Flydubai Flight 981 Crashes on Landing in Rostov-on-Don, Russia

[Gabriel]

There should be. There isn't. Just stick shaker. This isn't a 21st century aircraft.

What??? You need a 21st century airplane for an alarm that says "pitch, pitch"?
I wonder how they did to implement the "bank angle, bank angle" one back in the 737-300 (because it is there where I experienced it).

It probably takes as much as re-flashing the firmware from version 75.7.25.103 to 75.7.25.104.

Perhaps there should just be a warning that says "PRIMARY FLIGHT DISPLAY" in such situations.

Well, I always say that the Primary Flight Display has the words "primary" and "flight" in its name for a good reason. A PF, pilot flying, that is not looking and the PFD, is not flying.

[Gabriel]

I suspect much of the real intent is to avoid lawsuits.

If someone on the board says "we think the plane crashed because the pilot is a dummy" and it turns out the crash was due to a bomb, the person on the board (and probably the other board members, the authority that convened the board, the owner of the building where they met, all the board members' pets...) is guaranteed to get sued.

If the board says nothing and lets people *think* the plane crashed because the pilot was a dummy, the board doesn't get sued.

They don't need to say anything, and especially not any opinion. They just have to release the PDF and FDR FACTUAL information.

[Evan]

They don't need to say anything, and especially not any opinion. They just have to release the PDF and FDR FACTUAL information.

If they know at this point that there was disorientation behind this and that this resulted from a departure from standard procedure, they need to release that tidbit as well. At least within the industry.

On the other point, a 21st-century airplane places less blind faith in the pilots and gives the airplane the means to defend itself (and the cargo). I think they did the roll alert back then because even those F-100 pilots coming out of the military might not understand the implications of roll in a large swept-wing airliner, whereas every cropdusting cowboy knew about the dangers of pitch. Just a hunch.

[Gabriel]

If they know at this point that there was disorientation behind this and that this resulted from a departure from standard procedure, they need to release that tidbit as well. At least within the industry.

That would be pretty obvious from the CVR and FDR factual information. Especially for the industry.

[TeeVee]

Indeed.

If it was the plane at fault, that would be forthcoming rather quickly, especially if a fleetwide concern existed. If it was the pilots, there is no particular hurry. Why not? What's the difference here? A threat is a threat. If FlyDubai is running a substandard training regimen, or if a certain universal human stealth factor has been revealed, that information needs to be shared as soon as possible. I realize that it takes quite a while in most cases to make a solid determination, but why not error on the side of safety and at least raise an alarm in the industry?

Oh, right, lawyers.

if every time a pilot effed up they ran out and screamed pilot error, do you think that would radically change the behavior of other pilots? if so, how is it that you can cite multiple, similar pilot error type crashes? Survey says!: evan is wrong and merely screaming out to the world that these two DEAD pilots caused the crash will change nothing. fact is, unless there are directives issued, airlines are not going to require pilots to read interweb fora and most pilots do not read the shit about other pilots' eff ups unless required to.

your last comment proves that you are truly a jack-donkey. err on the side of safety by running out half-cocked makes more sense. and as a lawyer i agree that everyone should run out half-cocked screaming shyte so that others will be safe.

[elaw]

...there was disorientation behind this and that this resulted from a departure from standard procedure...

Still seems to me a high probability that somatogravic illusion was behind this.

Okay now I'm really confused.

Researching "somatogravic illusion", everything I could find states that it results from the effect of unnatural acceleration forces on the human vestibular system... not from departure from standard procedure. How does following a standard procedure prevent somatogravic illusions?

[3WE]

Okay now I'm really confused.

Researching "somatogravic illusion", everything I could find states that it results from the effect of unnatural acceleration forces on the human vestibular system... not from departure from standard procedure. How does following a standard procedure prevent somatogravic illusions?

It's his black and white thinking.

He's sort of right, because in my cowboy mind you want to.watch the instruments and maintain a very high situational awareness to prevent or correct incidences of $20 disorientation...

Just make sure that's in the FCOM of the 737-8 for hand flown go arounds in 12 knot crosswind components, because general rules have no.value.

[Evan]

Okay now I'm really confused.

Researching "somatogravic illusion", everything I could find states that it results from the effect of unnatural acceleration forces on the human vestibular system... not from departure from standard procedure. How does following a standard procedure prevent somatogravic illusions?

A go-around executed correctly will not place high acceleration forces on the vestibular system. Procedure provides for this in a number of ways.

Every procedure I've seen for go-around uses autothrottle (unless unavailable). The 737 autothrottle is designed to produce a reduced N1 go-around thrust with the first push of the button sufficient to establish a 1000-2000fpm climb. If, for some reason, you need TOGA N1, you can push the button a second time once you have reached the initial limit. If you improvise the go-around, you might simply firewall the thrust levers (especially if you are very frustrated after waiting two hours in a holding pattern and have to abandon yet another approach). This will result in pitch coupling and amplify your pitch commands to place you in a steep climb gradient. If you suddenly level off from this situation, you may very well experience somatogravic effects.

Every MANUAL go-around procedure I have seen requires careful adherance to the flight director (with the autothrottle). If you improvise, you might get into an atitude that, well... see the last paragraph.

The SAFEST procedure, if ILS and both autopilots are available, is to use all of them. Once established on the ILS with both autopilots you have a) a fail-passive approach to the threshold and b) an automatic go-around that is fail-passive. The odds of losing situational awareness due to vestibular illusions has now been reduced to virtually zero.

