Dash 8/Q400 Stall Recovery stats...

[Firebird2XC]

I had a thought the other day and wanted to see if it pans out any.

In the 3407 crash, the press repeatedly mentions that the Captain 'overrode' the stick pusher.

My question is this- first off, can you actually override it, or just yank the yoke back once it activates?

Secondly, assuming you're in a full stall and it activates the pusher. Given the approximate conditions:

A) how long would the airplane take to recover and begin building airspeed

B) how much altitude is lost?

My real interest is this- I remembered being allowed to ride a stall all the way to the pusher in the embraer sim, and the altitude loss was somewhat alarming. Granted, fixed wing jet performance in a full stall will have different aerodynamics.. but.

Considering the very low altitude AGL in the vicinity of impact and that man made obstacles might reach much higher, the actual clearance for 3407 was very, very small for an airplane that size.

Add in that the event took place in the dark when see and avoid is much harder and visual illusions come into play.. well.. yeah.

If Renslow knew that the airplane encountered a full stall nose-down recovery and would lose a certain amount of altitude, he knew that the only way to recover the airplane safely at that point was to avoid altitude loss at all costs.

If that's the case, then 3407 was doomed once that airspeed bled off, and it was not Renslow's corrective action that sealed their fate- just the crew's inattention to their original airspeed loss.

This does, of course, assume that most of what is believed about the stall-spin scenario is completely correct and that there were no other contributing factors. I'm by no means pinning anything on the crew at this point in time.

Here's why I ask- If it was the airspeed loss that doomed them, then momentary inattention, not actual action, is what did it. The argument for pilot fatigue and performance thereafter becomes much more of an issue than before.

It's also interesting to note how much a few seconds of additional warning may have helped the crew. Could in fact some sort of aural warning to alert the crew to airspeed loss prior to shaker/clacker activation prevented their loss of Situational Awareness? Some less less dramatic, perhaps.

Thoughts, everybody?

 

[derg]

A good reference while reading the above thread:

[yt]lxywEE1kK6I[/yt]

 

[TUCKnTRUCK]

Yes, when you exceed a set resistance threshold, it will automatically kick the pusher off line. This is to allow the pilot to override the pusher.

What I have been told is that if you override the pusher, it also turns off the elevator feel, as the system would not be sure which actuator was broken. Also, I also do not know if the pusher would activate on both yokes should they be separated for a stuck surface, or if that alone would disable the system.

The gearing on the props keeps the engines turning right in the "poweband" so power is available almost instantly.

I have no idea how much altitude is lost, as I've never done that in training. I've never been close to the "pusher" sim or otherwise.

Honestly, I am pretty sure if they had slammed the power up, and touched nothing else, it likely would have recovered. The Q floats and balloons with every little gust of wind. It has a very high aspect ration wing, which causes it to act like a glider in thermals and gusts. Trying to keep the airspeed at ref+10 on approach is a short ticket to the looney bin. For a while Colgan pilots were limited to Vm0 - 10kts... the speed jumps around so much people would over speed very easily.

From this, I am sure that the NTSB is going to present a long, long line of Domino's... and many places where somebody ( Pilot, dispatch, atc, training etc) could have pulled one out of the line, and prevented this from happening.

 

[ctab5060X]

If that's the case, then 3407 was doomed once that airspeed bled off, and it was not Renslow's corrective action that sealed their fate- just the crew's inattention to their original airspeed loss.

I have thought that was the underlying problem from the first time I read the CVR transcript. Looking at the overall picture from the accident, the Q400 crashed from a stall, but what caused the stall? If you think if it in those terms, it starts to indicate it was the uncorrected loss of airspeed.

It's also interesting to note how much a few seconds of additional warning may have helped the crew. Could in fact some sort of aural warning to alert the crew to airspeed loss prior to shaker/clacker activation prevented their loss of Situational Awareness? Some less less dramatic, perhaps.

Thoughts, everybody?

Odd that you should post this today.

I was talking about something similar with another FO this morning. From what the NTSB released, the airspeed degradation was something like 50 knots in 20 seconds. Don't know if the Q400 is equipped with a speed trend vector like the CRJ, but with both being Bombardier products, it would be logical.

