Atlas Air 767 down/Texas

[NSEU]

Are you saying that control column deflection was in fact an FDR parameter for the accident flight?

Column deflection is an input into the B767 DFDR (as well as control wheel and rudder pedal position).There are two position transducers on the ends of the torque tube (which runs between the control columns) and, optionally, one force transducer (Ref AMM 31-31-08 and 31-31-19, 21-12-03). Also, reference Boeing Wiring Schematic 31-31-01 Page 101 Sheet 4 and Sheet 5.

Having said that, I cannot say that the DFDR records this data.

[MLHeliwrench]

would you be able to hear a jump seaters voice clearly on a CVR recording?

Do crews and /or jump seat pax go through security screening prior to boarding (I would think not)?

[niteflyr]

Do crews and /or jump seat pax go through security screening prior to boarding (I would think not)?

Absolutely 100%

[jimjim1]

it seemed improbable to push over and reach the extreme dive angle without bouncing off the top of the cockpit.





(Unusually) I am not certain that this has not been posted here earlier.

Vertical acceleration analysis I guess from the flightradar24 (or similar) data. Done a few days ago (27 Feb 2019) by Peter Lemme, satcom.guru. More than -1.5g. Quite a lot more seems likely. Presumably done by differentiating the altitude twice. Interesting discussion on Peter's page.

https://www.satcom.guru/2019/02/more-questions-raised-for-atlas-5y3591.html

https://1.bp.********.com/-lrRaWb1ZZCM/XHdwq5__CwI/AAAAAAAAFaM/XUALij5OF-EVO9bzD9_iLB3fxibPhjuigCLcBGAs/s640/Screen%2BShot%2B2019-02-27%2Bat%2B9.24.24%2BPM.png(sic)


[IMG]https://cimg3.ibsrv.net/gimg/pprune.org-vbulletin/640x368/satcomguru_91740d05c4135b16d2f13a23b9adb75d72d12ff2.png

[Chu Chu]

I'm guessing something along the lines of the 737 rudder hardovers would move the column along with the elevators? So you'd need to know control force (or have other evidence) to rule it out?

[FIRESYSOK]

I'm guessing something along the lines of the 737 rudder hardovers would move the column along with the elevators? So you'd need to know control force (or have other evidence) to rule it out?

This isn’t a Microsoft joystick.

[tdracer]

Are you saying that control column deflection was in fact an FDR parameter for the accident flight?

Short answer is I don't know for sure, but I'd be surprised if it's not.

[M2dude]

would you be able to hear a jump seaters voice clearly on a CVR recording?

Do crews and /or jump seat pax go through security screening prior to boarding (I would think not)?
Yes you would via the CVR Area Mic', and yes, crews DO have to go through security screening

[EDML]

Short answer is I don't know for sure, but I'd be surprised if it's not.

The NTSB reported that the DFDR stored 350 different parameters. That means it's using a comprehensive parameter set that will surely include each control position and TLA.

[Airbubba]

Are you saying that control column deflection was in fact an FDR parameter for the accident flight?

Short answer is I don't know for sure, but I'd be surprised if it's not.

It looks like pitch control input was a required parameter on this 1992 model B-763:

§ 121.344 (d)(1):

(d) For all turbine-engine-powered transport category airplanes (https://www.law.cornell.edu/definitions/index.php?width=840&height=800&iframe=true&def_id=ee9803083700896cd85aff74cb4f95ea&term_occur=9&term_src=Title:14:Chapter:I:Subchapter:G:Part:121:Subpart:K: 121.344) that were manufactured after October 11, 1991 -

(1) The parameters listed in paragraph (a)(1) through (a)(34) of this section must be recorded within the ranges, accuracies, resolutions, and recording intervals specified in Appendix M of this part by August 20, 2001. Parameters listed in paragraphs (a)(12) through (a)(14) each may be recorded from a single source.
____________________

§ 121.344 (a)(12):

(12) Pitch control input;



https://www.law.cornell.edu/cfr/text/14/121.344

[MLHeliwrench]

Yes you would via the CVR Area Mic', and yes, crews DO have to go through security screening

OK, wasn’t sure that cargo crews got screened.

[Old Boeing Driver]

This is a fairly old 767. Does the FDR record column movements/positions, or are the FDR readings based on the position of the stabilizer/elevator.

Meaning if the stabilizer/elevator were moving, the columns actually may not be.

Apologies if this was asked/answered.

[Airbubba]

This is a fairly old 767. Does the FDR record column movements/positions, or are the FDR readings based on the position of the stabilizer/elevator.

It should record both the pitch control input and the primary pitch control surface position according to the FAR I linked above.

[FIRESYSOK]

A thing called ‘startle factor’ is going to have played a huge role here, IMO. It doesn’t take but a bad instrument reading to precipitate an overwhelming, improper reaction.

[Intrance]

No no no I can’t believe that.

Look up the CRJ200PF crash in Sweden from 2017. PF stopped watching instruments for a bit to do a briefing, when finished he looks back to the instruments and sees a 15 degree pitch up on the PFD attitude and practically instantly pushes column forward. PNF and standby instruments had normal pitch. Quick crosscheck would have most likely prevented that crash. But we are humans and are easily suprised/startled into action.

[BobbyHowie]

Does full power on the 767 generate much pitch up. In other words if you had run out of ideas to raise the nose might you try it? Clutching at straws here......
The Turkish DC10 which crashed in France following explosive decompression used full power to near enough level off before getting shredded in a forest.

[Power]

Do crews and /or jump seat pax go through security screening prior to boarding (I would think not)?

no, not if they went through KCM entry

[GlueBall]

This is not even speculation, just a "scenario" that might fit that reported throttle and control-input reading.

Massive incapacitation of PF (bird through the windscreen, medical, other), body pitches forward onto both column and (hand) throttle levers. Nose-over forces make it impossible for the other pilot and jumpseater to clear the controls until too late (that possible last-second attempt to get the nose up).

EDIT: I see I'm not alone - but it would certainly require a "Black Swan" event.

​​​​​​Below 10,000 feet the pilots would normally be strapped in by the shoulder harnesses, making a full slump-over into the control column less likely.

[suninmyeyes]

I agree with Glueball that with a standard 5 point Boeing seat harness and in turbulence an incapacitated pilot would not be able to make any forward control input let alone the hefty sustained input required to override an autopilot and pitch down 45 degrees. My only suggestion (and this is pretty remote) would be that someone could have been adjusting his seat and in turbulence it slid forward and jammed the control column forward. I hope it does not turn out to be a nefarious act as it would increase a lot of already inappropriate security restrictions on pilots.

