Aeroflot Superjet 100 fire and evacuation at UUEE

[3WE]

3WE...I fear there are still many out there who think like him.

...pilots who think that one should know procedures well, AND NOT forget fundamental aerodynamics either...that maybe relentless pull ups are bad, airspeed should be watched on short final (with a hand on the power levers), and that a go-around near decision height should involve appropriate, near-full power and a measured, but skillfully applied, crisp pull up while monitoring key flight parameters?

You do have strange ideas.

[ATLcrew]

No, don't "close the investigation". Often investigations reveal other vulnerabilites. I would still recommend:

1) Exploring the reason for more active control inputs by pilots when in DIRECT law (is it training? Is it psychological? Is there something about the aircraft response in DIRECT mode?)

(The latter is unlikely as the aircraft was flown extensively during flight testing in DIRECT mode and some of the test pilots commented that it is very benign and even that they prefer the handling in this mode)

2) Exploring ways to prevent the EIU units from rebooting after a lightning strike, thus keeping the aircraft in Normal Law. And making the avionics generally more fail-passive during a similar lightning strike.

3) Exploring ways to ensure that the gear doesn't damage the spar web during multiple impacts in excess of 4G.

4) Exploring the human factors involved that led to very poor decision-making.

5) Exploring ways to overcome risk-taking and overconfidence and strengthening respect for the danger posed by severe weather systems.



How many are within your control? The industry already made this determination. Stable Approach Criteria is gospel now. Not going around when passing a certain stabilization gate outside the criteria is explicitly dangerous and a violation of safety culture (and pilot-passenger trust). Man, that's progress! No more dive and drive. No more passenger roulette. And the same goes for a predictive windshear warning, let alone two of them.

Sure, there will always be exceptional circumstances where this does not apply. But none of those circumstances were present here. If there were, the report makes no mention of them.

As far as this report tells us, the crash was the result of pilots bent on task completion and willingly gambling with passenger lives on the outcome. That cannot be tolerated. They lost about 40 chips on this hand.



The incident crew inputs are far more pronounced in the final moments. The other flights may have been more stabilized. The principal difference is that the other flights landed intact and did not bounce.

What is different about them? Perhaps the other flights didn't try to advance thrust or make abrupt pitch occillations to the stops at the last moment in a desperate attempt to stabilize rather than go-around.

I'm actually surprised that you would defend them Gabriel after all we've learned. You blow off the stable approach criteria (or fly into known wind-shear*), all bets are off and anything can happen. Everything after that is just what happened, not why it happened...

* known wind-shear = the predictive wind shear has told you there is wind shear ahead and the possibility strongly exists.

Gratefully noted and acknowledged.

[Evan]

I will take a vow of silence for a while.

I've presented the finding of the report.
There is nothing in it to justify an immediate landing when doing so was that risky.
There is nothing in it that prevents them from holding, burning off fuel and methodically working the problem.
There is nothing in it that prevents them from completing the required checklists (which they skipped).
There is nothing in it that justifies ignoring warnings and disregarding CRM.

There is something in it that confirms that company pilots were repeatedly using bad stick technique is DIRECT mode (but nothing that explains or justifies it or suggests that it is due to control difficulties).
That needs further investigation because SOMETHING needs to be fixed there. I believe that something is, as you have pointed out, a lack of training and practice in that mode.

HOWEVER,

There is nothing in that correlates amplified stick handing in DIRECT mode with a bounced landing. Of the 7 flights on the plot, 6 of them did not bounce on landing.
There is nothing in it the corelates amplified stick handling in DIRECT mode with excessive landing speed. Of the 7 flights on the plot, 6 of them landed below 150kts.