What would it take to write some code for say: If the Speed Trend Vector is greater than 20 knots for x number of seconds, a single chime master caution and associated message are set off/displayed.

 

[wheelsup]

What would it take to write some code for say: If the Speed Trend Vector is greater than 20 knots for x number of seconds, a single chime master caution and associated message are set off/displayed.

They already have that...it's called your eyes and your brain...

 

[Firebird2XC]

They already have that...it's called your eyes and your brain...

True. The old maxim of 'speed is life' certainly applies here.

I'm just curious as to how perceived aircraft response in a 'real stall' vs. an approach to stall scenario might have influenced the Captain's actions on 3407.

Perhaps, like windshear, an FAA mandated requirement for fully developed stall demonstrations in the simulator might help. Impressing upon pilots that bigger airplanes might induce such large stall recovery events at low levels might be enough to sufficiently promote airspeed awareness.

 

[Firebird2XC]

A good reference while reading the above thread:

[yt]lxywEE1kK6I[/yt]

Here's another.
http://204.108.4.16/d-tpp/0905/00065IL23.PDF

It's the ILS approach plate.
2300 feet for MSL altitude prior to glideslope intercept.
That's 1572, assuming the altitude is flat all the way from the airport to that point.
At slope intercept above the LM the MSL altitude is 2206.

That's 1478 AGL as well.
Given that man-made objects rise well above the terrain, and obstacle clearance at that point is even less.

Something else I JUST noticed.

When the CA calls for 'Flaps 15', the flaps indicator in this demo goes to 10 and stops. While flap extension/retraction is not immediate, there's enough of a delay before the FO retracts the flaps that they should have reached another 5 degrees. So either that indication is wrong, or that flap extension was abnormal. If you watch the speed awareness tape (I have no idea how that works in the Q400 vs. other planes) it shows penetration into shaker territory as the flap call is being made.

Since stall speed will change based on flap extension, this seeming errant flap extension may have been contributory. Note also- a Dash driver here once noted that it's normal to raise one notch of flaps should a stall recovery be required. Whether there were 10 degrees or 15, the flaps went all the way up, without a call for retraction from the pilot.

This calls to question something interesting- Speed limitations for flap extensions and profiles, etc. If the anti-ice protection system was on, this could have change stall speed protection criteria, improper flap extension/retraction could have influenced this, whether it was pilot error or mechanical malfunction.

Thoughts, everybody?

 

[EDUC8-or]

I was talking about something similar with another FO this morning. From what the NTSB released, the airspeed degradation was something like 50 knots in 20 seconds. Don't know if the Q400 is equipped with a speed trend vector like the CRJ, but with both being Bombardier products, it would be logical.

Yes, 50 knots in 20 seconds and another 50 in the 10 seconds to follow, very scary.

 

[TUCKnTRUCK]

The Q has a trend vector, but using it would be like trying to follow a bouncy ball down a flight of stairs. You end up playing with the power so much chasing it that everything else goes to crap.

 

[casey]

When the CA calls for 'Flaps 15', the flaps indicator in this demo goes to 10 and stops. While flap extension/retraction is not immediate, there's enough of a delay before the FO retracts the flaps that they should have reached another 5 degrees. So either that indication is wrong, or that flap extension was abnormal. If you watch the speed awareness tape (I have no idea how that works in the Q400 vs. other planes) it shows penetration into shaker territory as the flap call is being made.

Read the CVR transcript. He calls for flaps 15 and as the FO is setting that the shaker starts up and she deselects flaps 15 (though i dont remember if she says what she sets it to). The timeframe this happens in the flaps probably never made it past 10 before she set them back from 15.

And you can try to find excuses wherever you want about what happened but at some point you'll just have to admit it was likely a mistake and human error that caused the stall. I realize its harder to accept that this was in all likelyhood a completely avoidable accident and it was the (in)action of the pilots that caused it rather than being able to place the blame on something outside of our control (severe ice, mechanical, etc).

Perhaps the NTSB will make me eat my words, but I doubt it.

 

[Firebird2XC]

Read the CVR transcript. He calls for flaps 15 and as the FO is setting that the shaker starts up and she deselects flaps 15 (though i dont remember if she says what she sets it to). The timeframe this happens in the flaps probably never made it past 10 before she set them back from 15.