[dgordon42]

My only suggestion (and this is pretty remote) would be that someone could have been adjusting his seat and in turbulence it slid forward and jammed the control column forward
Jump Seat pax getting out of a 'front seat' after sampling the B767 perhaps?
(See post #327 in this thread)

Dave.

[last747fe]

​​​​​​Below 10,000 feet the pilots would normally be strapped in by the shoulder harnesses, making a full slump-over into the control column less likely.
I seem to remember that ATC offered a vector around weather when the crew said which way they wanted to go they were told to descend Expedishly to 3000 could've been hooking up straps jump seater leaning forward to watch the approach thrown onto pedestal and throttle.

[Airbubba]

Jump Seat pax getting out of a 'front seat' after sampling the B767 perhaps?
(See post #327 in this thread)

Dave.

Not likely in my experience, and why would they swap seats down low approaching a line of weather?

But, most of the other scenarios including an intentional crash don't seem likely either.

There was an infamous incident in 1994 where a crewmember tried to crash a freighter:

https://en.wikipedia.org/wiki/Federal_Express_Flight_705

[golf yankee one one]

Jump Seat pax getting out of a 'front seat' after sampling the B767 perhaps?
(See post #327 in this thread)

Dave.

This scenario sounds to me the most plausible explanation for this tragedy. There is a possible similarity to the RAF Voyager incident over Turkey, which was recovered because that occurred at high altitude.

The possibility of a deliberate act seems less likely when details are compared with Germanwings; in that case the pilot was alone in the cockpit, his colleague unable to return because of the security door, and, in retrospect, the pilot had a clear history of mental illness. In this case there is no suggestion that one pilot had left the cockpit, and in addition there was, exceptionally, a third person very close by, making this flight the least likely one during which to commit suicide. Of course we do not know if any of the three individuals had any relevant medical history, and that could tilt the balance of probabilities...

[ManaAdaSystem]

I agree with Glueball that with a standard 5 point Boeing seat harness and in turbulence an incapacitated pilot would not be able to make any forward control input let alone the hefty sustained input required to override an autopilot and pitch down 45 degrees. My only suggestion (and this is pretty remote) would be that someone could have been adjusting his seat and in turbulence it slid forward and jammed the control column forward. I hope it does not turn out to be a nefarious act as it would increase a lot of already inappropriate security restrictions on pilots.

I believe the seats on the 767 have electric adjustment.

[Airbubba]

I believe the seats on the 767 have electric adjustment.

On a freighter? I doubt it.

[Zlinguy]

I believe the seats on the 767 have electric adjustment.
I flew the 200/300 20 years ago and I'm pretty sure ours were all manually adjusted (I'm not saying that it wasn't a possible option, though..).

[tdracer]

The shoulder portions of the five point harnesses on the 767 are the inertia type - so depending on the 'rate' it's quite feasible for a pilot to 'pitch' forward into the controls. Besides, he didn't need to suddenly 'pitch' forward - slumping forward would be sufficient and is unlikely to be fast enough to trigger the inertia reels.
Airbubba - I don't know if it was basic or an option - but as I recall the 767s I did flight tests on over the last 15 years all had electric seat adjusts - and most were new build freighters.

[DaveReidUK]

On a freighter? I doubt it.

The airframe in question was a passenger aircraft until a few years ago.

[TRey]

Has any determination/guesstimation been made about the ceiling at the time and location of the accident? Or, the likelihood if they were IMC at the time of the "upset"?

[Volume]

Below 10,000 feet the pilots would normally be strapped in by the shoulder harnesses,
Not from my jumpseat experience. Maybe 30% of the pilots do actually use them.
Having some poor weather ahead with a lot of red on the radar however, I would expect most of them to strap in properly.

[Broomstick Flier]

I believe the seats on the 767 have electric adjustment.

Two of ours have electric adjustment and all the other have manual adjustment. All built originally as freighters. I guess it must be a customer option.

[sequ]

The shoulder portions of the five point harnesses on the 767 are the inertia type - so depending on the 'rate' it's quite feasible for a pilot to 'pitch' forward into the controls. Besides, he didn't need to suddenly 'pitch' forward - slumping forward would be sufficient and is unlikely to be fast enough to trigger the inertia reels.
Airbubba - I don't know if it was basic or an option - but as I recall the 767s I did flight tests on over the last 15 years all had electric seat adjusts - and most were new build freighters.

This particular 767 had manual seats.

Saludos,

SEQU

[GroundProxGuy]

Pretty sure all wind shear warnings are inhibited at that altitude.

Predictive Windshear alerts from the weather radar should still function. Airbubba is right, reactive windshear alerts from the EGPWS are inhibited above 1500 feet AGL. So if they were responding to a WS alert at that altitude, it was from the WXR.

[MarkerInbound]

no, not if they went through KCM entry

There's no KCM to the MIA ramp areas.

[MarkerInbound]

Has any determination/guesstimation been made about the ceiling at the time and location of the accident? Or, the likelihood if they were IMC at the time of the "upset"?

Galveston was running about 2000 overcast, Hobby was 3500ish BKN and IAH was just reporting cirrus behind the weather.

[flyingchanges]

Predictive Windshear alerts from the weather radar should still function. Airbubba is right, reactive windshear alerts from the EGPWS are inhibited above 1500 feet AGL. So if they were responding to a WS alert at that altitude, it was from the WXR.

Might want to check your manual on that one.

[ManaAdaSystem]

The NG starts to scan for predictive windshear below 2300 ft RA, and will issue warnings/cautilns below 1200 ft.
Reactive below 1500 ft.
Not the same aircraft, but probably similar WS system.

[Murexway]

Don't know if it's true, but I saw an article yesterday saying that Grieg Feith (former NTSB "Mud Stud") mentioned that "the autopilot was still engaged when the aircraft was on its rapid descent, meaning that the pilots were fighting the automation". As an old timer, I can't imagine sitting there below 10,000, watching the throttles go to 100% and the nose dropping to 49 degrees nose low without disconnecting the automation, throttles to idle, deploying the spoilers, and pulling for all I'm worth.

[tdracer]

Don't know if it's true, but I saw an article yesterday saying that Grieg Feith (former NTSB "Mud Stud") mentioned that "the autopilot was still engaged when the aircraft was on its rapid descent, meaning that the pilots were fighting the automation". As an old timer, I can't imagine sitting there below 10,000, watching the throttles go to 100% and the nose dropping to 49 degrees nose low without disconnecting the automation, deploying the spoilers, and pulling for all I'm worth.

"Disconnecting" the autopilot on a 767 is trivial - just push or pull with a few pounds of force and it disconnects (with an aural alert in case it's inadvertent). Similarly it takes about 2 lbs. force to move a throttle (4 lbs for both throttles) regardless of what the autothrottle is doing. It's also pretty easy to differentiate an autothrottle movement from a manual movement on the DFDR. There is a max A/T rate (six degrees/second at the throttle lever IIRC) so about ten seconds to go from idle to full forward (obviously less if the throttle is at an intermediate setting).

