A320 unreliable speed . . . only if you are a nerd like me

[oicur12.again]

Hi all,

An interesting knowledge gap came up during my sim program at my new employer.

I have seen unreliable speed simulated several times throughout the years on the 320/330 but the exercise has always been the result of ADR faults resulting in loss of speed info (red speed tape on PFD)requiring pitch and power to continue safely flying. Not difficult.

Yesterday I went in to the sim armed with what I thought was sufficient knowledge to cope with the unrel speed exercise but I was wrong.

The instructor simulated blocked pitot tubes at 330 knots climbing thru F150 that went undetected by the ELAC’s and FAC’s. 2 ADR’s gave subtly false IAS and only one gave correct.

Of course the incorrect (Cap) side increased into the red and because the fault was undetected the aircraft remained in normal law and gradually pitched up despite our efforts to lower the nose. At no time did it fault the ADR’s and trigger alt law.

I was under the impression that the speed discrepancy would be detected immediately and the ELACS would trigger the NAV ADR disagree followed by altn law as per the wording of the FCTM.

But in this situation the only exit strategy would have been to switch all 3 ADR’s off almost immediately, BEFORE consulting the QRH. The event unfolded very quickly and my efforts to do the right thing by the QRH got us into trouble.

Can this really happen where a speed difference of 30-40 knots can be undetected and put the crew into a situation where normal law refuses you the ability to manage the pitch attitude?

Regards

[Goldenrivett]

Can this really happen where a speed difference of 30-40 knots can be undetected and put the crew into a situation where normal law refuses you the ability to manage the pitch attitude?
Affirm. If two ADRs indicate the same speed (despite being in error) then system logic will accept them and reject the 3rd one.
If you have totally blocked pitots and are still climbing, then the indicated airspeed will increase (due lower static pressure) and if overspeed protection activates then the aircraft will pitch up uncontrollably in Normal Law.

Don't be afraid to put the aircraft into ALT LAW to regain control (unless you are like the crew of AF447 or QZ 8501).

[C_Star]

i'm not sure if your scenario is possible in real a/c. However, two things should be noted here:

- Hi AoA protection has priority over high speed protection, so you shouldn't stall as long as you stay in Normal Law.

- switching off 2 ADR's is sufficient to force Alternate Law. While it's is not the procedure for UAS (it is for the spurious AoA port), I think it's quite justified in your scenario- after all, you have to stabilize the path first, then troubleshoot...

[RUMBEAR]

Oicur12

Recently I had the same situation in a recurrent training sim. The only difference was the fault was set just after lift off. So after setting TOGA and 15 deg nose up, the airspeed rapidly indicated overspeed. ( and consequently triggered high speed protection). The situation unravelled very quickly. It was only in the debrief with the time to consider the scenario that I fully understood what happened. I think it surprised the instructor also. Must say I hope nobody has to face this in a real life situation.

[737Jock]

Had it in the sim as well after lift-off, simply switching 2 ADRs off forces alternate law. Completely lost the inexperienced FO though when I asked him to do this. So had to do it myself as pitch was going above 25 degrees with full downstick and 10000 ft/min climb. I realised the problem was the climb making the overspeed worse, but in normal law I could no longer level off. recovery in alternate law was easy.

This machine needs a big shiney red button to eliminate normal law.

Part of the problem was terrain that required a climb, but through the climb ADR 1 and 2 reached overspeed condition, which triggered the protection. this made us climb faster thus creating a bigger "overspeed" and thus a greater reaction from the protection. Making the situation worse.
Instructor was surprised I could not lower nose with full downstick in normal law.

We tried the same scenario without switching ADRs off just to see what could happen, outcome wasn't pretty. Would have hit the ground at 12000 ft per minute if it wasn't for sim freeze.

Anytime the aircraft normal law goes weird just kill 2 ADRs and fly pitch power. It's not just for spurious aoa vanes.

[oicur12.again]

Thanks for the feedback folks, yes nideed it was a learning experience, albeit a slightly embarrasing one.

"This machine needs a big shiney red button to eliminate normal law"

I agree.

