Airasia Plane Missing?

[Mitch Cronin]

Wow... shows to go ya.... rumour reigns supreme in the absence of fact.

[Don Hudson]

Hi Mitch;

Re, "envelope protection" as computed by the FACs - I'll dig into an A320 AMM in the morrow for possible details, but the protections aren't limitations on the flight controls as "rudder limiters" or "aileron lockouts" are. The FACs compute stall, minimum, manoeuvering & maximum speeds for the aircraft in Normal Law and send the data to the flight control computers. I understand what you're saying in terms of a sudden alteration of flight control position due to a change in incoming data to the box but that's not how the FACs work.

The release by the NTSC of what is known so far would go a long way to stop these kinds of stories from becoming "instant realities" by major news organizations wanting to get their story out first. All they do is end up embarrassing themselves and angering those directly involved in this tragedy. No wonder so many do not trust media any more - they have proven themselves again unworthy of serious reporting and broadcasting.

This was the story behind the headline I posted yesterday. True? partly true? Who knows? I happen to think this article has an element of truth in it...something was "lost in translation" and the world's media ran with rumours, and the NTSC and the Indonesians are doing nothing to sort matters out and help understanding. The process is a gong-show, a shameful treatment of those directly affected by this accident; - the inability to get the message out to the world is an embarrassment to the process of accident investigation.

QZ8501: Circuit breakers and pilot seats – Turning snowballs into avalanches

February 2, 2015

gerrys

When the National Transportation Safety Committee revealed the factual information on the QZ8501 accident, their 18 point press release and the questions and answers made on 29 January, revealed nothing that was previously unexpected. However, not long after that, Bloomberg and later Reuters, began running a story that the Captain was not in his seat at some point in the accident.

The Snowball

The Flight Augmentation Computer provides electronic control for the A320 rudder. (source: Airbus)

The revelation comes from unnamed sources “close to the investigation”, stating that the crew were switching off the aircraft’s Flight Augmentation Computers (FAC). Initially, most of the media wrote that “crew switched off critical computers before crash.” Well, the FAC is critical but switching them off, shouldn’t cause the airplane to crash. Some media did attempt to make the aircraft’s reconfiguration to Alternate Law for the flight control system as making the airplane hard to control, leading to the crash. Flying in Alternate Law is no big deal, although you must be aware that the aircraft no longer protects you from stalling, overspeeding, or rolling over inverted.

The media spin, has turned this item into a snowball.

The Avalanche

On Friday, Bloomberg reported that “one person/people familiar with the investigation” said that the after trying to reset the device the captain pulled the circuit breaker on the device (FACs). Reuters seemed to have made the jump from “switching the device off” to “pulling the circuit breakers.” This was the start of the avalanche.

• The circuit breaker for the FAC, cannot be reached easily from the seat, so someone had to leave or stretch out quite far from his seat.
• There is no Airbus procedure that asks for the crew to pull the circuit breakers to switch off, or reset the Flight Augmentation Computers on the A320.

Having a 2 reputable global media on this, obviously led to a global newsfeed frenzy pumping news of “captain left his seat” to all corners of the world.

My own little investigation

The skeptical me thought something does not seem right. Pilots do not make up procedures in troubleshooting. If global media can use unnamed sources, then why can’t I call my unnamed source. The difference being that my source isn’t someone familiar with the investigation, he is in the investigation team. So I asked.

The answer from my source, was that Bloomberg and Reuters may have been misled or got lost in the translations. The investigating team have not made such a conclusion or have not found evidence of the captain leaving his seat, presumably based on the conversations heard in the CVR. My source said that both crew were troubleshooting (with one flying and one troubleshooting), that FAC fault (unclear single or dual) was mentioned, and that the crew did not make up troubleshooting procedures as they went along.

As previously it was reported that the stall warnings did go off, it can be assumed that the aircraft was not flying in Normal Law (where the airplane is protected from stalls, overspeeding, or rolling over) but in Alternate Law instead. Combine this with previous news that the FACs were both switched off, let’s have a quick look at the DUAL FAC Fault procedure:

Enough talk of the circuit breakers. Let’s see if turning off the FACs require pilots to leave their seat. Here is the A320 cockpit and where the FAC switches are on the overhead panel:

As you can see, it is reachable from the pilot seats. However, the captain may need to stretch or loosen his seat harness to reach the FAC2 switch, and the same for the first officer to reach the FAC1 switch. This, is very different from “leaving his seat.”