The standard procedure with ILS approach is to use both autopilots and to execute go-arounds on autopilot. Does that answer your question?

[elaw]

Actually yes it mostly does!

My one comment would be that I suspect the somatogravic illusion is not a binary thing... if a 2-G rearward acceleration produces a certain effect, a 1.5-G will produce a similar effect, just lesser in magnitude. So in that case using the automation functions would not result in *no* effect, it would just be less.

On the other hand, the human body does work in strange ways sometimes and there might be some kind of threshold to the effect, either physical or psychological (or both).

But I still believe that a somatogravic illusion (or something else of a similar nature) is not enough to explain an accident all by itself. Suddenly transitioning from a descent to a climb in IMC is something that is done in aircraft all the time, and was done successfully many times before automation even existed. There has to be some additional factor(s) at play.

[Evan]

Actually yes it mostly does!

My one comment would be that I suspect the somatogravic illusion is not a binary thing... if a 2-G rearward acceleration produces a certain effect, a 1.5-G will produce a similar effect, just lesser in magnitude. So in that case using the automation functions would not result in *no* effect, it would just be less.

On the other hand, the human body does work in strange ways sometimes and there might be some kind of threshold to the effect, either physical or psychological (or both).

But I still believe that a somatogravic illusion (or something else of a similar nature) is not enough to explain an accident all by itself. Suddenly transitioning from a descent to a climb in IMC is something that is done in aircraft all the time, and was done successfully many times before automation even existed. There has to be some additional factor(s) at play.

Yes, pilot error! Of course a fully manual go-around is perfectly safe if there is no loss of situational awareness and no unusual attitudes involved. But the conditions are now fertile for these things to happen. Just add fatigue, inexperience, poor training etc. etc. and things can get bent. Aviation safety is about avoiding these areas of 'fertility'.

Even under vestibular disorientation, a pilot can follow the FD and be fine, or even the PFD without the FD. But the effect is apparently quite convincing if they do not.

Under automation (assuming the automation is functioning correctly) there are no significant somatogravic effects because nothing is abrupt: before leveling off at the MCP altitude, the autopilot pitch mode becomes ALT ACQ, which is an transitional mode, then ALT HOLD. By design, thrust comes on smoothly, attitude is controlled smoothly. Aerospace engineers have known about somatogravic issues for a long time. I'm not sure I can say the same about certain pilot training programs...

[Gabriel]

http://avherald.com/h?article=495997e2&opt=0

On Apr 8th 2016 the MAK reported in Russian, that the actual weather conditions were as forecast and reported by the weather systems, all weather instruments were working normally. According to flight data recorder the first approach was performed in manual control in adverse weather conditions with winds from 230 degrees at 13m/s gusting 18m/s (25 knots gusting 35 knots), light rain, mist and light to moderate windshear. At an altitude of 340 meters/1100 feet the crew initated a go around due to a sudden change of direction and strength of wind and entered a hold to wait for improvement of weather conditions. The second approach was also flown in manual control, at a height of 220 meters/720 feet about 4km/2.2nm from the runway threshold the crew initiated a go around again selecting TOGA on the engines. At an altitude of 900 meters/2950 feet there were simultaneous actions, the crew returned the flight controls to neutral and the pitch trim moved 5 degrees nose down causing a rapid nose down movement of the aircraft resulting in -1G of vertical acceleration. The subsequent crew actions did not permit to pull the aircraft out of the resulting nose dive, the aircraft impacted ground at a speed of about 600kph/325 knots with the nose more than 50 degrees below the horizont.

The MAK doesn't mention anything special between the initiation of the go-around at 720ft and the start of the upset at 2950 ft. Which doesn't necessarily mean that there wasn't anything special.

[Evan]

http://avherald.com/h?article=495997e2&opt=0



The MAK doesn't mention anything special between the initiation of the go-around at 720ft and the start of the upset at 2950 ft. Which doesn't necessarily mean that there wasn't anything special.

More questions than answers....

_Why the manual approaches? DId they think the conditions exceeded autopilot capability? Was the retiring pilot trying to get in some final stick time?

_Was the autothrottle engaged? If not, was the autothrottle armed?

_How did they "select TOGA"? Did they push the button(s) or did they advance the thrust levers manually?

_Was the AND pitch trim commanded? (is pickle switch position an FDR parameter?)

_What were the "subsequent crew actions"?

[Evan]

_Was the AND pitch trim commanded? (is pickle switch position an FDR parameter?)

Actually, that wouldn't necessarily tell us anything. If the trim was runaway and the cause was the switch, it would appear commanded on the FDR plot (assuming there is one).

[ATLcrew]

Survey says!: evan is wrong and merely screaming out to the world that these two DEAD pilots caused the crash will change nothing...

I'd even take it a step further: nothing ANYONE says on this forum will change anything.

[Gabriel]

They went from say 15 degrees nose up to more than 50 degrees nose down in a bunch of seconds. They went from +1G to -1G probably a much smaler bunch of seconds. There is no way to miss that. Spatial disorientation, somatogravic or somatogyral illusions, fatigue, and what not may help explain why this was caused in the first place and maybe why they could not recover, but they immediately knew that there was something very very bad going on.

This is very speculative and fuzzy, but it looks that something wrong happened with the trim (runaway or unintentional command in the AND direction) and that they could not recover from the subsequent pitch-down (and then dive) be it because they didn't recognize the trim issue or because they could not correct it, and elevator alone was not enough.