If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
More mostly irrelevant updates in AvHerlad. The latest one reinforces the scenario where everything was business a usual and then they were gone in a blink.
Whatever happened seems to have happened very very fast.
--- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
--- Defend what you say with arguments, not by imposing your credentials ---
It sounds like the autothrottle failed on one side, causing a gradual asymmetry and the crew failed to notice due to their concentration on weather and navigation. It then sounds like the autopilot added compensation until reaching the operational limit and disconnected, causing a sudden roll to the right which was not recovered by manual pilot inputs. The factors I suspect were stealth, startle and IMC spatial disorientation. The question that I have is whether the left or right thrust was the one malfunctioning. Since the initial event was a L rollback during climb, which was compensated by autopilot during a subsequent level-off, and then the a/p disconnect and more severe roll occurred after a readback to resume the climb, it would make more sense to me if the R thrust was functioning and the L thrust was not. The resumption of climb thrust would decrease, not increase the thrust assymmetry if the problem was with the R engine thrust being 'stuck', but intensify it if the problem was with the L engine being inoperably retarded.
Aviation Herald - News, Incidents and Accidents in Aviation
This is interesting. If a condition exists where a thrust unbalance can have the AP first trying to compensate for it with partial success and then give up and self disconnect causing a sudden and violent upset when the controls are neutralized (by the AP giving up), I would call that a latent unsafe condition.
This condition has caused a few accidents in the past, some related to icing where again the AP was compensating for asymmetric lift until it gave up, let go on the controls (including the compensation it was doing) and the airplane suddenly rolled inverted. In a stretch, we may even include AF 447 in a variant of this condition. While the upset was self-inflicted by the pilot due to startle, the sudden knowledge that the AP had just disconnected (with no previous warning) may very well have been a factor.
For a long time I had the opinion that, wen the AP is keeping a more-or-less stable situation (or at least helping it not to become more unstable), it should give a warning to the pilots and "invite" the pilots to grab the controls and disconnect the AP a few seconds before disconnecting itself.
So, for example:
Current case: Due to an increasing thrust asymmetry, the AP is using increasing aileron inputs to try to keep the target bank until it reaches the limit of control input it is designed to provide and self disconnect. The pilots receive an AP disconnect warning at the same time than a violent roll as the compensation that the AP was doing is suddenly removed with the AP self-disconnect.
Suggested case: Due to an increasing thrust asymmetry, the AP is using increasing aileron inputs to try to keep the target bank until it reaches the limit of control input it is designed to provide. The AP keeps that max input (even when it is insufficient to keep the plane from rolling, but rolling slowly) and give a warning to the pilot that it is about to self-disconnect in 10 seconds. The pilot grab the controls, notice that they yoke is partially turned say to the right, notes that nonetheless the plane is rolling to the left, grab the yoke firmly and disconnect the AP. The pilot is still surprised by the yoke trying to center itself and the plane trying to roll in the opposite direction but he already has his hands on the yoke and has a much clearer situational awareness and is able to compensate.
The above in no way removes the pilot's responsibility to monitor what the plane is doing and what the automation is doing. These pilots apparently missed, for about 1 minute and 30 seconds, and during a critical phase of flight, the split throttles and split engine indications and necessarily a reduction in airspeed, or climb rate, or both.
--- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
--- Defend what you say with arguments, not by imposing your credentials ---
On Feb 2nd 2021 the KNKT chairman said in a press conference, that reports distributed by western media about a possible autothrottle malfunction causing asymmetric thrust are wrong.
Well, guess what? They were not wrong. That did happen and that caused the chain of events that led to the loss of control and accident.
--- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
--- Defend what you say with arguments, not by imposing your credentials ---
For a long time I had the opinion that, wen the AP is keeping a more-or-less stable situation (or at least helping it not to become more unstable), it should give a warning to the pilots and "invite" the pilots to grab the controls and disconnect the AP a few seconds before disconnecting itself.
As have I. A master caution, a specific warning tone, an amber message like AP CTRL LIMIT or something, but no disconnect for 10 seconds. It's seems like a no-brainer. And pilots will insist that it adds unneeded complexity. But I would be very happy to see that done (don't expect it on legacy aircraft however).
The factors I suspect were stealth, startle and IMC spatial disorientation.
Late in my PRIVATE pilot training, we did recovery from unusual attitudes while on instruments.
(Training for an inadvertent IMC & Cb encounter.)
I know that the statistics say that (by some measures) that training is inadequate...
The point I am making, though is that recovery from unusual attitudes is trained for from time to time AND a current pilot is expected to be able to...recover.
(That, and monitoring basic stuff every FEW seconds (as opposed to 1.5 minutes).)
Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.
It sounds like the autothrottle failed on one side, causing a gradual asymmetry and the crew failed to notice due to their concentration on weather and navigation. It then sounds like the autopilot added compensation until reaching the operational limit and disconnected, causing a sudden roll to the right which was not recovered by manual pilot inputs. The factors I suspect were stealth, startle and IMC spatial disorientation. The question that I have is whether the left or right thrust was the one malfunctioning. Since the initial event was a L rollback during climb, which was compensated by autopilot during a subsequent level-off, and then the a/p disconnect and more severe roll occurred after a readback to resume the climb, it would make more sense to me if the R thrust was functioning and the L thrust was not. The resumption of climb thrust would decrease, not increase the thrust assymmetry if the problem was with the R engine thrust being 'stuck', but intensify it if the problem was with the L engine being inoperably retarded.
here he goes again... who said anything about IMC? in fact,
"The superimposed ADS-B-based flight profile with radar weather image at 1438 LT provided by the BMKG indicated that the radar intensity level along the flight profile was not more than 25 dBz9, which means that the flight path did not indicate any significant development of clouds."
anyway, what is absolutely astonishing is that the report makes no mention of any crew action at all. AP disconnected at 45 degree bank angle. i know we've been through this before, but i still find it hard to believe that two pilots with over 15,000 on-type experience between the two of them didn't notice 1) left throttle lever PHYSICALLY moving and 2) a bank angle of 45 degrees. i mean holy shyte! WTF were they doing up there?
here he goes again... who said anything about IMC? in fact,
"The superimposed ADS-B-based flight profile with radar weather image at 1438 LT provided by the BMKG indicated that the radar intensity level along the flight profile was not more than 25 dBz9, which means that the flight path did not indicate any significant development of clouds."