My Conclusions on the Avalanche

How did the story went from switching off the FAC to pulling the circuit breaker, and from stretching to reach a switch to become the captain leaving his seat?

My simple answer: it got lost in the translation!

Simple confusion from a writer not realizing that that resetting the FACs and switching it off involve the same push-button switch could lead to the belief that switching it off means pulling the circuit breaker.

The need to stretch to reach a switch, given language barriers (eg: if source and the media speak in different languages), can result in difficulties of the context of the captain’s movement in his seat. Combined with the confusion over switch vs circuit breaker, put the two together, and there you have it, “he must have left his seat to reach the circuit breaker.”

Mainstream journalists are mostly not aviation journalists, and often even their news bureau rely on aviation experts to convert the technical language into layman’s terms. This adds another translation barrier for the context to be conveyed correctly.

On sunday morning, one of the local media, Tempo, put out a piece stating that the NTSC denies the stories that the captain left his seat (link in Indonesian only). The Investigator In Charge was quoted, saying “Fiction! There was no such thing!” Unfortunately the article is titled “NTSC denies that Captain left the cockpit”, but luckily I know the author so I asked him if he meant “left the cockpit” or “left his seat”. The answer is: BOTH!

Let’s see if Bloomberg and Reuters will update their stories quoting the NTSC’s refuting that the captain left his seat.

UPDATE:
Late on Monday 2 February, Reuters has put out a story that the investigators say no evidence that the captain left his seat.

[boestar]

Don:

In the description above it does in fact state that the rudder limiter and Yaw Damper are controlled directly by the FAC. It also states, emphatically, RUD WITH CARE > 160kts.

So then if the PF was making a modest input to the rudder at cruise speed and the FACs were then both switched OFF (PB not CB). would it then not be possible for the Rudder to increase its deflection to the full commanded deflection (No limiter)? This would in turn induce a large yaw component and roll component in the direction of the deflection. This would be a bad thing on most aircraft unless you were trying to induce a snap roll.

Again I am just speculating based on what I know of the basic control systems. I do not have a strong knowledge of what LAW the aircraft would be in at this point or the protections that it would offer.

[Don Hudson]

Thanks for the correction boestar- yes indeed, the FACs control the rudder and aren't just "input" for the ELACs & SECs.

I was too closely focussed on the notion that a sudden change in input to the FACs would result in a sudden change in the control surface position; I didn't think that such a thing could occur, an impression that I still hold. But let me ask around. Some comments I've read have indicated that the aircraft in Roll Direct Law is sensitive enough such that, at cruise altitude, one could possibly upset it to the point of LOC. I dont' think so but after AF447s crew's performance I am somewhat open to more theories... The main reason I think this wouldn't occur is based upon the way the airplane is designed in the first place. There is only one case that I am aware of on record where a partial loss of control occurred on an Airbus aircraft due to the performance of the flight control computers and that was an A330 of QANTAS, (QF72), where incorrectly-filtered data-spikes caused an errant PRIM, (something like an ELAC & SEC combined) to nose-dive the aircraft. The reason, IIRC, is in the same category as the recent AD issued by EASA for the Airbus A320/A330/A340 series regarding frozen AoA probes possible causing a pitch-down (ostensibly to recover from what the ELACs "thought" was too-high an AoA) - under the conditions that is may occur, the failure is generally more gentle than a sudden hard-over.

You'll recall the performance of a B2 bomber at Guam after the flight control computers received incorrect information from a plugged pitot probe. The incidents aren't unknown.

Again, thanks for waking me up

[DEFCON]

I'm confused by all the media stories, which seem to be frequently followed by corrections and retractions. Do we know for certain which variety of 'computer' may have been of interest to the Captain?

In the Lufthansa case, the probes froze while the aircraft was flying at a normal for conditions AoA. The impaired system computed / interpreted the data and decided the AoA was too high. Hal forced a nose down attitude accordingly. Is there any reason the opposite scenario couldn't be possible?

[boestar]

Thanks Don. I am just trying to understand it all. I am always curious to see what happens when we depart from "Normal" in a process.

[Don Hudson]

You're confused?!....

Re, "Is there any reason the opposite scenario couldn't be possible?"

Anything is "possible" but not everything is "probable" so it makes a bit more sense to talk about probabilities.

The introduction to the recent AD issued by EASA states:

An occurrence was reported where an Airbus A321 aeroplane encountered a blockage of two Angle of Attack (AoA) probes during climb, leading to activation of the Alpha Protection (Alpha Prot) while the Mach number increased. The flightcrew managed to regain full control and the flight landed uneventfully.