As for the crash itself, we all know that bouncing is the result of being unstable in the final and that SOP is to go-around as a minimum cautionary requirement. That lesson was hard-learned through repeated tragedy and w/o aircraft. Go-arounds not only prevent 100% of bounced landings, they are the only sure way to prevent them. You can argue that better recovery technique will suffice but we both know that the required level of skill under pressure is more than we can expect of pilots across the industry, so it can't be the solution. The only solution is to never land in that unstable condition if a go-around is a safe option. The solution is to never proceed beyond MDA in that condition. The solution is to continue to defend the passengers against pilot over-confidence and risk-taking (which remains very entrenched in the culture).

3WE can continue to be the apologist for rule-breaking and risky decision-making. He is actually a valuable member of this forum because he represents the rebellious and cavalier mentality behind this very entrenched problem and the continued resistance to learning humility from the mistakes of history. He's an interesting case-study. Thankfully, he's not an airline pilot, but I fear there are still many out there who think like him.

[Gabriel]

I will take a vow of silence for a while.

[3WE]

***you can't expect 10 minutes of direct-mode flight every 6 months to kick in in an emergency when in the mean time you have flown hundreds of hours and hundreds of take-offs and landings in normal mode.***

***I am just giving them the benefit of the doubt.***I am just not being as quick as you at judging them.***

There, I quoted you verbatim. (There's probably another couple of key one liners I should snip out and add)

That's essentially my whole point.

***The question is WHY they didn't go around. Evidently they were in a rush to land or they would have burned fuel before landing overweight.
So again, don't paint me the picture WE HAVE NOW about the condition of the plane. Tell me what picture THEY HAD THEN, and whether that picture was justified or not given the information THEY had at that time.***

Bingo. My point, Gabieee verbatim. I WILL say that I have confirmation bias that FORCING the landing may very well have been a reasonable choice AT THE TIME...some of your sensitive electronic stuff is fried by a big ass lightning bolt that is not supposed to fry stuff. THEY WERE OUTSIDE OF THE NICE NORMAL WORLD IN THE FCOM THAT WAS WRITTEN AT 0 KTS AND 0 FT AGL...and Evan cannot comprehend that the pilots might have REASONABLY suspected other important stuff was fried and on borrowed time...great to figure all that out AFTER they landed...but they were kind of busy WHILE they were flying.

MAYBE EVEN WATCHING FOR THAT AWFUL WIND SHEAR THAT IT SEEMS MAY NEVER HAVE HAPPENED!

[3WE]

I am not saying that it WAS adversely affected. I am saying it a concerning possibility that needs to be investigated.

It is interesting how both you and Evan take my I-don't-know's trying to use them to make your points. You are both exercising confirmation bias by ignoring parts of what I say, taking other parts, and twisting/interpreting them in specific ways that fit your positions and did not exist when I said what I said.

Ok, Mr. Hair splitter. Correct, I am not quoting you verbatim. (Friendly razz, no anger)

You said that six? crews were using much bigger control inputs when on alternate law- in particular in pitch...that the Alternate Law inputs and feelings were markedly different from normal law. Correct? (not verbatim, but you did say essentially that).

Correct(no question mark), neither of us KNOW that that made THAT plane bouncy...

BUT

Would you agree that that is CONSISTENT with a situation that would TEND to support bouncing.

AND

Would you agree that a plane that is already fast, and already at a reduced flap setting, that something like that is likely to MAYBE make the plane more bouncy?

I'm not asking if you KNOW...I'm asking, "Would that be CONSISTENT with making the plane more bouncy"

What you and I know that Evan cannot comprehend is that you go fast and have fewer flaps and the plane feels and acts differently...and you are a little bit (emphasis on a little bit) off your A game...you can get into a bounce...Not that much different than the Narita guys.

Sure, study it further- but it's not MY confirmation bias as much as Evan's total dismissal that maybe alternate made it feel different and might have contributed. And Evan, please find where I said that was THE PRIMARY CAUSE...I don't work in black and white like you do.