A flight path that does "not indicate any significant development of clouds" is not necessarily VMC. The flight was delayed on the ground for 30 minutes due to heavy rain. There were significant clouds around the pattern when they took off. I would not expect cloud-free skies that point. It's not unlikely that they may have entered IMC at that moment. Take a look at the white stuff on the satellite image taken ten minutes after the last radar contact. (IMC = anything not meeting VMC minima).
You ask how they could not notice a 45deg bank in VMC. Unless their attention was directed downward the entire time, I don't think they could miss that. Unless the visual horizon was obscured by... some white stuff...
anyway, what is absolutely astonishing is that the report makes no mention of any crew action at all. AP disconnected at 45 degree bank angle. i know we've been through this before, but i still find it hard to believe that two pilots with over 15,000 on-type experience between the two of them didn't notice 1) left throttle lever PHYSICALLY moving and 2) a bank angle of 45 degrees. i mean holy shyte! WTF were they doing up there?
There is room for ambiguity in the 45 degrees statement.
At 14:40:05 LT, the FDR data recorded the aircraft altitude was about 10,900 feet, which was the highest altitude recorded in the FDR before the aircraft started its descent. The AP system then disengaged at that point with a heading of 016 degrees, the pitch angle was about 4.5 degrees nose up, and the aircraft rolled to the left to more than 45 degrees.
The aircraft was already rolled more than 45 degrees when the AP disconnected? Or did the airplane roll to more than 45 degrees when the AP disconnected?
And the power lever... it seems that is was slowly closing.
At 14:38:42 LT, the FDR data recorded that as the aircraft climbed past 8,150 feet, the thrust lever of the left engine started reducing, while the thrust lever position of the right engine remained. The FDR data also recorded the left engine N1 was decreasing whereas the right engine N1 remained.
At 14:40:05 LT, the FDR data recorded the aircraft altitude was about 10,900 feet, which was the highest altitude recorded in the FDR before the aircraft started its descent. The AP system then disengaged at that point with a heading of 016 degrees, the pitch angle was about 4.5 degrees nose up, and the aircraft rolled to the left to more than 45 degrees. The thrust lever position of the left engine continued decreasing while the right engine thrust lever remained.
While this gave the pilots more time to detect and react, the situation is more subtle and inconspicuous. You would for sure perceive the sudden yaw and sudden loss of thrust that comes with a sudden engine failure. But a rolling back through 1 minute and 20 seconds with the AP compensating for it, it is much more hard to detect. Yes, they have to look at the instruments, and it is inexcusable if they didn't. But perhaps the cues in the PDF were not so strong (yes, either the speed or the climb rate must have been going slowly down, but perhaps the attitude was kind of normal), they would have to see the split lever (rather than the lever "moving" if it was moving so slow) or the split engine instruments, which are not so much the focus of attention as the primary flight instrument. Add yes, they were likely distracted looking the weather radar and talking with ATC to negotiate headings and altitudes to avoid weather and conflicting traffic, and probably making manual altitude and heading entries in the AP. Again, not excusable, one of them should have been flying the plane and monitoring the instruments. But I can see that happening.
Safety in aviation doesn't so much depend on everything working as expected but on assuming that things will NOT work as expected and having multiple layers covering your arse.
If the only safeguard for 1.5 minutes of pilot inattention causing the sudden disconnection of the AP and the neutralizing of the compensating control inputs that it was stealthy doing and the airplane suddenly rolling upside down is to have the same distracted pilot startle, get their shit together, make the correct control inputs, and complete the recovery (when the pilots were not even so well trained in upset recovery to begin with), I call that a latent safety issue.
--- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
--- Defend what you say with arguments, not by imposing your credentials ---
That's what they did (though not optimally). It didn't work.
--- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
--- Defend what you say with arguments, not by imposing your credentials ---
I know most of you have already seen this. Watch it again! And keep in mind that this is from 1997. How much more have we automated in 24 years! https://youtu.be/V5_GENu6aqg
A B-717 spins out of crossed up slow flight. The pilots take PROMPT corrective action.
25 seconds and 10,000 feet lower (ironingly similar) they have recovered.
Gabriel makes a very strong argument that given a very short delay of the SURPRISE in the 737 versus the 717 guys ALREADY hand flying in a mentally primed CRITICAL flight mode, and the chances of a realistic recovery are thin.
My only thought is that proper upset recovery would include PUSHING “up” while inverted, as opposed to the near vertical dive.
Gabriel, you may have to repeat this because I think Bobby ignores me from I called him out for ragging on ignorant outsiders after insisting that anti skid didn’t function when it was most needed.
Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.
We process personal data about users of our site, through the use of cookies and other technologies, to deliver our services, personalize advertising, and to analyze site activity. We may share certain information about our users with our advertising and analytics partners. For additional details, refer to our Privacy Policy.
By clicking "I AGREE" below, you agree to our Privacy Policy and our personal data processing and cookie practices as described therein. You also acknowledge that this forum may be hosted outside your country and you consent to the collection, storage, and processing of your data in the country where this forum is hosted.
Tweet
Comment