And you can try to find excuses wherever you want about what happened but at some point you'll just have to admit it was likely a mistake and human error that caused the stall. I realize its harder to accept that this was in all likelyhood a completely avoidable accident and it was the (in)action of the pilots that caused it rather than being able to place the blame on something outside of our control (severe ice, mechanical, etc).

Perhaps the NTSB will make me eat my words, but I doubt it.

I agree that human error is likely the cause, but like you, I won't know for sure until I read the final report. My leaning on this is this.. If you look at the CVR transcript:

22:16:07 — CAM: (sound similar to landing gear deployment)
22:16:11 — Shaw: Over to tower you do the same, 3407.
22:16:14 — Flight Crew Audio Panel (HOT): (sound of two double chimes)
22:16:21 — Shaw: Gear’s down.
22:16:23 — Renslow: Flaps fifteen before landing checklist.
22:16:26 — CAM: (sound similar to flap handle movement)
22:16:26 — Shaw: Uhhh.
22:16:27 — CAM: (sound similar to stick shaker lasting 6.7 seconds)
22:16:27 — HOT: (sound similar to autopilot disconnect horn repeats until end of recording)
22:16:27 — CAM: (sound of click)
22:16:31 — CAM: (sound similar to increase in engine power)
22:16:34 — Renslow: Jesus Christ.
22:16:35 — CAM: (sound similar to stick shaker lasting until end of recording)
22:16:37 — Shaw: I put the flaps up.
22:16:40 — CAM: (sound of two clicks)
22:16:42 — Renslow: (sound of grunt) (unintelligible) -ther bear.
22:16:45 — Shaw: Should the gear up?
22:16:46 — Renslow: Gear up oh (expletive).
22:16:50 — CAM: (increase in ambient noise)
22:16:51.9 — Renslow: We’re down.
22:16:51.9 — CAM: (sound of thump)
22:16:52.0 — Shaw: We’re (sound of scream)

The question in my mind is this: Did she ever even attempt to select Flaps 15?

If not, why did she retract the flaps to 10, then all the way up? There seems to be some ambiguity as to where the flaps should are and/or should be. Furthermore, her retraction of the flaps were not called for by the Captain.

I'd like to see the raw data in this case to determine if flap extension time, etc, was out of whack.

I totally concede the point of human error, but I want to beat the bushes to find out if there were other problems that were unexpected that compounded human error, or a possible flaw in the flap speed/extension profile.

While human error is at play, discovering if there's a procedure that allows people to make mistakes or compound ones they've already made would be helpful.

The human factors in this case are very interesting.

 

[SpiraMirabilis]

She selected flaps 15, when they were extending from 10' to 15' is when stuff hit the fan.

Part of the recovery procedure in the 400 from a stall is retracting flaps one notch up.

 

[casey]

The question in my mind is this: Did she ever even attempt to select Flaps 15?

If not, why did she retract the flaps to 10, then all the way up? There seems to be some ambiguity as to where the flaps should are and/or should be. Furthermore, her retraction of the flaps were not called for by the Captain.

My take on this (and it is speculation on my part) is similar to what I've seen on the line with missed approaches. Take two guys who last did a missed in the sim and put them in situation with a little stress and an unexpected need to go around and you might get something like:

PF: uhh go around!
PM: ...... Here's your flaps 9
PF: oh yea, thanks, gear up.
PM: ...positive rate, here you go.

where the calls should have gone something like

PF: go around, max thrust, flaps 9
PM: positive rate
PF: gear up, heading, low bank.

If I heard the stick shaker going off my mind is going to say "STALL!" and my reaction is going to be "max thrust, flaps 9". Perhaps she was in the same mindset and since the captain isnt saying anything, she is trying to clean up the airplane on her own in accordance with their stall recovery procedure. Just an idea.

 

[Firebird2XC]

She selected flaps 15, when they were extending from 10' to 15' is when stuff hit the fan.

Part of the recovery procedure in the 400 from a stall is retracting flaps one notch up.

Right. One notch up. She went ALL the way up.

Why?

 

[Snuggle]

Part of the recovery procedure in the 400 from a stall is retracting flaps one notch up.