[Evan]

Don't try to convince me of what I am already convinced. The question is WHY they didn't go around. Evidently they were in a rush to land or they would have burned fuel before landing overweight.
So again, don't paint me the picture WE HAVE NOW about the condition of the plane. Tell me what picture THEY HAD THEN, and whether that picture was justified or not given the information THEY had at that time. It was just pure disregarding of the rule? Or was there something else? (including human factors like load saturation, tunnel vision, etc)

Ok, let's paint.

Was there anything in the report about a CVR conversation between the pilots regarding control difficulties? No.
Was there any report to the tower about the state of the aircraft? Yes!

Upon the controller ' request: "Any assistance required?", the crew responded: "No, so far
everything is normal, standard". Upon the controller's request for clarification, the crew responded
that they were having problems with the radio communication and that the automatic flight control
had been lost.

Did the subsequent CVR conversations seem acutely stressed, as if the pilots were flighting to control the aircraft? None that I can detect.

At 15:26:05, the crew had the following conversation: the FO "Should I set the transponder
to 7700? Or leave as it is? Ok, roger.". The PIC: "We could have done it long before, actually."

I also wanted to give the pilots the benefit of the doubt and I did, by waiting until the report came out. But here it is, and it's pretty damning...

I am not saying they did right or wrong. I just want more information before judging.
I am just giving them the benefit of the doubt. By now. I am just not being as quick as you at judging them.

Are you sure about that, cuz...

Ok, let's see...

Don't take-off in adverse weather conditions.
Don't fly into a knowingly active storm.
Burn fuel before landing.
Of course go around when you got the wind-shear warning and when the speed, vertical speed and pitch became unstable shortly before touchdown (which is the ONLY thing that you claim can avoid this)
Don't use the sidestick as a binary input device (full-up, full-down).
Go around after the first bounce.
Go around after the second bounce.
Don't shove the nose down after the 1st bounce.
Don't shove the nose down after the 2nd bounce.

If that's what you call not judging them...

As you know, while I am judging their actions and judgments, I am interested only in why they were bad ones. Poor training? Poor safety culture? Human factors? I only want to know what is broken and can can be fixed.

[Gabriel]

Keep in mind they were hit by lightning and had some sensitive electronic crap knocked out AND Gabriel is telling us that the handling was adversely affected...

I am not saying that it WAS adversely affected. I am saying it a concerning possibility that needs to be investigated.

It is interesting how both you and Evan take my I-don't-know's trying to use them to make your points. You are both exercising confirmation bias by ignoring parts of what I say, taking other parts, and twisting/interpreting them in specific ways that fit your positions and did not exist when I said what I said.

[Gabriel]

No, don't "close the investigation". Often investigations reveal other vulnerabilites. I would still recommend:

1) Exploring the reason for more active control inputs by pilots when in DIRECT law (is it training? Is it psychological? Is there something about the aircraft response in DIRECT mode?)

(The latter is unlikely as the aircraft was flown extensively during flight testing in DIRECT mode and some of the test pilots commented that it is very benign and even that they prefer the handling in this mode)

2) Exploring ways to prevent the EIU units from rebooting after a lightning strike, thus keeping the aircraft in Normal Law. And making the avionics generally more fail-passive during a similar lightning strike.

3) Exploring ways to ensure that the gear doesn't damage the spar web during multiple impacts in excess of 4G.

4) Exploring the human factors involved that led to very poor decision-making.

5) Exploring ways to overcome risk-taking and overconfidence and strengthening respect for the danger posed by severe weather systems.

Yes, for example those. I would also like them to investigate the handling qualities in direct mode for normal pilots used to fly 100% in normal mode. Not for test pilots who are super trained in handling airplanes with no previous experience in type, different flight modes, and remaining in control in situations prone to oscillation (including PIO but also loss of stability and loss of damping). The way that all other crews used the stick is at least concerning and worth investigation further. Of course, also training in direct mode, but you can't expect 10 minutes of direct-mode flight every 6 months to kick in in an emergency when in the mean time you have flown hundreds of hours and hundreds of take-offs and landings in normal mode. Also, what information did the pilots had about the real condition of the plane, how was their decision making, what was their mindset, how was the judgement process, and why there was a sever breakdown in CRM.