When Alpha Prot is activated due to blocked AoA probes, the flight control laws order a continuous nose down pitch rate that, in a worst case scenario, cannot be stopped with backward sidestick inputs, even in the full backward position. If the Mach number increases during a nose down order, the AoA value of the Alpha Prot will continue to decrease. As a result, the flight control laws will continue to order a nose down pitch rate, even if the speed is above minimum selectable speed, known as VLS.

This condition, if not corrected, could result in loss of control of the aeroplane.

If the AoA probes freeze say, at 5deg, that AoA is valid during the climb. As TAS & Mach Number increase in the climb, the AoA reduces, until in cruise it is around 2.5deg. The stall AoA at cruise Mach Numbers is about 5, perhaps 6 degrees. When the AoA reaches the frozen AoA value, in Normal Law the ELACs will (most likely, relatively gently as the airplane approaches the triggering AoA) respond by pitching the airplane down to reduce AoA and keep the airplane out of Alpha-Floor protection. The triggering of Alpha-floor and application of TOGA thrust is controlled by the FACs. This will be in the flight data if this is what happened. The AD requires that the crew turn off two / three ADRs which takes the airplane into Alternate Law which deactivates Alpha-floor protections.

One possible explanation for the focus on the FACs is this problem, because the turning off (vice just resetting) the FACs also deactivates the Alpha-protections - the FACs can only be turned off using the CBs, and even then, only on the ground. Remember, this AD only came out in the first two weeks of December, (even though it was originally issued two years ago, for both the A320 & A330/A340) and the captain, reputed to be very sharp, may have known that it was the FACs that computed the envelope speeds and activated the Alpha-floor autoflight response. As can be understood from the AD, the probability of the opposite occurring, (pitch-up, instead of pitch-down), could be considered as being extremely low, and unlikely in this accident because the airplane was at cruise altitude and Mach number prior to the increase in indicated altitude.

This is all just tea leaves now...

boestar...re understanding, you and me both! It's been seven years since I flew anything and, as Rich observes, the memory ain't what it used t'be, gol'darnit.

I have no clue now what happened, but I don't like the scent of what I intuit happened - I was hoping this wasn't going to be even close to another AF447. And I dont' trust anything the Indonesians put out anymore, perhaps not even their NTSC. The NTSB still hasn't been formally included, which increases mistrust.

[boestar]

Im not entirely sure I would want to be in or near a storm with two inactive FACs and the associated increased workload that comes with it. I guess we can still ask the question "What is it doing now?" when we are humming along.

While you are curious from one point of view I am curious from the aircraft systems point of view. All those nicely lit button are connected to the magic. I am curious as to whether the magic has gone missing somewhere.

You do know all those boxes down below the floor are powered by smoke right? If you let the smoke out of the box they don't work anymore.

[Boney]

Notwithstanding this AD, when stuff happens, who is in control? The pilot or the computer?

I know, a silly question, but that last post of that AD vis~a~vis the AoA, sure make one wonder, what was airbus thinking.

My bird used the CC, so in mostly all cases, except the few issues of a rudder hard over a few years ago, the pilot can override the autopilot.

[boestar]

Its is not a CC or SS issue. A CC just as easily feeds a set of computers to fly the plane. the issue or non issue is PHILOSOPHY. Plane protects the Plane or pilots protect the pilots. One philosophy goes....If we make the computers override the pilot command and keep that command from taking the airframe outside of defined parameters then we can protect the airframe from undue stresses and make it lighter and more efficient. While the opposing philosophy says If I let the pilot do what he wants then when a situation get scary the pilot, if necessary, can overstress the airframe in order to save his and everyone behind him ass. Therefore the airframe needs to be stronger and thus heavier and slightly less efficient. (Known as the 150% rule) The plane will take stresses up to 150% of what we say it will as a margin for safety.

I like the second philosophy. Break the plane to save the people works for me. Preventing me from doing that gives me cause for concern.

Let the fire breathing begin

/duck and cover/

[Woody Pusher]

Don,

Boestar asked earlier about the affect the loss of two FAC's would have on the Rudder travel. As the FAC's provide the computations for the RTL, I would assume that at CRZ FL the RTL would be around 4 deg, and would remain that way.

Would slats/flaps or gear down free up the rudder travel? I haven't seen anything on this, and wondered if you know.