[FCeng84]

Don't know if it's true, but I saw an article yesterday saying that Grieg Feith (former NTSB "Mud Stud") mentioned that "the autopilot was still engaged when the aircraft was on its rapid descent, meaning that the pilots were fighting the automation". As an old timer, I can't imagine sitting there below 10,000, watching the throttles go to 100% and the nose dropping to 49 degrees nose low without disconnecting the automation, deploying the spoilers, and pulling for all I'm worth.

One of the things that may need to be looked into on this is how the 767 control system responds if the pilot pushes the column while the autopilot is engaged. More recent models recognize pilot intervention via significant controller displacement as time to disconnect the autopilot. On older models the autopilot does not immediately disengage in response to pilot input. There may have been a period of time where both Hal and Row 0 were providing significant control inputs. Learning that the autopilot remained engaged would not necessarily mean that Hal was at fault.

[Old Boeing Driver]

This event was going on for 18 seconds. I would assume that one or both pilots had hold of the yoke.

I can't imagine that either did dot push the A/P disconnect on the yoke.

Note: I am assuming there is an A/P disconnect on the yoke as on previous Boeing's.

[Airbubba]

This event was going on for 18 seconds. I would assume that one or both pilots had hold of the yoke.

I can't imagine that either did dot push the A/P disconnect on the yoke.

Note: I am assuming there is an A/P disconnect on the yoke as on previous Boeing's.

There is an autopilot disconnect button on the yoke and a disconnect bar on the glareshield Mode Control Panel on the classic B-763.

Also, the NTSB release said the throttles went to full power before the pitch over.

I also can't imagine not pulling the throttles and speedbrake lever back with the nose pointed down for whatever reason.

[Murexway]

One of the things that may need to be looked into on this is how the 767 control system responds if the pilot pushes the column while the autopilot is engaged. More recent models recognize pilot intervention via significant controller displacement as time to disconnect the autopilot. On older models the autopilot does not immediately disengage in response to pilot input. There may have been a period of time where both Hal and Row 0 were providing significant control inputs. Learning that the autopilot remained engaged would not necessarily mean that Hal was at fault.No, I'm sure they were trying everything humanly possible, exactly the way we all would have been. But if the darn automation wouldn't relinquish control, that would be a horrible feeling.

[niteflyr]

Why the speedbrake lever? I would rather build up speed and altitude until I figured out what happened.

When you"re 49 degrees pitch down and going thru redline?

[A4]

??????......er....if I’m pointing down at the ground (let alone 49 degrees down) I’m going to close the throttles, pull the speed brake and pull.....probably the only way you’re going to get any chance of gaining altitude......once you’ve arrested the descent.....

A4 (10,000+ hrs Airbus)

[Murexway]

Also read that their speed increased from their assigned 230 kts to 430 kts in the descent, and that at impact it was close to 500 kts.

[Icelanta]

??????......er....if I’m pointing down at the ground (let alone 49 degrees down) I’m going to close the throttles, pull the speed brake and pull.....probably the only way you’re going to get any chance of gaining altitude......once you’ve arrested the descent.....

A4 (10,000+ hrs Airbus)
then do not use the speedbrakes but very slowly, increase pitch and convert your energy into climb.
putting out the speedbrakes increases your load on the airframe even more, increasing the chance of an inflight breakup.

[A4]

I see that.....but accelerating towards the barbers pole (and beyond) also isn’t going to be great for structural integrity! Additionally, it’s dependent upon how much altitude you have to play with......from 20,000 it’s no issue......from 5,000....different scenario. Does the 777/787 have load factor protection like the Airbus? At least you can pull to the limit with out risking structural integrity (if the aircraft (Bus) is in Normal Law). The 767 would obviously require a more nuanced approach to handling.

A4

[Australopithecus]

From around 2000’ and 500 kts it would be pretty much a 3 G pull-up. Moot point though if something is preventing full pitch-up control inputs.

[RoyHudd]

Deliberate act possible. Examination of pilots' histories needed.

This possible cause should be evaluated in the same way as a technical issue. Methodically and objectively.

[Hotel Tango]

Deliberate act possible.

Well, in fact, almost anything is possible until we hear differently!

[Sailvi767]

then do not use the speedbrakes but very slowly, increase pitch and convert your energy into climb.
putting out the speedbrakes increases your load on the airframe even more, increasing the chance of an inflight breakup.


Can you explain this statement. How do speedbrakes increases airframe load?

[Sailvi767]

I see that.....but accelerating towards the barbers pole (and beyond) also isn’t going to be great for structural integrity! Additionally, it’s dependent upon how much altitude you have to play with......from 20,000 it’s no issue......from 5,000....different scenario. Does the 777/787 have load factor protection like the Airbus? At least you can pull to the limit with out risking structural integrity (if the aircraft (Bus) is in Normal Law). The 767 would obviously require a more nuanced approach to handling.

A4

The Airbus will limit you to two G’s in the situation you describe. The airframe is required to take nearly double that load at max certified weight. 2.5 G’s is the requirement with no damage and 3.75 before structural failure. At landing weights those limits would be even higher.

[Running Ridges]

Can you explain this statement. How do speedbrakes increases airframe load?

They move the lift distribution outboard, so if you're pulling 2g with the speedbrakes out the root bending moment will be higher compared to a 2g manouevre with a clean wing.
In any case I doubt you'd still be accelerating if you're pulling 2g+ as the drag would be significantly higher

[aterpster]

This event was going on for 18 seconds. I would assume that one or both pilots had hold of the yoke.

I can't imagine that either did dot push the A/P disconnect on the yoke.

Note: I am assuming there is an A/P disconnect on the yoke as on previous Boeing's.
Yes, there is. The 767 doesn't have "high tech" computer "augmentations." It's a straight-forward, easy airliner to fly. A/T easy to disconnect or override. Autoflight easy to turn off.

[Sailvi767]

They move the lift distribution outboard, so if you're pulling 2g with the speedbrakes out the root bending moment will be higher compared to a 2g manouevre with a clean wing.
In any case I doubt you'd still be accelerating if you're pulling 2g+ as the drag would be significantly higher

At a given AOA the lift the wing can produce is significantly reduced with the speedbrakes out which would unload the wing. If you pulled to a higher AOA to maintain the same G force it may be possible you would shift the bending moment but I doubt it has much impact.

[Machinbird]

Written from a fighter pilot viewpoint. When applying to a transport, there are going to be additional factors to consider such as control hinge moments, wing bending moment, and control system design. (For Airbus FBW, all such decisions have been already made for you.)

