Flight envelope protection event involving an Airbus A320

[Ichiban]

Full report at:

Investigation: AO-2014-049 - Flight envelope protection event involving an Airbus A320, VH-VQY, near Mildura Airport, Vic on 12 March 2014

On 12 March 2014, at about 0920 Eastern Daylight-savings Time, an Airbus A320 aircraft, registered VH-VQY, departed Melbourne, Victoria on a ferry flight to Darwin, Northern Territory, with a captain and first officer on board.
After about 5 minutes in the cruise at FL 360, the captain temporarily left the cockpit. When abeam Mildura, Victoria, the first officer received a clearance from air traffic control (ATC) to climb to FL 380.
Approaching FL 380, at about FL 373, the first officer observed the airspeed increase and the airspeed trend indicator approaching the maximum operating Mach number (MMO). He attempted to reduce the airspeed by selecting the speed back to M 0.76 however he observed the airspeed and the trend continue to increase. The first officer reduced the thrust to idle, which disconnected the autothrust, in an attempt to reduce the airspeed. He extended the speed brake and disconnected the autopilot to adjust the pitch attitude of the aircraft in an attempt to maintain the selected altitude.
When at about FL 383, the first officer re-engaged the autothrust, and returned the thrust levers to the climb detent. He applied forward pressure on the sidestick to lower the nose attitude of the aircraft in an attempt to recapture FL 380.
The aircraft then descended and the airspeed slowed below the VLS speed. The first officer then applied rearward pressure on the sidestick in an attempt to regain FL 380 and reduced the thrust levers towards idle but short of idle stop position. The application of back pressure increased the aircraft’s angle of attack. At the Alpha Protection speed, the Alpha Floor function activated.
This incident provides a reminder of flight crew of highly automated aircraft, to understand the implication of the intended and actual level of automation applied.

[outside limits]

Unbloodybelievable.
The result of this was that both pilots need to be at the controls for enroute level changes. This doesn't address the incompetence but merely confirms that only one pilot in the cockpit knows what he's doing.

[emeritus]

I find some of these incidents strange, speaking as an old pre glass/ fully automated pilot.

2+2 always equals 4 the same as cruise power for the applicable flt level always equals the applicable cruise speed.

If the ASI says otherwise then the ASI is lying. The only way I would be looking at idle thrust at high altitude is if I were wanting to descend.

All that is needed to fly at FL380 in fact is an altimeter, a thrust indicator and a manual of power settings.

Emeritus

[waren9]

emeritus makes a good point but these things arent usually as simple as they might be.

not defending the crew but

climbing in to a suddenly increasing headwind, some weight and not a lot between mmo and vls and perhaps an atmosphere different to isa all can conspire against a young airbus pilot

the report doesnt go that far, so who knows.

[Capn Bloggs]

Fascinating stuff.

I'm disappointed at the rabid condemnation that regularly occurs on Proone when one of these incidents occurs. I had a look a couple of days ago on the ATSB website and there are quite a number of incidents; good targets for sledgers but equally good learning material for those of us trying to better ourselves. So pull ya heads in and learn.

This incident provides a reminder of flight crew of highly automated aircraft, to understand the implication of the intended and actual level of automation applied.

And also a reminder that accurate handflying is still a critical skill and may be required at any instant.

What are the

He observed the TOGA LOCK display on the engine warning flight director

immediately scanned the primary flight director (PFD)

Nowhere near enough detail in the report about the FCOM procedures for overspeed nor when it was changed. As per bluesky, maybe it's a French thing that a procedure you may have to do instantly from memory is not a Memory item. I know it's cool to have only a few memory items but...

Would also have been nice to have the trace of the N1 for those of us who don't understand what the Climb Detent does at various times, and AP status trace.

I don't understand why the aeroplane didn't level at FL380 the first time with the AP still engaged?

Want more reading on High Altitude Overspeeds? There's a 7-page thread on it in Tech Log:

http://www.pprune.org/tech-log/53879...overspeed.html

Originally Posted by Jabba

Another PPR

I'll bite! What's a PPR (other than Prior Permission required)?

[clunckdriver]

As one of the early Bus pilots in North America , this merely demonstrates the flawed logic behind much of the Airbus design philosophy, but don't try telling this to Mr. Ziegler unless you want to be in court till Hell freezes over, he is always right no matter what, this in spite of the industry finally waking up to the importance of high standards of hand flying, be it via cables, hydraulics, or FBW.

