the heck is "if safe conduct of the flight is affected"

In my opinion, every time you have an unreliable speed safety is inherently affected. For all you know, you could be on the verge of stall, or overspeed, or a spiral dive or whatever.

If you have unreliable speed at 390, in turbulence at night with hail hitting the windshield, your body giving you all kind of misleading inputs to your brain... Is the safe conduct of the flight affected? I would say "yes it is" so i would have to take over manually and set pitch at 5º and CLB thrust? I don't think so...

They have to enhance the procedure (maybe including more "emergency" pitch-thrust settings: for high level climb, cruise or descend. These are on the tables but not on the memory items.

They can also develop a tool to give inertial V/S in the MCDU (altitude is secondary. V/S is what we need)

Then we can always level off.

One last thought:
What if the Air France 330 was in overspeed, or even well beyond it and when AP disconnected they set 5º up all of a sudden? let's say a 4º increment in pitch at that level and that speed. This could explain the pitch up input.

Inertial VS doesn't work without barometric inputs. See:

http://www.pprune.org/tech-log/452973-fpv-fpv-cage.html ;)

From the 744 ground school notes.

If there are accelerometers and valid aTTitude, inertial V/S is easy to obtain.

Or they can install a dedicated accelerometer. There must be a thousand ways of achieving a reliable instrument to achieve zero vertical speed.

Or else, enlighten us how to rule out (or not) failed altitude/V/S data

No, it isn't. See the thread discussion quoted above.

On a recent A320 sim-check, we had the following scenario :

In T/O just after rotation, we got 2 unreliable ADR`s, that both had the same failure of increasing airspeed, rapidly increasing to above 400 kts.

That means, that even though they are faulty, they wote out the remaining good ADR.

And what happens on the A320 in an overspeed situation : VMO/MMO + xx kts/0.xx mach, the A/C starts to pitch up, even though you manually try to pitch down !

And even more scary, with A/THR engaged, it reduce to idle due to higher speed than targetspeed. (We are not quite sure if this a "simulator thing" that it reduces to idle, can`t find anything about it in the books. Anyone confirm ??)

So with a rapidly increasing pitch and thrust at idle, things start to evolve very very quickly ! (and in this scenario very close to the ground). You have to do the following quickly, not a memory item (we are talking seconds) :

Solution : Select 2 ADR`s to off (doesn`t mean anything which two) and then you can apply the pitch/power procedure.

hadn't thought of that one!

regarding the thrust, if in SPEED, thrust will kept at idle as long as speed is much faster than target.

So in this case the high speed protection is trying to kill us. Taking over manually solves the A/THR comanding idle thrust problem, but what about the high speed protection!

how come thre is nothing in the procedure about this? what are we missing?

after rejecting one ADR we still have normal law, so it makes sense.

In Normal Law, Alpha ('Stall') protection "overrides" High Speed protection.

In Alternate Law, if High Speed protection is available (it is not always), it is "reduced", which means (IIRC) you can "overpower" it with e.g. full fwd stick.

Thus within the Flt Ctrl Laws, the High Speed protection should not be able to pull the nose up into/beyond the stall.

Of course, other factors e.g. THS position, CG, further malfunctions, Crew reaction time, may mean a stall can occur... all I am saying is the simple "aircraft uncontrollably pitched up into a stall and nothing anyone could do about it" line is not 100% valid ;)

NB I am talking A320 series here - A330 may be different (but doubt it fundamentally).

Why do people get wound up about unreliable airspeed in the cruise?

When you get to cruise make a note of the pitch attitude and N1 setting?

If you get Unreliable airspeed, you have the couple that was working fine to start with.

Or turn to 2.21 of the QRH, go to the bottom of the page and cross reference......

CLEAN - FL - SPEED - WEIGHT - PITCH/THRUST at your current weight.

Read across the pitch/thrust tables until yours match! Its all in front of you and always has been.