With the nose buried downward, and with concern about hitting the ground, you would like to achieve a minimum radius turn, but without tearing the aircraft apart.
The minimum radius turn is generally achieved where max AOA and max allowable G meet on the performance curves.

You can consider using up some of your safety margin when planning what G level to attain. Then look up the performance curves and determine a cornering velocity based upon your most probable gross weight.
Below cornering velocity, you can slightly improve performance by accelerating slightly and pulling into buffet. Above cornering velocity, you must slow down to improve turn performance and meanwhile avoid going beyond your max acceptable G (without a G meter).

Knowing your cornering speed is strictly emergency knowledge and would be used solely to ensure you are not completely out of the ballpark in your pull out efforts.
No doubt, the test pilots and aerodynamics guys can flesh this out better. As I recall, the F-4 cornering velocity was 420 knots. You definitely will not want to be anywhere near that speed when you pull out in your transport.

[Sailvi767]

Transport aircraft like the 767 have a reasonably light wing loading relative to fighters. I suspect that at the 230 knots assigned they could generate 3.75G’s. Since there is no where to read G force unless they had recent aerobatic flying that would be very subjective.

[EDML]

At a given AOA the lift the wing can produce is significantly reduced with the speedbrakes out which would unload the wing. If you pulled to a higher AOA to maintain the same G force it may be possible you would shift the bending moment but I doubt it has much impact.


Correct. Less lift less load. That is how gust load relief via the ailerons works on the bus.

The interesting question would be, how the speed brakes influence the cornering radius pulling out of the dive. However, even here they won't make things worse I would guess.

[Machinbird]

Since there is no where to read G force unless they had recent aerobatic flying that would be very subjective.
So true, but not a problem IF you are below your cornering velocity. If well in excess, I would think that pulling throttles to idle and putting out speedbrakes would make a substantial improvement in pullout altitude, even if at the expense of maximum g load.
The most valuable seconds in a dive recovery are the first few seconds. Delay in getting max g on will cost you altitude.
We haven't even discussed what you do when you realize that you were successful and will miss the ground.;)

[FIRESYSOK]

Absolute rubbish. In a nose-low unusual attitude, yes, you should pull the Gs smoothly, and earlier than later (before the speed has a chance to increase further), but never use speed brakes. A G load of 4 (which is well beyond certification limits) will increase stall speed by a factor of two. Suddenly you have a stall speed of 350+ knots and you guys want to destroy the laminar flow further? I’m not sure what to think of the comments here. I’ve never ever heard of this. My opinion is deploying the spoilers in a high-G dive recovery is going to put the airplane in the ground twice as efficiently. But go ahead and think it’ll help.

[henra]

I’m not sure what to think of the comments here. I’ve never ever heard of this. My opinion is deploying the spoilers in a high-G dive recovery is going to put the airplane in the ground twice as efficiently. But go ahead and think it’ll help.
Have you read @Machinbirds post?
He perfectly explained why in certain circumstances temporarily activating the spoilers can make sense.
They make sense when you are above cornering speed, i.e. when the wing could aerodynamically produce more lift than the structure can take. In that case the radius of the pull up maneuver will increase (or you shed the wings). In that case slowing down will reduce turn radius (yes the spoilers will reduce lift somewhat so that would slightly increase the radius. There is a balance Probably some 20-30 kts above cornering speed you get a net gain with the brakes applied.

[gums]

Salute!

@ FIRE
I think what 'bird and I and maybe Henra are saying is there is an optimum speed and gee for best turn radius. The AoA is surprisingly lower than stall AoA for many planes.
Your gee toward the center of the circle is what determines the ft/sec from your flight path a millisecond ago. So stall AoA is great for max lift, maybe gliding, but norrmally comes about at a slower CAS than best turn radius. It's not an intuitive relationship, but we have seen it in many of our lites For example, the F-16 gets best turn at about 360 kt CAS and 9 gees - turn radius of about 1400 feet!!!! . You can get up to 25 degrees AoA slower than that but not at 9 gees.

Without a gee limiter, I would pull for all its worth in a big plane and worry abut ripping the wings off later. Most charts I have seen show overspeed and flutter to be a bigger problem than over gee.

GUms opines...

[Murexway]

Absolute rubbish. In a nose-low unusual attitude, yes, you should pull the Gs smoothly, and earlier than later (before the speed has a chance to increase further), but never use speed brakes. A G load of 4 (which is well beyond certification limits) will increase stall speed by a factor of two. Suddenly you have a stall speed of 350+ knots and you guys want to destroy the laminar flow further? I’m not sure what to think of the comments here. I’ve never ever heard of this. My opinion is deploying the spoilers in a high-G dive recovery is going to put the airplane in the ground twice as efficiently. But go ahead and think it’ll help.I'm sure you're right when you have acft control, as in a high-angle release. But when your nose is suddenly pointed directly at the ground below 10 and the throttles are stuck at max, popping the boards immediately rather than later may buy you a precious few seconds (and feet) to try and correct the situation. If you can get the nose started up, you can always ease the boards in. But as this accident proved, if nothing else is working, you haven't got much to lose by trying anything and everything.

[Murexway]

Salute!

@ FIRE
I think what 'bird and I and maybe Henra are saying is there is an optimum speed and gee for best turn radius. The AoA is surprisingly lower than stall AoA for many planes.
Your gee toward the center of the circle is what determines the ft/sec from your flight path a millisecond ago. So stall AoA is great for max lift, maybe gliding, but norrmally comes about at a slower CAS than best turn radius. It's not an intuitive relationship, but we have seen it in many of our lites For example, the F-16 gets best turn at about 360 kt CAS and 9 gees - turn radius of about 1400 feet!!!! . You can get up to 25 degrees AoA slower than that but not at 9 gees.

Without a gee limiter, I would pull for all its worth in a big plane and worry abut ripping the wings off later. Most charts I have seen show overspeed and flutter to be a bigger problem than over gee.

GUms opines...Never heard of "Gees."

[Airbubba]

But when your nose is suddenly pointed directly at the ground below 10 and the throttles are stuck at max, popping the boards immediately rather than later may buy you a precious few seconds (and feet) to try and correct the situation.

Which begs the question, were the throttles still at max all the way down and if so why?

And was the autopilot really on until impact?

[gums]

Salute!

Sorry Murex, about the "gees", I am used to phonetically spelling out the force of gravity for those foreign to the aviation world.
'course, that's what many women I have entertained haved exclaimed after a few minutes!

Gums sends...

[Old Boeing Driver]

Which begs the question, were the throttles still at max all the way down and if so why?

And was the autopilot really on until impact?

I have the same questions. Since this event was over a span of 18 seconds, I would think that they were man handling the controls and forgetting about the automation running the show.