[Don Hudson]

Boney, boestar, I know a little about C* and FBW and that is always a dangerous thing. Better, perhaps some answers can be found in two papers, especially the one called "Fly-by-wire Augmented Manujal Control - Basic Design Considerations", and a Cranfield paper on C* (C-star) fbw laws (link to the Cranfield Page), called, "The Application of a C* Flight Control Law to Large Civil Transport Aircraft", (the paper itself). ;-)

Hi Woody - I'll check in the morrow!

[Boney]

Thanks Don

[boestar]

Thanks Don

[Don Hudson]

Hi Woody;

Rudder travel is indeed limited at cruise speeds, to 3.4° deflection.

If both FACs are lost, (I suspect "disengaged" using the PBs, not the CBs), the rudder travel limit is locked when/if the second FAC failure occurs.

When the slats are extended, (but I suspect, not the gear), the limitation is set at the low-speed, (maximum deflection authorized) setting.

[boestar]

Thanks Don:

That makes sense to me that it would revert to basic operation based on configuration. That's what I was looking for. I just wondered if the FAC was removed from the equation what would happen to the RTL. Failing SAFE makes sense.

[Don Hudson]

boestar - works the same way on the A330/A340 - I should've remembered this from AF447.

[Woody Pusher]

Hi Woody;

Rudder travel is indeed limited at cruise speeds, to 3.4° deflection.

If both FACs are lost, (I suspect "disengaged" using the PBs, not the CBs), the rudder travel limit is locked when/if the second FAC failure occurs.

When the slats are extended, (but I suspect, not the gear), the limitation is set at the low-speed, (maximum deflection authorized) setting.

Thanks Don.

Looks like the RTL curve would be much the same as that we are familiar with for the A330.

[Don Hudson]

Looks like the RTL curve would be much the same as that we are familiar with for the A330. - Yes, exactly.

[boestar]

Yup that blows my possible theory out the window....thankfully.

[DEFCON]

If the aircraft is forced to do a flapless / slatless landing and gear extension doesn't reset the rudder travel limit, are approach, landing & go-around speeds high enough for restricted rudder travel to provide the authority necessary to handle yaw / roll requirements?

[Don Hudson]

Understand the point...one would think that the extension of the gear would reset the RTL to maximum deflection authorized but I don't see it in the manual.

If it isn't so, (gear down, resets RTL), perhaps Airbus was thinking probabilities and the electrical & hydraulic redundancies in the design?

First, from the videos I've seen on Youtube showing certification tests landing the aircraft with a significant crab on, (Boeing 777), loss of rudder authority doesn't prevent an approach & landing even in a crosswind!

Second, "slatless and flapless" plus an RTL set at minimum authorized deflection of 3.4°, (which means a loss of slat-flap capability essentially at cruise speeds), would be a "10^-9 probability" or more. The 3.4° is worst-case; any loss of both slat & flap systems at lower speed is going to have more deflection authority.

Loss of the slats and flaps could mean loss of all three hydraulic systems, (blue system powers slats, yellow system powers flaps, green system powers both) in which case the outcome is all but assured.

I'm trying to recall other circumstances which would cause the loss of both systems but for independent reasons, say, loss of the green system plus slat and flap assymetries which would trigger both the slat and flap wing-tip brakes, (which lock the slats or flaps when an assymetry is sensed), or two failed SFCCs...(Slat-Flap Control Computer).

I think we could agree that all of these circumstances would be in the 10^-9 probability realm.

Academically, I think a slatless/flapless approach and landing is doable, particularly if both engines are operating normally and if there's little crosswind... ;-) The aircraft would be in Direct Law and would be more sensitive but as AF447 showed one can get used to such sensitivity pretty quickly, (right after the loss of airspeed, Bonin got the initial PIO in Roll-Direct under control very quickly...but then pulled up...).

A go-around would be a very bad day indeed, but academically speaking I think it would be straightforward in the sense that the airplane is still flyable with the smallest hydraulic system, the blue one. But, worst-case, it would take deft, experienced handling and the discipline to maintain a high degree of focus for an extended period of time.

For reference...the A330's RTL scheduling schematic is something like:

[DEFCON]

Thanks Don.

Just left wondering who might have calculated the 1B to one odds?

[Mitch Cronin]

If you really want to drive yourself crazy, try calculating the odds that you actually survived all the obstacles to your potential existence and are really here to contemplate that fact!

[DEFCON]

http://www.ibtimes.co.uk/airasia-flight-qz8501-bleached-bones-found-recovered-fuselage-wreckage-1490088