Not for a flaps anywhere from 0 to 15 stall recovery. At least not at my airline.

 

[Firebird2XC]

My take on this (and it is speculation on my part) is similar to what I've seen on the line with missed approaches. Take two guys who last did a missed in the sim and put them in situation with a little stress and an unexpected need to go around and you might get something like:

PF: uhh go around!
PM: ...... Here's your flaps 9
PF: oh yea, thanks, gear up.
PM: ...positive rate, here you go.

where the calls should have gone something like

PF: go around, max thrust, flaps 9
PM: positive rate
PF: gear up, heading, low bank.

If I heard the stick shaker going off my mind is going to say "STALL!" and my reaction is going to be "max thrust, flaps 9". Perhaps she was in the same mindset and since the captain isnt saying anything, she is trying to clean up the airplane on her own in accordance with their stall recovery procedure. Just an idea.

I agree that most people will act outside standard calls when motivated by self preservation. I'm just trying to understand the flap calls/extension procedures, as I think there were some very odd things going on there.

Example- you're an Embraer pilot. In a stall recovery, max thrust, flaps 9 is the correct configuration, etc. If what Spira said above is true, then whether or not she was at flaps 10 or 15 matters, and whether or not she should have stayed at flaps 10, gone to flaps 5, or flaps 0 is in question.

One notch or not, why'd she go ALL the way up? Were the flaps still that far retracted?

I'd just really like to see the whole set of data, etc, as I see a crew's coordination melting down and no clear ruling on where the flaps are and where they should be.

The 'human factors' issue of what Renslow thought might happen if he allowed the pusher to activate or just lowered the nose is of interest, too.

I'm not saying they didn't do what they did- I'm asking why exactly they chose what they did, or was it just all-out panic? If there was a reasoning behind it and they were taught to respond a certain way, maybe that logic or procedure is flawed.

I think we all have a vested interest in asking the complicated questions and making sure nothing gets swept under the rug.

 

[derg]

Airspeed and Situational Awareness are two things that keep blasting like a megaphone into my head.

Plus, I wonder if Colgan teaches an SRT/"Standard Recovery Technique"?

Here are a couple of snippets from a major airline's training manual when it comes to stalls:

Do not retract flaps during the recovery. Retracting the flaps from the landing
position, especially when near the ground, causes an altitude loss during the
recovery.

Ground Contact Not a Factor

At the first indication of stall (buffet or stick shaker) smoothly apply maximum
thrust, smoothly decrease the pitch attitude to approximately 5° above the horizon
and level the wings. As the engines accelerate, counteract the nose up pitch
tendency with positive forward control column pressure and nose down trim. (At
altitudes above 20,000 feet, pitch attitudes of less than 5° may be necessary to
achieve acceptable acceleration.)

Accelerate to maneuvering speed and stop the rate of descent. Correct back to the
target altitude.

Ground Contact a Factor

At the first indication of stall (buffet or stick-shaker) smoothly advance the thrust
levers to maximum thrust and adjust the pitch attitude as necessary to avoid the
ground. Simultaneously level the wings. Control pitch as smoothly as possible. As
the engines accelerate the airplane nose will pitch up. To assist in pitch control,
add more nose down trim as the thrust increases. Avoid abrupt control inputs that
may induce a secondary stall. Use intermittent stick shaker as the upper limit for
pitch attitude for recovery when ground contact is a factor.

When ground contact is no longer a factor, continue to adjust pitch as required to
maintain level flight or a slight climb while accelerating to maneuvering speed for
the existing flap position.

Recovery from a Fully Developed Stall


An airplane may be stalled in any attitude (nose high, nose low, high angle of
bank) or any airspeed (turning, accelerated stall). It is not always intuitively
obvious that the airplane is stalled.

An airplane stall is characterized by any one (or a combination) of the following
conditions:

• buffeting, which could be heavy
• lack of pitch authority
• lack of roll control
• inability to arrest descent rate.

These conditions are usually accompanied by a continuous stall warning. A stall
must not be confused with the stall warning that alerts the pilot to an approaching
stall. Recovery from an approach to a stall is not the same as recovery from an
actual stall. An approach to a stall is a controlled flight maneuver; a stall is an
out-of-control, but recoverable, condition.