Just the yanking on the yoke would probably disconnected the A/P and surely someone saw the thrust levers full forward. Especially since they were accelerating from 240KTS to 425KTS.

RGDS

[Zlinguy]

I still believe that the most effective way to prevent/minimize large speed exceedances would be the extension of the landing gear at the onset ( although the ability for a crewmember to physically reach the gear lever and the time required for extension outside of the design parameters would have probably negated any benefit in this situation...).

If I recall correctly, the 767 doesn't physically limit gear extension above VLo ( I think the override button is there to allow retraction when air/gnd logic would otherwise prevent it, but, extension would be dependent on aerodynamic loads), and, obviously it is not without risks, such as loss of gear doors and attendant hydraulic failures, and an unknown pitching moment at high speed as the gear extends. The gear itself is robust and unlikely to be damaged and more importantly offers a source of "clean" drag without the attendant concerns of varying lift distribution via speedbrakes, etc.,etc...

[Educated Airman]

I am still thinking about the missing bolt on the horizontal stab actuator theory. If this happened the stab would most likely have moved beyond the design limits and could possibly have caused contact with the elevator control cables resulting in an overload of one of the many pulley quadrants. This would cause the cables to go slack and the elevators to assume and retain any arbitrary position. If there is a control column position channel on the FDR and it does not agree with the elevator position, a probable cause could be stab movement beyond design limits. Once the horizontal stab stalled out due to pitch momentum, the excessive nose down AOA could have a wing area effect on the stab (Due to center of lift and hinge location) to blow it into a nose up position resulting in the final pull up.

[runner1021]

WSJ reporting that pilot error is now suspected as cause of crash.

[FCeng84]

I still believe that the most effective way to prevent/minimize large speed exceedances would be the extension of the landing gear at the onset ( although the ability for a crewmember to physically reach the gear lever and the time required for extension outside of the design parameters would have probably negated any benefit in this situation...).

If I recall correctly, the 767 doesn't physically limit gear extension above VLo ( I think the override button is there to allow retraction when air/gnd logic would otherwise prevent it, but, extension would be dependent on aerodynamic loads), and, obviously it is not without risks, such as loss of gear doors and attendant hydraulic failures, and an unknown pitching moment at high speed as the gear extends. The gear itself is robust and unlikely to be damaged and more importantly offers a source of "clean" drag without the attendant concerns of varying lift distribution via speedbrakes, etc.,etc...

Seems to me that when pointed down hill at such an angle with the ground rushing at you, generating lift (normal load factor) to turn that flight path vector back up above the horizon is the most critical part of recovering. Adding drag that does not come along with additional lift probably does not help much. Anything that limits achieved "gees" is reducing the chances of escape rather than enhancing them.

[Vessbot]

Salute!

Sorry Murex, about the "gees", I am used to phonetically spelling out the force of gravity for those foreign to the aviation world.
'course, that's what many women I have entertained haved exclaimed after a few minutes!

Gums sends...
There are 2 completely different double entendres at play here, I'm not sure which one you meant but my hat's off if it is both!:}

[Vessbot]

Seems to me that when pointed down hill at such an angle with the ground rushing at you, generating lift (normal load factor) to turn that flight path vector back up above the horizon is the most critical part of recovering. Adding drag that does not come along with additional lift probably does not help much. Anything that limits achieved "gees" is reducing the chances of escape rather than enhancing them.

A few thoughts.

First, if the speed is below Va and therefore lift is AOA-limited, then you're right and strictly speaking, the boards will further limit lift and therefore reduce the pullout-radius-shrinking normal force. However, without a AOA meter and a confident pilot willing to pull all the way to the limit, it is extremely unlikely that the full AOA lift potential will be used to begin with. In that case, the speed brakes' effect on reducing pullout-radius-growing airspeed, will be overall beneficial.

Second, if the speed is above Va and therefore lift is G-limited, then the speedbrakes will have no negative effect and only positive. Third, with the nose significantly below the horizon, if airspeed is below Va, it will be above it very soon anyway.

[Airbubba]

WSJ reporting that pilot error is now suspected as cause of crash.

Here's the article from Andy Pasztor:

Pilot Error Suspected in Fatal Atlas Air Cargo CrashInvestigators exploring likelihood that crew accidentally increased thrust on approach to Houston airport, sources say


https://cimg9.ibsrv.net/gimg/pprune.org-vbulletin/1322x1200/pasztor_1_36ab859adc8e2b30318791a937d28d1f042e5ec8.jpg
https://cimg1.ibsrv.net/gimg/pprune.org-vbulletin/1302x1715/pasztor_2_c8a121b1f6fdfc5916c1e26eba7e8270a89f4558.jpg
https://cimg0.ibsrv.net/gimg/pprune.org-vbulletin/1231x368/pasztor_3_462e2d227b4c57760687dfbaf3ef37951401f77c.jpg

[log0008]

Turbulence causing a pilots hand to push the engines to t/o power and then a massive pitch down control input sounds like the most strange and extraordinary cause of a crash I've read, has to more to it surely?

[runner1021]

"The seemingly disoriented crew failed to regain control—despite commands to pull up from the jet’s high-speed dive—and the wide-body plane plowed into a marshy area".

It would appear that this information came from the CVR. Hard to believe it would take more than a few seconds to get the throttles back to idle and initiate a recovery. Negative G causing sustained forward pressure on the yoke???

[extreme P]

Turbulence causing a pilots hand to push the engines to t/o power and then a massive pitch down control input sounds like the most strange and extraordinary cause of a crash I've read, has to more to it surely?

Agreed. The turbulence would have had to further jostle a pilot's thumb to disconnect the A/T.

I hope they can recover some useful info from the CVR.

[Murexway]

Which begs the question, were the throttles still at max all the way down and if so why?

And was the autopilot really on until impact?Don't know for sure, but the quote about fighting the automation was from Greig Feith, former NTSB guy (https://aeronauticsonline.com/ntsb-releases-new-details-about-atlas-air-767-crash/) He must still have contacts on the board, I would think. Supposedly, according to the article, they managed to reduce the pitch to only 20 degrees nose low, but it sure doesn't look like that on the video clip that was posted online showing it just prior to impact. The story originally said that the speed was close to 500 kts at impact, but that part seems to have been edited out of the story now. Sure hope it wasn't intentional.

[Lonewolf_50]

This is very hard to believe, Airbubba. Not taking you to task, but very skeptical of Andy the twitter entity.
49 degrees nose down is a non trivial departure from anything remotely resembling normal corrective action in a transport aircraft. It's the kind of pitch angle down that you'd see in an aerobatic display.
While I will await further amplifying information, massive nose down pitch of that magnitude is an over-control that is way out of limits for the flight regime the crew were in: getting themselves set up for a standard approach into an airport.
If my hand bumps the control, engine or otherwise, it takes a certain amount of time of "nobody is near the yoke" for that to go from normal flying to 49 degrees nose down and for that to not be immediately corrected.
We'll see what further NTSB releases reveal.