[Ned Gerblansky]

Mis-handling or mis-understanding?

There is an old joke about the difference between an experienced Airbus pilot, and those of lesser experience and it goes like this:

An in-experienced Airbus pilot exclaims "What's it doing now?!!"
An experienced Airbus pilot calmly says "Oh, it's doing THAT again!"

IMHO the error here was that the F/O, assuming an autoflight malfunction, did not take full manual control, and got lost in the limbo that exists between partial autoflight and full control by the PF or the AP. "Mode mis-match" it was called in the early days.

An airmanship issue could be argued that, having experienced difficulty with what one thought the automation was doing, the climb could have been delayed and the aircraft returned to straight and level. My point is either YOU fly it or you let it fly you.

I remember once in the cruise at about F360, scanning around the displays, I happenned to catch both engines winding back (commanded by the autothrust) to idle. I immediately brought this to my F/O's attention and said words to the effect of "Hey, check this out!!"

We were in clear skies, no change of wind or temperature, no turbulence and both engines were running. I guarded the throttles, ready to push them forward of the climb detent, which overides autothrust. The airspeed decreased nearly to Vmin, when the engines roared back to climb power, stabilised @ the desired mach no and no further incident. It's an airbus - plastic heap of Euro****te.

Bottom line is for newbies on that thing-that-isn't-an aircraft, if it don't do what you want it to; 3 clicks! Auotpilot off, FD off, Autothrust off and then you're in a Cessna 150. Alpha prot and TOGA lock are the aeroplane trying to avoid a stall - and that can only be brought about by mis-handling, deliberate or not.

Hope this helps, cheers Ned.

[old rope]

The report doesn't mention when the PF actually retracted the speed brakes. Were they left extended until auto retract at Alpha Floor? Am aware of this scenario occurring previously with similar results.

Tough crowd this pprune thing, can't help but wonder how well some of the commentors would have handled this situation in the early days of their high performance jet careers.... FL380, a rather reduced flight envelope, alone on the flight deck with a PFD full of carnage. Not an easy place to be...

Leave the automatics in, let it overspeed, there will be considerably less paperwork.

[Capt Kremin]

I don't fly the 320 but am wondering if there is an issue with the autothrust response at light weights here.

I once did a ferry flight in a 330 where the autothrust was cycling between full thrust and idle as it attempted to hold the cruise speed at FL410. The aircraft was very light. Descending to 370 fixed the problem.

Perhaps there is a similar issue on the 320 as it seems like the autoflight system didn't cope very well in the initial parts of the incident.

[maggot]

Originally Posted by old rope

Leave the automatics in, let it overspeed, there will be considerably less paperwork.

yeah, sure - do what you can to prevent/avoid the situation but has an aircraft ever crashed from a minor speed excursion in the cruise? no need to rush into things.

general comment.

also, in climb the engines are targeting N1 - engaging a speed mode through 'alt' or whatever can reduce the thrust without mixing things up (as ned says above), add speedbrake if needed.

[Mobi LAME]

VH-VQY

You bloody Tech Crew! There was nothing wrong with the aircraft, other than a dunny flush CIDS problem, when we fly it out from the factory, it was about 30,000 flight hours ago though.

[LightItUp]

Cadet.....?

[TurningFinalRWY36]

Can you please explain how a newly hired direct entry would have done any better in this situation. I fail to see how burning around in a twin piston at 8000ft compares to dealing with a complicated jet aircraft at FL380 at M0.8

[Mud Skipper]

I fail to see how burning around in a twin piston at 8000ft compares to dealing with a complicated jet aircraft at FL380 at M0.8

And "therein lies the problem" as they say.

[Wally Mk2]

Always interesting reading when it comes to high Alt jet upset events.

'old rope' was correct to some degree, "leave the auto's etc' in" but having said that hind sight is always a wonderful thing but humans when presented with the abnormal/unusual/different often react or revert back to other ways of thinking despite current training etc,(Air-France A330, perfect Eg) IE reduce thrust, get some spoiler action happening (phew saved the overspeed)& then focus on not busting any ATC laws (going thru an Alt) this compounds the whole situation with all the time in the back of ones mind.....sh1t am in trouble now with the CP, you know 'just culture' & all that crap!!!.
The 'Bus' loves to hang the driver/s & it pays never to take yr eyes off the speedo for too long, in fact in any A/C obviously but with the 'Bus' the very soft auto-flt logic means it feels likes it's way behind the 8 ball just to keep the boguns comfy!:-)

Remember guys these fancy machines are designed by humans, fixed by humans & flown by humans & they are sadly still flying into the side of mountains in perfect mechanical condition so that means there is always gunna be cock-ups from the stick in the sand design stage all the way up to test pilots, how else do we find out errors whether that be human induced or design induced & learn from them?:-)

We ALL learn, this event is just another class-room day.