(While you're at it, note the LRC and green dot driftdown flight levels in case.)

Of course

Im sure that maintaining altitude with 2.5 or 3 degrees and about 85% N1 will give you about .76 at most of the typical flight levels and weights in cruise.

But they should include a pitch and thrust for cruise in the memory items.

And the level off critical question remains?

"Am I levelling off or is something in the air data system fooling me and leading me to disaster?"

I wonder if there is any means (other than GPS) to rule out a problem in altitude indication before leveling off for the trouble shooting

Hi Microburst2002,
I think this is because, pitots can clog with ice, nesting wasps, bird strike, or be unreliable with radome damage etc.
The only static ports problem I can think of is if they have been taped over by maintenance / engineering. Your outside check before flight should prevent the last case.

Otherwise if IMC, and no GPS data - I can't see how you would determine if you have levelled off either.

For me, the "level off" part means you are not climbing/descending at XXXX ft/min, which could affect your speed (a'la AF447's climb at 7000ft/min). Setting pitch at ~2.5 deg and thrust of 85% (or the pitch/thrust setting from the QRH), ensures that you are leveled off, or at least close enough for practical purposes and is more important than what the altimiter says. Even if you have one, you should hold the pitch, power, not the altitude...

Setting 2.5Deg pitch and 85% N1 close to coffin corner and slow on speed, hockey stick territory (which you wouldn't know if you didn't have the cue of yoke pressure) will not suffice in recovering the situation.

Therefore holding the pitch to recover the stall isn't always best advice especially at high alt with little margins

I see your point, Shaka

so there can't be a pitch/thrust setting for a high altitude scenario?
in that case, maybe the extra memory item in the procedure could give pitch and thrust to establish in a safe descend?

See what I mean? there should be more memory items for different situations (not only take off or go around); and
the issue of the static should be addressed, although what Rudderrat says makes sense.

Any other input about that (the unlikely probability of having erroneus static air data other than due to covered ports on the ground?)

"On a recent A320 sim-check, we had the following scenario : In T/O just after rotation, we got 2 unreliable ADR`s, that both had the same failure of increasing airspeed, rapidly increasing to above 400 kts. That means, that even though they are faulty, they wote out the remaining good ADR."

Not a pilot, but in another life I wrote medical/control software.
You MUST build in extensive sanity checks for rational behaviour and rational trends.

When in level flight/engines running normally, if the logic is suddenly informed of the sort of deceleration/speed loss more often associated with cumulo-granite encounters, it would be wise for the logic to be sceptical about the information before passing it on all loud and flashing to the flight-deck.

Eg:- Blood pressure readings from 3 independent sources

S1 shows a blood pressure and trend within the expected range
S2 shows an instantaneous BP increase to insane levels
S3 shows an instantaneous BP increase to insane levels
For S2 and S3 (irrational) to be able to outvote and drop S1 (rational) from the BP display is neither reasonable nor sensible.

As another example it is probably sensible to reject patient ages over 220 years, burns of 150% or 50kg newborn babies....

Much better to say sweetly;"Hi guys! I've just lost all my indicators and the AP has dropped out. 2 seconds ago I knew you were at M0.82, wings level, trimmed at 3deg NU and all was well, so don't do anthing hasty."

Mac

Set standard settings, as per QRH, or your brain, or as they were before the issue, afraid of coffin corner then descend. Whats the confusion, were pilots not just SOP book worms.

Pitch attitude and thrust, simple, you are trying to maintain control, not an altitude or speed...:ugh:

As Shaka pointed out, if you are deep in the sh*t when you realise that speed is unreliable it can be absolutely useless to set cruise pitch and speed.

If it is not, or there is a pitch and thrust that can establish the airplane in a safe flight path (be it level or not) then we should know those settilngs by heart, and they should be included in the memory items of the procedure.