[Educated Airman]

Having experienced a severe windshear that was some what anticipated in cavu on approach to Sedona Arizona, I can tell you that it is instinctive to push the yoke and throttles when the nose drops un- commanded. For my experience, that was the right thing to do even if I was only 500 AGL. The question with Atlas is what caused the un-commanded nose drop.

[fdr]

I am still thinking about the missing bolt on the horizontal stab actuator theory. If this happened the stab would most likely have moved beyond the design limits and could possibly have caused contact with the elevator control cables resulting in an overload of one of the many pulley quadrants. This would cause the cables to go slack and the elevators to assume and retain any arbitrary position. If there is a control column position channel on the FDR and it does not agree with the elevator position, a probable cause could be stab movement beyond design limits. Once the horizontal stab stalled out due to pitch momentum, the excessive nose down AOA could have a wing area effect on the stab (Due to center of lift and hinge location) to blow it into a nose up position resulting in the final pull up.

The 767 is a conventional geometry aircraft: The stab provides a negative lift force to balance the nose down pitching moment of the wing. If the actuator becomes fully free of constraint, the aerodynamic load on the stab will drive it to stabilizer leading edge down, which increases the negative lift, and results in a pitch up. The normal force from the section is forward of the remaining hinge line, and will result in the stabiliser increasing LE down (pitch up).This was not the case for an aircraft that pitched nose down in wings level flight. (Sorry...)

[if you get up close and personal with the stab, you will see that it is a cambered section, with the camber towards the bottom, not the top]

[Eric Janson]

What about inadvertent selection of TOGA mode? - hand resting on the thrust levers then turbulence causes one of the two TOGA switches to be pushed.

Would make more sense than the thrust levers being manually pushed forward.

[Murexway]

Turbulence causing a pilots hand to push the engines to t/o power and then a massive pitch down control input sounds like the most strange and extraordinary cause of a crash I've read, has to more to it surely?I agree. Like everybody else, I've been in some pretty heavy turbulence and I don't remember that it ever caused me to jam the throttles to the stops. And the acft supposedly was on the A/P, so it would had to have disconnected in order to get a pitch up from the increased thrust, in any case. Even if all that had happened, I don't see them pushing massively on the yokes to correct was only a few hundred foot climb. Sounds more like a Keystone Cops movie.

[TowerDog]

What about inadvertent selection of TOGA mode? - hand resting on the thrust levers than turbulence causes one of the two TOGA switches to be pushed.

Would make more sense than the thrust levers being manually pushed forward.

No,
The 767 is an easy, docile Jet, no risk of inadvertent selection of TOGA mode, none whatsoever unless you are totally incompetent and I doubt these guys were. (Unless one of them was suicidal, then anything and everything is on the table)

[Murexway]

What about inadvertent selection of TOGA mode? - hand resting on the thrust levers than turbulence causes one of the two TOGA switches to be pushed.

Would make more sense than the thrust levers being manually pushed forward.I don't buy this hypothesis reported by the Wall Street Journal writer. I really doubt that they were hand flying while driving around in some chop. And if TOGA was selected accidentally for some reason, they would have simply pulled the throttles back. It wouldn't have caused them to have go ape....

[Educated Airman]

The hinge line is aft of the center of lift, but forward of the planform area center. Once the stab stalls out, the air force on the top of the surface will push the stab back to a stab nose down position.

[Murexway]

There are 2 completely different double entendres at play here, I'm not sure which one you meant but my hat's off if it is both!:}Hilarious :)

[Bleve]

Reminds me of a scenario I saw in a 767 Sim a few years ago. A standard Go-Around from about 200ft AGL with autopilot engaged. Missed approach altitude was 3000ft. Everything appeared normal, we had altitude capture and the nose lowered as we approached 3000ft. BUT the nose kept dropping and dropping. At about 20 degrees nose down I disconnected the autopilot and manually flew a nose low upset recovery. It turns out the automatics had not captured 3000ft but a much lower altitude (we suspected about 1000ft). So the mechanics of what happen was as follows:
- After pressing the Go-Around switches the engines went to Go-Around thrust (normal).
- The Autopilot raises the nose to climb away (normal).
- The Autopilot trims the STAB nose down to counteract the large nose-up pitching moment due to Go-Around thrust (normal).
- The automatics captured ~1000ft instead of 3000ft (NOT normal).
- As we were climbing (or had climbed) through 1000ft, the autopilot commanded a nose down attitude to regain 1000ft.
- With the nose attitude decreasing and Go-Around thrust set, the speed started increasing rapidly.
- The autothrottle then set idle thrust in attempt to control the speed.
- Now here's the kicker: The autopilot had trimmed nose down to counteract Go-Around thrust. With the thrust rapidly reduced to idle, the now unbalanced pitching moments caused the pitch attitude to rapidly drop through 20 degrees nose down. The autopilot did not react quick enough to counteract the rapid nose down pitch change.

That might be a factor in this incident. Go-Around thrust activated. Nose down trim applied to counteract the nose-up pitching moment caused by Go-Around thrust. Nose lowered to regain the vertical path, but speed rapidly increases due Go-Around thrust. Thrust rapidly reduced to control speed. Pitching moments are now unbalanced and nose rapidly drops to a very nose low attitude. Startle effect and confusion overwhelms the crew.

[runner1021]

18 seconds from the initial loss of control to impact. Even considering the startle factor and an assumedly fatigued crew, something doesn't add up.

[stilton]

What about inadvertent selection of TOGA mode? - hand resting on the thrust levers then turbulence causes one of the two TOGA switches to be pushed.

Would make more sense than the thrust levers being manually pushed forward.


I don’t believe they had flaps or slats extended in which case TOGA is not even armed


i don’t buy the WSJ theory for a moment, makes absolutely no sense



In the picture I saw the left side of the horizontal stabilizer looked damaged, perhaps they hit a drone

[jugofpropwash]

I find it very hard to believe that someone's arm could have been "jostled" to that extent - but what if one of the pilots (or the jump seat passenger) was out of their seat, and sudden turbulence knocked them off their feet and into the controls? It seems that an individual stumbling into the controls would be more likely to knock levers or other controls out of position, and if the individual landed in the PF's lap, maybe by the time that was sorted, the situation might have been unrecoverable?

[Raffles S.A.]

I guess they should put some pilots in a sim and try out a bunch of scenarios.