And a big red button that turns off auto trim on final when its gusty!!!!!

cheers

[Willie Nelson]

A few people have argued for the 'big shiny red button' I think having three ADR buttons, neatly out of the way is just fine. I think we all realise that this is not quite the way Airbus intended things however should the above sort of situation start to unfold then the PIC or their delegate would be well within their rights to switch off two ADR's as per the OEB, just don't switch off the IRU's by mistake!

[ExV238]

The reason that increasing numbers of aircraft have flight envelope protections (starting with stick pushers, decades ago) is to protect against inadvertent approach to potentially catastrophic flight regimes. In general, pilots may encounter the flight envelope protections because:

1. They have lost awareness of the aircraft's attitude, flight path and/or energy state for human factors reasons or incorrect flight data, or

2. Upset due to an external influence.

In Case 1, de facto the pilot probably does not understand the situation. Providing a 'shiny red button' to immediately disable flight envelope protections may very well be the last thing that we should provide, because it is much more likely that those protections are working correctly than it is that they are the cause of the event.

The AF447 report is required reading to those of us flying advanced aircraft. But so is the report into the recent RAF A330 Voyager tanker upset, caused by a loose object pushing the stick full forward. In the latter case, the captain, at a loss to understand the behaviour of the aircraft, considered switching off the ADRs. However, the high speed protection was active at that moment and protecting the aircraft. Had the ADRs been switched off, or a 'shiny red button' been used, the aircraft would almost certainly have been lost due to grossly exceeding design speeds.

Be careful what you wish for, and remember that those devising emergency procedures and OEBs had much more time and information to come up with the best response to such events than a pilot would have in the unlikely event of such a malfunction.

[oicur12.again]

Exv238

yep, good points.

[Goldenrivett]

ExV238
Providing a 'shiny red button' to immediately disable flight envelope protections may very well be the last thing that we should provide, .....
Had the ADRs been switched off, or a 'shiny red button' been used, the aircraft would almost certainly have been lost due to grossly exceeding design speeds.
Suggest you stay well away from B777 & B787 aircraft because they have a "shiny red button" (called PFC disconnect switch on the overhead panel in the flight deck) and the aircraft design utilises envelope protection in all of its functionality rather than envelope limiting. (see para 11.3 for an explanation http://www.davi.ws/avionics/TheAvionicsHandbook_Cap_11.pdf

[Smilin_Ed]

ExV238 has just summed up what appears to me to be a major problem with today's pilots: They are entirely to content to let the airplane do what the computers tell it to do and they do not recognize when they are telling it to do the wrong thing. AF447 is probably the ultimate example of this.

[ExV238]

Goldenrivett,

Indeed, the PFC switch is in an equivalent position to the ADR buttons in the A320/330. However, the point is that the Boeing would not have been preventing an over speed in an upset such as the one I mentioned and might have exceeded VD regardless of the status of the flight control system. Therefore, the use and effect of operating that switch in the protection regime is very different. This is a philosophical difference between Airbus and Boeing that has been thrashed around for years on this forum and elsewhere.

However, the 'protection' and 'limitation' terms are misnomers. Both designs provide protection In different forms, but the Airbus does not limit the operational flight envelope (despite what some here might say!)

Whilst I'd love a go in the 777/787 for comparison (although I've flown the sim and tried the protections), I remain to be convinced that it's superior to my 'Bus!

[ExV238]

:) Ed,

Sorry, don't follow ya...

:sad:

[737Jock]

sorry but I think you misunderstand. Unreliable airspeed does not trigger the OEB actions.

But the protections will happily stall the aircraft into an unrecoverable state in case of an unreliable airspeed which senses an overspeed on 2 ADR's.
Unreliable airspeed and you think it's a brilliant idea to just shut down 2 ADR's completely at random losing all the airspeed data.

I'll have that shiny red button any time of the day.

That A330 captain could have turned off 2 ADR's, but he didn't. Why would he have pushed the shiny red button?
In any case it would be a guarded switch that requires confirmation of both pilots, versus one pilot just electing to switch off some ADR's........

[ExV238]

737,

If it's me you're responding to, then, with respect, I think you've misunderstood me! Perhaps I didn't make express myself very well. These are important points, so I'll answer them.