How many are within your control?

Ok, let's see...

Don't take-off in adverse weather conditions.
Don't fly into a knowingly active storm.
Burn fuel before landing.
Of course go around when you got the wind-shear warning and when the speed, vertical speed and pitch became unstable shortly before touchdown (which is the ONLY thing that you claim can avoid this)
Don't use the sidestick as a binary input device (full-up, full-down).
Go around after the first bounce.
Go around after the second bounce.
Don't shove the nose down after the 1st bounce.
Don't shove the nose down after the 2nd bounce.

The industry already made this determination. Stable Approach Criteria is gospel now. Not going around when passing a certain stabilization gate outside the criteria is explicitly dangerous and a violation of safety culture (and pilot-passenger trust). Man, that's progress! No more dive and drive. No more passenger roulette. And the same goes for a predictive windshear warning, let alone two of them.

Don't try to convince me of what I am already convinced. The question is WHY they didn't go around. Evidently they were in a rush to land or they would have burned fuel before landing overweight.
So again, don't paint me the picture WE HAVE NOW about the condition of the plane. Tell me what picture THEY HAD THEN, and whether that picture was justified or not given the information THEY had at that time. It was just pure disregarding of the rule? Or was there something else? (including human factors like load saturation, tunnel vision, etc)

I am not saying they did right or wrong. I just want more information before judging.

As far as this report tells us, the crash was the result of pilots bent on task completion and willingly gambling with passenger lives on the outcome.

As you can see, I am not convinced of that. Neither of the opposite.

I'm actually surprised that you would defend them Gabriel after all we've learned.

I am not. I am just giving them the benefit of the doubt. By now. I am just not being as quick as you at judging them.

[Evan]

Keep in mind they were hit by lightning and had some sensitive electronic crap knocked out AND Gabriel is telling us that the handling was adversely affected...

Gabriel?

[3WE]

***Keep in mind the QRH and FCOM state that you may disregard the warning ONLY if there is no wind shear hazard and there are no other signs of wind shear.***

That's cute.

Keep in mind they were hit by lightning and had some sensitive electronic crap knocked out AND Gabriel is telling us that the handling was adversely affected...

And, I guess there's NO wind shear warnings here, so hey perfectly OK: https://www.youtube.com/watch?v=w4EQuM_t8Fo

Yeah, doesn't matter that there's no ACTUAL reported wind shear in the final report, those pilots violated the POHFCOMQRH- put a little dot on the sentence and call it done.

https://www.youtube.com/watch?v=f0Xt-Zwh9nY

[Evan]

I suspect Bobby, ALT or VNAV5 could have salvaged the landing.

I suspect Bobby, ALT or VNAV5 would have gone around when there was still time.

...and then we see no real evidence of actual wind shear.

1) A standing warning:

At 13:47 wind shear warning 2 for Sheremetyevo for the period from 14:00 till 18:00 on
May 05 2019 issued: wind shear in the approach forecasted: UUEE WS WRNG 2 051347 VALID
051400/051800 WS IN APCH FCST.

2) A predictive windshear warning (that wasn't even discussed!!).

At 15:28:26, at 1600 ft (490 m) QNH (at the radio altitude of 1100 ft (335 m)) the
"WINDSHEAR" warning was recorded, it is generated by the crew alerting system based on the
wind shear forecast data of the weather radar, accompanied by the voice message "GO-AROUND,
WINDSHEAR AHEAD". This warning informs the crew about the possible wind shear enroute. The
alert lasted for 11 seconds and included 2.5 warning cycles (5 voice messages, each cycle consisting
of 2 messages one after another, with the 1 second interval in-between). The crew had no discussion
of this warning triggering.

3) The obvious presence of CB activity, such as the one you just recently flew into.

Keep in mind the QRH and FCOM state that you may disregard the warning ONLY if there is no wind shear hazard and there are no other signs of wind shear. Clearly there was a wind shear hazard and I would consider the proximity of CB's to be 'other signs of windshear'.