[AviatorDave]

Here's the article from Andy Pasztor:




https://cimg9.ibsrv.net/gimg/pprune.org-vbulletin/1322x1200/pasztor_1_36ab859adc8e2b30318791a937d28d1f042e5ec8.jpg
https://cimg1.ibsrv.net/gimg/pprune.org-vbulletin/1302x1715/pasztor_2_c8a121b1f6fdfc5916c1e26eba7e8270a89f4558.jpg
https://cimg0.ibsrv.net/gimg/pprune.org-vbulletin/1231x368/pasztor_3_462e2d227b4c57760687dfbaf3ef37951401f77c.jpg

I don’t buy this story which sounds like a very bad student pilot screwup. These guys were no rookies. Both CA and FO had total time well in the thousands, and while the FO appeared to be relatively new on type with 520 hours, the CA had around 1250 hours.
A technically well functioning 767 is not known as a problematic or difficult to handle aircraft.
I have been training rookie student pilots to deal with unexpected power surges and losses in light pistons, but none of them would have augured in while trying to recover from a sudden pitch-up due to unexpectedly firewalling the throttle,

[log0008]

I agree. Like everybody else, I've been in some pretty heavy turbulence and I don't remember that it ever caused me to jam the throttles to the stops. And the acft supposedly was on the A/P, so it would had to have disconnected in order to get a pitch up from the increased thrust, in any case. Even if all that had happened, I don't see them pushing massively on the yokes to correct was only a few hundred foot climb. Sounds more like a Keystone Cops movie.

So much doesn't make sence, I'm still a flight instructor so can't comment on jets but i just can't see it happening. Sure there have been plenty of times where turbulence, has lead to a small un intended power movements but these are always felt and able to be reacted to quickly and without any effect on the flight, and thats in a piston aircraft, i'm sure a jet responds even more slowly. Also again not being an airline pilot but would a pilots hand even be anywhere near the throttles with the aircraft on autopilot at 7000ft or so?

I don't have the data but to me I just can't see how this situation could be possible, I'm thinking maybe someone has been misquoted and the auto throttles somehow activated t/o power. That or someone has the order stuffed, a nose pitch down to 49 degrees now rapid enough that could cause an unintentional fwd movement of the throttles.

[Mr Angry from Purley]

18 seconds from the initial loss of control to impact. Even considering the startle factor and an assumedly fatigued crew, something doesn't add up.
Flight was daylight, should we not wait until we see the schedule before making "assumed" statements of fatigue

[donotdespisethesnake]

"according to people familiar with the details"

These sort of unattributed leaks should be taken with a bag of salt. Happens all too often that journalists looking for a good story are willing to listen to anyone with gossip to share.

However, there have been several cases of somewhat inexplicable, and inexcusable handling errors over the past few years.

[CaptainMongo]

I would be surprised if Pasztor is far off the mark. He is one of the if not the most senior aviation writer at the Wall Street Journal. Historically he hasn’t thinly sourced his articles.

[AviatorDave]

I would be surprised if Pasztor is far off the mark. He is one of the if not the most senior aviation writer at the Wall Street Journal. Historically he hasn’t thinly sourced his articles.

Still, I would want to see the CVR transcript first. It‘s hard to believe that a 11.000 hour captain cannot recover from a power-surge related pitch up at 6000ft in daylight if that‘s the only condition the crew would have had to deal with.
But we‘ll see.

[GlueBall]

Educated Airman says: "The question with Atlas is what caused the un-commanded nose drop."

. . . Could you conceptualize that it may not have been uncommanded?

[Ian W]

I don’t believe they had flaps or slats extended in which case TOGA is not even armed


i don’t buy the WSJ theory for a moment, makes absolutely no sense



In the picture I saw the left side of the horizontal stabilizer looked damaged, perhaps they hit a drone

I must admit that was my initial thought. As it fits with the claimed background voice in the last response to ATC. Pull up to avoid a head on with a large drone then damage from the impact and LOC

[henra]

It‘s hard to believe that a 11.000 hour captain cannot recover from a power-surge related pitch up at 6000ft in daylight if that‘s the only condition the crew would have had to deal with.

That is if they immediately identified the thrust surge and the reason for it. If not, in IMC it would drastically amplify the Nose Up feeling. The somatogravic illusion of a combination of a sudden and massive thrust increase (from almost flight idle to TOGA?) combined with the resulting real pitch up moment will be massive. Combined with a startle effect while their minds were focussed on going around weather this could have led to temporary confusion. In the Rostov Crash the Crew even deliberately set TOGA and still drove it manually into the ground, despite a planned thrust change. In IMC surprise plus illusion is a serious adversary...
And 10 - 15s (when they came out of the clouds they apparently started the recovery but obviously too late) is not that long, if we look at LoC accidents of the last 10 Years where Crews were confused for up to minutes when outside reference was missing.
So, unfortunately, I would not completely rule out such a scenario.

[henra]

I must admit that was my initial thought. As it fits with the claimed background voice in the last response to ATC. Pull up to avoid a head on with a large drone then damage from the impact and LOC
Guys, you can be sure that by now NTSB will know if the ND attitude was due to Control input or caused by aerodynamic disturbance due to damage (from a drone/bird/whatever).
It is honorable but seriously clutching at straws by now. I just hope (and also firmly believe) that it was purely accidental and not deliberate.

[ironbutt57]

Looks like somebody just got over-excited with the elevator in the turbulence

[Murexway]

Only other thing I can think of, is that they ran under the bottom of a cell that Approach wasn't painting for some reason. Could have been a microburst?.... something similar to Delta at DFW in 1985. They can form fast. I was headed from DAL to MRY in a Falcon Jet that afternoon and was vectored right over the top of DFW westbound at 4,000. It was a nice sunny VFR afternoon with scattered puffy white clouds just above me and it was an unusual sight to be directly overhead DFW. When I got to MRY the line guys were all talking about Delta crashing in a thunderstorm at DFW. I couldn't believe that they had the right airport, since I'd passed overhead about 45 minutes prior to the crash. Of course, that was in August, and Delta was on the ILS.... a lot lower than Atlas.

[RatherBeFlying]

High Thrust
-> Nose Down Trim from Autopilot or MCAS
-> High Speed
-> Insufficient Elevator Authority to pull out of dive + pulling back thrust takes away positive pitch moment when most needed + air load can stall trim.

Seems to be a common factor with Rostov, Houston, Lion, Ethiopian?

[runner1021]

Flight was daylight, should we not wait until we see the schedule before making "assumed" statements of fatigue
If you are familiar with ACMI cargo operations, one can assume fatigue most likely played a factor. It's a fact of life in the industry.

[cappt]

Very hard to believe pilots of this freighter sat and watched their jet plunge nearly straight down while not taking corrective action. When descending 2.5 degrees is very noticeable, 5 deg. is a very steep descent, 10 degrees starts filling the ADI with all brown, you only see this in the SIM world. Any airliner heading down at 49 degrees is out of control and will not likely recover without 20K feet and very careful inputs.
When I heard the ATC "pull" I immediately thought someone seen something out front.