I am a pilot and i know what I would do, but Shaka has given me food for thought. I mean: i am positively sure that i would have never reacted with a sidestick pull in an impending stall if it happened to me. Many other pilots have done so, which to me is absolutely outrageous. the concept of stall I have since the first paragraphs I read, the first flight lessons i had, is very simple and clear. Why on earth did they do that?

But if I set 2.5º up and 85%... most likely these settings are useless if I am already stalled. I can even be unaware that i am stalled. i don't know how stall buffet feels in the 320. maybe it is simlar to turbulence... there should be some tool in a pilot's toolbox for that contingency, and the procedure is not providing us with that tool.

hence this thread

Pretty difficult for memory settings, given the wide range of types / weights / engines one can fly.

However, at Crz levels, I would suggest the safest option might be Idle and 0 - 1nd, this seems to fit the drill power / pitch for descent, and is getting you into a safer zone i.e. down. Going up is just causing you issues with Min/Max speeds increasing.

I would still maintain the biggest issue with UAS is not "dealing" with it. It is recognising it, getting your colleague to agree, and then executing the drill. I believe most of the accidents have never really got to that stage?

NoD

MB2002
I think it's time somebody tried to answer your question! And perhaps a few others..

Without GPS you have to find a way of validating an altitude indication.
The a/c has three ADR's utilising two Static sources, and either an ISIS or Standby Altimeter which use a separate Standby Static source.
Valid altimeter cross-checks:

ADR 1 or 3 (Capt Static) with ADR 2 (F/O Static).
Any ADR with ISIS/Standby Altimeter.

Then you can decide which altitude readout to rely on. If only one of them works... well Lindbergh crossed the Atlantic with only one....

(The QRH also contains ADR 1+2+3 Fault... you turn off all three ADR's and rely on ISIS.)

It's a pilot judgement call ..In jaja's simulator scenario, and in the Type Rating UAS sim session the events occur on take-off so the memory items must be followed because safe flight is obviously affected. Established level flight above MSA would not require actions other than flying level and disengaging AP and A/THR before following the QRH procedure.

Memorise those tables? Give us a break.. you should just maintain the pitch and thrust values that you have been monitoring for xx minutes/hours while you
a) arrange a sensible level-off with ATC if required
b) get the thrust/pitch value from the QRH
Remember there is always the SEVERE TURBULENCE table in the QRH which gives a M.76/275/250 N1 for GWT and FL.

jaja
Seems logical to me - which is why A/THR ... OFF is the second memory item.

I think such a perfectly symmetrical scenario unlikely - but a double bird-strike might do it, and I once hit five lapwings on take-off...

Stuck in an ATR
Wrong - the UAS procedure requires you to fly level, and determine the a/c speed from the pitch attitude. If not the desired speed you adjust thrust until the pitch is correct, then set an estimated correct thrust. You may also find a reliable IAS indication.

Starting to think that maybe a new line should be added to the memory item:

Above FL300 - THR IDLE / PITCH - 2 degrees. Establish safe descent and level off for trouble shooting above MSA.

This incident that shows the greater of two evils at higher flight levels is indeed the stall. There are very few if any areas of the world where you couldn't descend for a good 5 minutes from a high level whilst gathering your thoughts and aiming to level off at say FL200 for trouble shooting :confused:

NoD hits the nail on the head. ----Recognition of the event and the time of recognition are paramount.

I've been in TAT probe icing on a 777 flying over Indonesia and neither of us knew what to look for and didn't call for the correct QRH drill.
Now, the aircraft didn't do anything peculiar therefore we never strayed away from the ''safe'' baseline.
But lessons learnt.

What it comes down to is:
1) knowing what your aircraft is doing at all times. 2) have a working knowledge of pitch/power settings during all phases of flight if things go tits up 3) be prepared to sacrifice on these rules if on recognition of problems you realize you are not in a position to set ''normal'' pitch power settings

Not a good idea in crowded IFR airspace. PITCH -2 is OK at light weight but risks an overspeed if you are heavy. You should descend out of RVSM airspace but in a careful organised manner having consulted the tables.