[Evan]

The oversimplification of the week, close the investigation. In this case, the little dot is not a period (or a full stop). Is a "to be continued".

No, don't "close the investigation". Often investigations reveal other vulnerabilites. I would still recommend:

1) Exploring the reason for more active control inputs by pilots when in DIRECT law (is it training? Is it psychological? Is there something about the aircraft response in DIRECT mode?)

(The latter is unlikely as the aircraft was flown extensively during flight testing in DIRECT mode and some of the test pilots commented that it is very benign and even that they prefer the handling in this mode)

2) Exploring ways to prevent the EIU units from rebooting after a lightning strike, thus keeping the aircraft in Normal Law. And making the avionics generally more fail-passive during a similar lightning strike.

3) Exploring ways to ensure that the gear doesn't damage the spar web during multiple impacts in excess of 4G.

4) Exploring the human factors involved that led to very poor decision-making.

5) Exploring ways to overcome risk-taking and overconfidence and strengthening respect for the danger posed by severe weather systems.

How many endings can you imagine for the sentence "the crash would not have happened if ______"?

How many are within your control? The industry already made this determination. Stable Approach Criteria is gospel now. Not going around when passing a certain stabilization gate outside the criteria is explicitly dangerous and a violation of safety culture (and pilot-passenger trust). Man, that's progress! No more dive and drive. No more passenger roulette. And the same goes for a predictive windshear warning, let alone two of them.

Sure, there will always be exceptional circumstances where this does not apply. But none of those circumstances were present here. If there were, the report makes no mention of them.

As far as this report tells us, the crash was the result of pilots bent on task completion and willingly gambling with passenger lives on the outcome. That cannot be tolerated. They lost about 40 chips on this hand.

The report shows a very real and concrete contrast between the manual flight control inputs in normal law vs in direct law. For the accident crew and for another half dozen of other crews. The sample of control inputs in direct law are ALL of a much bigger amplitude and alternating both directions. For the accident crew and for the other half dozen or so.

The incident crew inputs are far more pronounced in the final moments. The other flights may have been more stabilized. The principal difference is that the other flights landed intact and did not bounce.

What is different about them? Perhaps the other flights didn't try to advance thrust or make abrupt pitch occillations to the stops at the last moment in a desperate attempt to stabilize rather than go-around.

I'm actually surprised that you would defend them Gabriel after all we've learned. You blow off the stable approach criteria (or fly into known wind-shear*), all bets are off and anything can happen. Everything after that is just what happened, not why it happened...

* known wind-shear = the predictive wind shear has told you there is wind shear ahead and the possibility strongly exists.

[Gabriel]

The quick, flippant, and poorly-supported answer is "not made it worse". As to exactly how they should have done that, I don't know.

I suspect Bobby, ALT or VNAV5 could have salvaged the landing.

You and ALT alluded to bounce mitigation techniques. They sounded interesting (although yours was pretty aeronengineerish).

Yes, but all that assumes that you have good control over the airplane (in particular in the pitch axis). They lacked that since the flare if not early.

[3WE]

***By the way, what is your idea of what the pilots should have done after the bounce?***

The quick, flippant, and poorly-supported answer is "not made it worse". As to exactly how they should have done that, I don't know.

I suspect Bobby, ALT or VNAV5 could have salvaged the landing.

You and ALT alluded to bounce mitigation techniques. They sounded interesting (although yours was pretty aeronengineerish).

As some of your recent comments indicated, simulated crews were making some bigger control inputs- indicating they might be slightly behind the plane...fascinating.

By the way- it's very possible that the crew botched something that an ordinary, properly-trained crew generally should not botch...but I dislike the notion that the dummies were saying yee haa, screw it that we aren't stabilized let's land this byotch just because we are improvisational cowboys is WRONG...they were dealing with a fair amount of crap including maybe a shred of fear for their lives...