Wmk2

[Keg]

can't help but wonder how well some of the commentors would have handled this situation in the early days of their high performance jet careers.

In the 'early days' of my jet career I wouldn't have dreamed of climbing without the other pilot on the flight deck....... but perhaps that's just me. I've done it just once in nearly 20 years in QF. I was in command two-ish years so perhaps 16+ years experience. Never again after that time. Realised after the first little bit of the climb that I was widening the risk margins significantly for nil significant benefit to me or others.

[maggotdriver]

Winter time and jet streams. SA is the biggie here, we have all had issues at times with Boeings and Airbus. To all the cynics, I have thousands of hours in both and they're both crap at anticipating the large value shear. We however are aware of it and can look at a sigwx chart. Reduce the Mach prior to or at the beginning of the climb not after it's almost too late. It depends on the size of the buffer pior to reaching buffet. The 738 is crap! Not sure about the 320.

On a separate note two things. I wonder how long before the mods shut this thread like the one into the Goldie and secondly, I hear this is the fourth occasion in the last twelve months they have had a TOGA LK or alpha prot issue. Can anyone confirm this as this is alarming if true?

[Wally Mk2]

'Maggy' I doubt this thread will last long as it is about the J word which is associated with the Q word both a protected species in here at times
I hope it doesn't get slammed shut though as these discussions can be beneficial to some.

I agree with yr analogy about reducing the Mach, I use the 'Superman' philosophy (up up & away!) as I tend to pull off a couple of points of Mach to achieve just that (better margins) & to use the current energy to zoom up & get there faster, I mean after all if it's due turb then why hang around at Crz Clb?

Wmk2

[Ned Gerblansky]

I do hope this thread continues

I find the whole incident fascinating, and avoidable. One of the great assets of a forum is the ability to share information and suggestions to avoid a repeat scenario. A couple of points from my dim dark past:

Unless the crew is supplemented, at some stage in a long sector a pilot will leave to go to the toilet. The pilot remaining will then be single pilot, obviously. To those who say that 2 pilots should be on the flight deck for a level change, I question what happens when ATS say "Due crossing traffic climb NOW to F360"? Do you reply "Unable due captain having a leak", and then accept the 90 degree turn? What does the captain think when they feel the aircraft change heading in that scenario? Personally I would cease "splashing the boots" and get back into the cockpit and ask "What were you thinking?!!"

If an F/O is checked to line, then they should be able to handle any situation when the captain is not present, from TCAS to level 3 warnings - if not then they should not have been checked to line. From past experience the limitations of the Airbus autopilots are not widely taught, but the microchips and technology date back to 1980. In those days your computer would have been a Commodore 64, TRS-80, or if you were flash a Sinclair Z-80! These same chips power Airbus A320s today, and the autopilots are woeful. If ATS gives you a speed on descent of, say, 320 knots - the autopilot will handle that plus or minus 20 kts!! Reportable speed violations!

Next, an SOP from a long-defunct airline. When a pilot leaves the flight-deck, a flight attendant comes in, if only to open the door to the pilot outside in the event of incapacitation. The F/A cannot obviously occupy a control seat, so stands or jump-seats. It's amazing how more conservative the solo-pilot will be with another pair of eyes on them, qualified or not.

Lastly, was this incident caused by the F/O trying changing level using V/S? That only works on descent, in climb you stall, or get close. If you are new to the toaster (not-really-an-aeroplane-but-it-does-toast-pretty-well) use open modes for climb and descent. If in "coffin corner" select an airspeed between the zippers, give yourself a margin, and remember that one of those zippers will kill you -especially T-tails!!- the other will generate paperwork. As a lazy man I choose neither - be a lazy pilot and let the aircraft look after NOW so you can concentrate on NEXT.

Cheers,

Ned

[Cost Index]

Great post Ned. Agreed.

I hope admin keep this open. I find these discussions invaluable