Note that I refer to emergency procedures AND OEBs; it may be thread drift (expansion?) but I was talking about more general flight control issues than just unreliable airspeed. My point is, indeed, that we need to be very careful about responding 'creatively' to an emergency by turning off ADRs/PFCs/stick pushers or whatever protection systems the aircraft has, in preference to following published procedures. There are echoes of the Staines Trident crash here; on that occasion the stick pusher was trying to protect the aircraft but was dumped, presumably because the crew's mental model of their predicament was wrong. Oh, and stick pushers were still relatively new technology and mistrusted by some.

My point re the A330 was that the pilot considered turning off the ADRs. On this occasion he didn't, but others in similar conditions might have done.

I know of no technical reason why a correctly loaded Airbus (or comparable Boeing, Embraer, etc etc) would enter an unrecoverable stall if the correct stall recovery actions are taken. Such things are tested in all flight control modes and wouldn't pass certification if this were so.

This is a really interesting subject, and there will always be differences of opinion on flight envelope protection. There are differences of approach across the manufacturers yet all of them are endorsed by the certification agencies, which shows that there's still a healthy opportunity for technology to develop and move forward.

[RHS]

Hi Guys,

Just so I can understand this a little better, I am reasonably new to type and this threw up some interesting scenarios.

So if my understanding is correct, after initially carrying out the memory drills "AP/FD Disconnect, A/THR disconnect, (in this case) CLB 5" because the pitots had blocked and therefore two ADR's were giving the same but incorrect information (aircraft overspeed) the Flight control protections kicked in and tried to force the nose of the aircraft upward as designed to exit the overspeed condition, but in this unique set of circumstances led toward a stall condition. The only way to get out of this was to switch off two ADRs thus forcing alternate law where manual control of pitch could be reestablished?

Thanks!

[Goldenrivett]

Hi RHS,
The only way to get out of this was to switch off two ADRs thus forcing alternate law where manual control of pitch could be reestablished?

Affirm. When we did it in the sim, we had TOGA Lock (due Alpha Floor activation) and overspeed "protection" fighting each other.

A "big red switch" would have been most welcome rather than the loss of a perfectly serviceable Altimeter Source (with 2 ADRs off) in order to regain control.

[CONF iture]

We tried the same scenario without switching ADRs off just to see what could happen, outcome wasn't pretty. Would have hit the ground at 12000 ft per minute if it wasn't for sim freeze.
Once activated, AoA protection should have taken priority ...

But so is the report into the recent RAF A330 Voyager tanker upset, caused by a loose object pushing the stick full forward. In the latter case, the captain, at a loss to understand the behaviour of the aircraft, considered switching off the ADRs. However, the high speed protection was active at that moment and protecting the aircraft. Had the ADRs been switched off, or a 'shiny red button' been used, the aircraft would almost certainly have been lost due to grossly exceeding design speeds.
This time the other guy had left the flight deck, but first step first, in all logic and in first priority he would have attempted to counter the sudden dive using his own sidestick, priority switch or not, kind of the way it happened for Lufthansa on takeoff.

Once a friend of mine took delivery in Toulouse and asked the question why didn't they install a red guarded switch to force direct law and kill any interference ?
The answer was : "Some pilots may use it"

[JosuaNkomo]

PUD SOD

Pitot blocked speed under reads in the descent. Thus over reads in climb. So the higher you climb the greater the over speed error until the Airbus normal law over speed corrections kick in and pitch up the aircraft until it stalls. Usual scenario in SIM is on take off.

I love the QRH statement "level off for troubleshooting".

In my mind if the protections stop you levelling off to troubleshoot turn 2 ADR,s off set a sensible pitch/pwr setting. CONF 1,2,3.... 7.5/57% but if CONF 1+ F pitch about 3.5. That's what I have in the back of my head.

Easier said than done as startle factor huge.

[CONF iture]

Pitot blocked speed under reads in the descent. Thus over reads in climb. So the higher you climb the greater the over speed error until the Airbus normal law over speed corrections kick in and pitch up the aircraft until it stalls.
Why should it stall ... Isn't it alpha protected as well ?

[tubby linton]

Any aircraft can stall but the bus tries very hard not to. If you are back somewhere beween alpha prot and alpha max and you encounter a windshear it could stall, that is why the fctm encourages you not to hang around in this corner of the envelope.