TPI suggested Idle, 0 to -1deg since they covered mid speeds for all weight for the A320.

I disagree with you re IFR / RVSM, since we are dealing with a real threat - we are now at 3 total losses from people unable to react sufficiently to UAS. The chance of a midair is tiny by comparison. A bit like an explosive decompression - you don't sit around at 38K' awaiting an ATC clearance to descend ;) But that is a point of view, not more or less valid than yours...

NoD

@Tyropicard

Just a small correction. ADR 3 static inputs are from STBY static probes. Same source for the ISIS but without ADR intervention. ;)

Aristoclis

Thanks Tyro and aristoclis

Still, if a level off is necessary to carry out the procedure there should be something in the procedure telling how to achieve that.

This is one thing I would include to improve the procedure.

Secondly, I would include something for the high cruise scenario, taking into account that by the time you realise that speed is unreliable you can have been already with weird pitch/thrust settings and in a stalled condition for al ong time, or well above overspeed, maybe suffering some nasty compressibility aeroelastic effects or who knows what... It is a very vulnerable situation, at high cruise. Just the same a take off is. So due consideration should be given to it.


NOd I agree that speed is unreliable is the most difficult part. In a simulator you are expecting things to go wrong. In real life, you do that, too, but in a different manner, and this failures are very very subtle...

and then you have to make the other pilot agree with you and start disconnecting AP/FD and A/THR and level off, which can be difficult if he sees nothing specially out of normal and he is the captain

Craggenmore
 

Amen!!!!

"Otto" may be in control, but you're still Pilot Flying... There is a hint there somewhere...

ECAM Actions.

If you are in a stall, setting that pitch and thrust might not be enough to get out of it.

It is very advisable to note pitch and thrust for every cruising level, so you can detect abnormal deviations from those values and help unreliable speed recognition.

but aside from that, we need a memory item "escape" pitch and thrust for high altitude provided by Airbus, such that we can confidently establish on a safe flight path and speed.

MB2002, I agree with you to some point that paper procedures do not extensively address Unreliable Airspeed at cruising level.

You keep mentioning Stall in your scenario. How/why a stall at cruising level? Unless someone is hiding behind a large newspaper, he will get cues that something is going on before...

Training lacks here: unless the speed/altitude (could be also TAT probes affected by ice particles impacting the A/THR...) discrepancy is obvious, we are very reluctant to disengage all automations.
There could be hundreds of different unreliable airspeed scenarii: if you're lucky, aircraft will go into ALTN law and disconnect everything for you...then if you apply pitch/thrust, you are SAFE (2.5° and 78% will work for any Airbus twin jet)

Past incidents have shown that those situations rarely exceed few minutes before everything goes back to normal...

If for any reason, you actually ended up in a stall with unreliable airspeeds...my question would be: how quickly can you determine what type of stall it is?...because correcting actions are not the same.

@aristoclis
Thanks for the correction - must have spotted the TAT line and ignored the rest! Every day's a school day...
TP

NoD
I agree that a descent is a good idea from high level cruise, and with your suggested pitch .. but it doesn't take long to make a Mayday and organise a descent.. and with a bit of luck everyone else has avoided the CB so you won't hit them...

shortfuel

that's the point

Is there a pitch/thurst setting which will do the job for any unsafe condition (stalled, impending stall, ovespeed...)? If it exists, they should tell us.

In the sim, I learnt how quickly you can be in a hazardous situation, even expecting something to happen. up there, with high and low limits so close, things can get hairy pretty quickly. I would be good to have a way out of any trouble, and it is not a problem if that "solution" implies initiating an immediate descent. Not more of a problem than an emergency descent, at least.

NoD
I would like to do some more research - assuming AF447 to be one of these which are the other two?
TP