Note: Anytime the airplane enters a fully developed stall, the autopilot and
autothrottle should be disconnected.

To recover from a stall, angle of attack must be reduced below the stalling angle.
Nose down pitch control must be applied and maintained until the wings are
unstalled. Application of forward control column (as much as full forward may be
required) and the use of some nose-down stabilizer trim should provide sufficient
elevator control to produce a nose-down pitch rate. It may be difficult to know
how much stabilizer trim to use, and care must be taken to avoid using too much
trim. Pilots should not fly the airplane using stabilizer trim, and should stop
trimming nose down when they feel the g force on the airplane lessen or the
required elevator force lessen.

Under certain conditions, on airplanes with underwing-mounted engines, it may
be necessary to reduce thrust in order to prevent the angle of attack from
continuing to increase. Once the wing is unstalled, upset recovery actions may be
taken and thrust reapplied as necessary.

If normal pitch control inputs do not stop an increasing pitch rate in a nose high
situation, rolling the airplane to a bank angle that starts the nose down may be
effective. Bank angles of about 45°, up to a maximum of 60°, could be needed.
Normal roll controls - up to full deflection of ailerons and spoilers - may be used.
Unloading the wing by maintaining continuous nose-down elevator pressure
keeps the wing angle of attack as low as possible, making the normal roll controls
as effective as possible.

Finally, if normal pitch control then roll control is ineffective, careful rudder input
in the direction of the desired roll may be required to initiate a rolling maneuver
recovery.

WARNING: Only a small amount of rudder is needed. Too much rudder
applied too quickly or held too long may result in loss of lateral
and directional control or structural failure.

 

[jtrain609]

It didn't matter how far the flaps went up, with the yoke in the captains lap there was no chance of recovering from the stalled condition.

How much power can those engines develop? Like 16,000 HP or something like that? By lowering the nose, keeping the wings level and dumping all the flaps I'm willing to bet that thing will claw its way out of a stall in short order with that much power being applied.

Look at that video of the FDR data again. The captain was putting in the proper control inputs to get the wings leveled again, but they were obviously doing nothing because the wing (and thus the ailerons) were stalled. The wing remained stalled because of how far he jacked the yoke back. Again, check it out, those inputs with the yoke neutral or forward and they fly out of this. Those inputs with the yoke back and they crash and burn.

 

[Firebird2XC]

It didn't matter how far the flaps went up, with the yoke in the captains lap there was no chance of recovering from the stalled condition.

How much power can those engines develop? Like 16,000 HP or something like that? By lowering the nose, keeping the wings level and dumping all the flaps I'm willing to bet that thing will claw its way out of a stall in short order with that much power being applied.

Look at that video of the FDR data again. The captain was putting in the proper control inputs to get the wings leveled again, but they were obviously doing nothing because the wing (and thus the ailerons) were stalled. The wing remained stalled because of how far he jacked the yoke back. Again, check it out, those inputs with the yoke neutral or forward and they fly out of this. Those inputs with the yoke back and they crash and burn.

You're probably right. I just thought it was worthwhile to examine this. I can't help but feel that ground proximity played some sort of role in his control inputs, and the sequence of flap extension/retraction may have played into that. Probing the mindset of both pilots is something I'm very interested in.

Maybe I'm barking up the wrong tree again, but barring pure panic, I can't help but wonder why he'd respond that way.

Occam's razor probably gets us all the explanation we need.. but still.

Truth be told, I'm looking for flaws in the process that we can use as leverage for change.
Something more specific than "humans make mistakes" would be good material.
I suppose the given "tired humans make worse mistakes" is a start, but something more specific would be helpful.

 

[Firebird2XC]

Airspeed and Situational Awareness are two things that keep blasting like a megaphone into my head.

Plus, I wonder if Colgan teaches an SRT/"Standard Recovery Technique"?

Here are a couple of snippets from a major airline's training manual when it comes to stalls:

At my carrier, we have a program for ground and simulator training regarding things such as extreme unusual attitudes, windshear/microburst encounters, etc. Perhaps that or something like it should be more standardized in the industry?

>sigh< I may have just seen too many 'House, M.D.' episodes lately, but I can't help but feel like we're all missing something.