[JosuaNkomo]

Airplane thinks it is going much faster than it actually is. As the speed erroneously goes too high the overspeed protection activates.

[Gary Lager]

Josua - but see C-Star and CONF's posts earlier - AoA protection takes priority over overspeed protection and AoA prot is controlled by AoA vanes, not pitot static system, so it is unaffected by blockage. System should not let you stall (though TL is right in that this is not a nice part of the envelope to hang about in).

I love the QRH statement "level off for troubleshooting".
the scenario under discussion is why the full sentence from the QRH reads: when at, or above MSA or circuit altitude level off for troubleshooting i.e. DON'T DELAY troubleshooting by climbing higher than absolutely necessary.

[Gryphon]

The only way to get out of this was to switch off two ADRs thus forcing alternate law where manual control of pitch could be reestablished?


It is not the only way but the recommended one. Anytime in Alternate Law you have reduced protections and you are able to decrease the pitch attitude with the side-stick.
Example: Theoretically you can switch Off both FAC and you have Alternate Law but you cannot find it as a procedure.

[JosuaNkomo]

"Flight control protections kicked in and tried to force the nose of the aircraft upward as designed to exit the overspeed condition, but in this unique set of circumstances led toward a stall condition. The only way to get out of this was to switch off two ADRs thus forcing alternate law where manual control of pitch could be reestablished?"

Better summing up than I could do.

From Stall Recovery A320 QRH. " In case of lack of pitch down authority, reducing thrust may be necessary"

So are we going to be really concerned which protection takes priority in Airbus logic or will I just think " Fn"3k it this is not doing what I want it to do" 2 times ADR OFF!

As I said scary...

[CONF iture]

Any aircraft can stall but the bus tries very hard not to. If you are back somewhere beween alpha prot and alpha max and you encounter a windshear it could stall, that is why the fctm encourages you not to hang around in this corner of the envelope.
In the meantime it is the procedure to specifically apply in case of windhear.
"If FD bars are unavailable, use an initial pitch attitude up to 17.5 ° with full backstick, if necessary. If needed, to minimize the loss of height, increase this pitch attitude."
Airbus wants you to rely faithfully on their protections and their management of alpha.

[sonicbum]

i'm not sure if your scenario is possible in real a/c. However, two things should be noted here:

- Hi AoA protection has priority over high speed protection, so you shouldn't stall as long as you stay in Normal Law.

- switching off 2 ADR's is sufficient to force Alternate Law. While it's is not the procedure for UAS (it is for the spurious AoA port), I think it's quite justified in your scenario- after all, you have to stabilize the path first, then troubleshoot...

I also believe that it is extremely unlikely that 2 speed indications are erroneous by exactly the same magnitude leading to an undetected condition that unfolds as you have described. I am honestly not 100% sure on how many cases of unreliable speed indication have occurred in that same exact context, but I believe none.

[sonicbum]

p.s. sorry I have quoted the wrong post, meant to quote the OP :8

[tubby linton]

Airbus also state the following CONF-

The PF must not deliberately fly the aircraft in alpha protection, except for brief periods, when maximum maneuvering speed is required.
If alpha protection is inadvertently entered, the PF must exit it as quickly as possible, by easing the sidestick forward to reduce the angle-of-attack, while simultaneously adding power (if alpha floor has not yet been activated, or has been cancelled). If alpha floor has been triggered, it must be cancelled with the instinctive disconnect pushbutton (on either thrust lever), as soon as a safe speed is resumed.

[CONF iture]

The PF must not deliberately fly the aircraft in alpha protection
Don't go there just for the fun of it.

If alpha protection is inadvertently entered
If entered by accident, don't stay there.

But never you'll see Airbus doubting their capacity to keep you out of a stall by applying their full back stick procedure. When requested, they just want you to go there without thinking.

[tubby linton]

Go there without thinking but leave just as quickly!

[CONF iture]

Because it is not desirable to apply a procedure longer than necessary but not because "If you are back somewhere beween alpha prot and alpha max and you encounter a windshear it could stall"
You imply doubt where Airbus has shown none.

[vilas]

Alpha MAX is not a fixed quantity. It is being adjusted to changing parameters. So even if a stall occurs or is occurring because of quick changing environmental factors FCC is taking action to reduce the AOA. In other words recovery is already in progress, so you are not likely to exceed approach to stall phase which is much more benign than full stall that can occur in alternate and direct law. Imagine being in wind shear in a non protected aircraft where you are pushing and pulling against the stick shaker without any assistance from the aircraft and you can easily pull the aircraft in a full blown stall. Basically the protections are for inadvertent pilot excursions. Can there be any protection when going down the Niagara?

[vilas]

Valpha prot shown on the PFD is actually AOA displayed as speed. What is not understood by many is that it is a latching conditions. The aircraft will maintain that speed/AOA and will not exit the regime unless the stick is pushed forward. This was the cause of the air proximity incident between A340 and A330 across the Atlantic.

[Goldenrivett]

vilas,
Imagine being in wind shear in a non protected aircraft where you are pushing and pulling against the stick shaker without any assistance from the aircraft and you can easily pull the aircraft in a full blown stall.

Er... you would be pulling against the natural speed stability of the aircraft as it wanted to return to its trimmed AoA. Since the stab trim remains where you left it, you would relax the back pressure when the stall warning activated.

Pilots call it flying.

[Chris Scott]

Quote:
"... Pilots call it flying."

And I would describe that as arrogant...

[vilas]

Surely it is called flying and all those who do it are called pilots then why all these discussions? If stability was all that required then nobody would have thought of envelope protection Boeing included.

[vilas]

Golden
If you loose A/Skid it is not the end of the world if you use proper braking technique but would you rather not have A/Skid because you have the skill? Aren't lesser mortals better of in an aircraft that tries to unstall itself? We discuss events where stable attitude was badly pitched up without any environmental provocation and then somebody comes around banging his chest about piloting skills in a wind shear. I won't bet on it.

[Goldenrivett]

vilas,

My comment to you about your statement Imagine being in wind shear in a non protected aircraft where you are pushing and pulling against the stick shaker without any assistance from the aircraft and you can easily pull the aircraft in a full blown stall was to reassure all those pilots (and passengers) on conventional Boeings that a "full blown stall" is not inevitable. The aircraft will assist you with it's natural speed stability, provided the stab trim stays where you left it. You seem to suggest that unprotected aircraft are more at risk.

We discuss events where stable attitude was badly pitched up without any environmental provocation and then somebody comes around banging his chest about piloting skills in a wind shear.
Crew responses to Alternate Law in AF447 and QZ8501 seem to suggest crews find the aircraft particularly difficult to control (due FCC & auto stab trim?) or they are out of practice at piloting skills or both.

See Aerodynamic Principles of Large-Airplane Upsets (http://www.boeing.com/commercial/aeromagazine/aero_03/textonly/fo01txt.html)
"An important concept for pilots to understand is that if the airplane is at a balanced, "in-trim" angle of attack in flight, it will generally seek to return to the trimmed angle of attack if upset by external forces or momentary pilot input. This is due to the longitudinal stability designed into that airplane."

Better tell Boeing, Lockheed & Douglas Flight training departments to stop banging their chests and that their comments are "arrogant"

[vilas]

Golden
I am not denying anything you are saying. I am comparing the two concepts. And Boeing also has switched to protected airplanes because it is definitely a help. Life of 737 you may be aware was prolonged to MAX only because of Southwest airlines ultimatum to switch to AB unless Boeing immediately provided an answer to neo.

[oicur12.again]

Re switching off 2 ADR's to get into alternate law.

In my recent sim PC we looked at the undetected unreliable speed scenario I mentioned in the opening of this thread as well as the abnormal v alpha prot OEB.

Both happened quickly and both took me by surprise. Both required the switching off of 2 ADR' very quickly in order to make the situation manageable.

As mentioned above by another poster, a blocked pitot situation during climb resulting in a false increase in IAS NOT detected can happen though is VERY unlikely. I dont know of any such event. The abnormal v alpha prot issue has happened though my understanding it has only occured once?

Does anybody know of any undetected unreliable speed events in the Airbus or know of how many times an abnormal v alpha prot event has occured?