[B2N2]

Let’s not go overboard with the “turbulence”.
This is a big aircraft, it will take a lot for it to be ‘extreme’ for a turbulence induced loss of control.

[Havingwings4ever]

Let’s not go overboard with the “turbulence”.
This is a big aircraft, it will take a lot for it to be ‘extreme’ for a turbulence induced loss of control.
Indeed. Having experienced severe turb in this ac ( eg. 700 ft drop in seconds and 51 degr bank angle, of course near Gander, jetstream turned 180 there), having to use max thrust to stay on the ILS G/S into MIA in a downdraft(microburst?) etc. gave me a lot of confidence in this ac. I am sure fellow airmen have seen worse...

[TheHardWay]

In the picture I saw the left side of the horizontal stabilizer looked damaged, perhaps they hit a drone


A picture of it after it followed the rest of the aircraft into the ground at high speed? What piece of wreckage have we seen a picture of that wasn't damaged?
Maybe I misunderstand your statement.

[FlexibleResponse]

When max thrust is applied at low level in an airliner at landing weight, the human vestibular balance system will perceive that longitudinal acceleration as a pitch-up, especially in the absence of strong visual cues (ie, good VMC flight conditions).

The instinctive pilot reaction to longitudinal acceleration is to push the control column forward to maintain what he perceives as a necessary control correction to maintain a constant aircraft pitch attitude. This reaction is especially strong if the pilot has been flying with visual references and then subtly loses such visual reference and enters instrument flight conditions but still believes that he is in visual flight.

The secondary effect of thrust application is also an upwards pitching moment on the aircraft and this reinforces the need for the pilot to instinctively push the control column forward (and trim) with any thrust increase.

This principal of fooling the human vestibular balance system is employed in 6 axis Flight Simulators to imitate the motions of aircraft flight.

Perception of an increasing pitch up by the pilot flying in response to the increase in thrust may have triggered him to push the control column forward until such time as he recognized the actual very steep pitch attitude on re-attaining strong visual cues on exiting the cloud base in the subsequent dive.

This principle might be a factor in the subject accident.

[Lonewolf_50]

@Flexibile:
Do you think it fair to presume that the pilot in this aircraft didn't have an instrument scan going? (Called by some a 'crosscheck')
If you are on instruments, then you fly with the AI/AHRS/HSI as your primary reference. You get trained to do that when you get an instrument rating.
(Granted, any of us can have a scan breakdown on a bad day).
So while what you say has been shown to be true, the remedy for that issue is also known:
establish a scan and use it to maintain attitude, performance, etc.

[FlexibleResponse]

What appears to have been a perfectly serviceable B767 flown by a competent and experienced crew crashed after entering a steep dive.

The reason for the accident is unknown at this stage.

I presume nothing.

[TRey]

When max thrust is applied at low level in an airliner at landing weight, the human vestibular balance system will perceive that longitudinal acceleration as a pitch-up, especially in the absence of strong visual cues (ie, good VMC flight conditions).

The instinctive pilot reaction to longitudinal acceleration is to push the control column forward to maintain what he perceives as a necessary control correction to maintain a constant aircraft pitch attitude. This reaction is especially strong if the pilot has been flying with visual references and then subtly loses such visual reference and enters instrument flight conditions but still believes that he is in visual flight.

The secondary effect of thrust application is also an upwards pitching moment on the aircraft and this reinforces the need for the pilot to instinctively push the control column forward (and trim) with any thrust increase.

This principal of fooling the human vestibular balance system is employed in 6 axis Flight Simulators to imitate the motions of aircraft flight.

Perception of an increasing pitch up by the pilot flying in response to the increase in thrust may have triggered him to push the control column forward until such time as he recognized the actual very steep pitch attitude on re-attaining strong visual cues on exiting the cloud base in the subsequent dive.

This principle might be a factor in the subject accident.

This very thing! You are not alone in this theory...

Having flown the 767 for sometime now and as others have alluded to, MAX thrust at that weight/altitude is no joke when it comes to acceleration. So +1 on that this could be a factor. Its plausible and it could be something that simple. The only thing I cant resolve is why/how did MAX thrust get commanded.


Cheers

[49d]

I think most professional pilots have internalized the scan...it takes place without thought...case in point, years ago we were ferrying a lear with the attitude indicators inop..using just the peanut gyro..we found you had to tape over the bad instruments, because you see and respond to the bad gyros even when your conscious mind knows not to

[V1rhot8]

@Flexibile:
Do you think it fair to presume that the pilot in this aircraft didn't have an instrument scan going? (Called by some a 'crosscheck')
If you are on instruments, then you fly with the AI/AHRS/HSI as your primary reference. You get trained to do that when you get an instrument rating.
(Granted, any of us can have a scan breakdown on a bad day).
So while what you say has been shown to be true, the remedy for that issue is also known:
establish a scan and use it to maintain attitude, performance, etc.


What if they had a instrument problem that caused the aircraft to present bad attitude information (nose high). Next the crew adds max thrust and pushes forward on the control column to lower AOA /the perceived unusual attitude towards “level.” Then they are trying to determine whether the attitude information is wrong or if they had a pitot static issue. Once the break out of the clouds they recognise the severity of the situation and start to pull up, but given the high load factor from doing nearly 400KIAS they can only recover to -20 degrees pitch.

[TRey]

What if they had a instrument problem that caused the aircraft to present bad attitude information (nose high). Next the crew adds max thrust and pushes forward on the control column to lower AOA /the perceived unusual attitude towards “level.” Then they are trying to determine whether the attitude information is wrong or if they had a pitot static issue. Once the break out of the clouds they recognise the severity of the situation and start to pull up, but given the high load factor from doing nearly 400KIAS they can only recover to -20 degrees pitch.


Pretty good theory...:) Especially for the "why max thrust" question.

[Lonewolf_50]

What appears to have been a perfectly serviceable B767 flown by a competent and experienced crew crashed after entering a steep dive. The reason for the accident is unknown at this stage. I presume nothing. Fair reply. *tips cap*

For V1rhot8
Interesting scenario, not beyond the possible depending on flight conditions at initiation of the event.
We'll see.

[EXDAC]

The only thing I cant resolve is why/how did MAX thrust get commanded.

Cheers

My speculation, and it is only speculation, is that turbulence was encountered while a seat swap was in progress and one of those involved in the swap put out a hand to steady themselves and pushed the throttles full forward. Speculation is based on lots of time in flight test where many seats swaps were observed. Who wouldn't have wanted to give the newbie a bit of stick time?

Of course that speculation is meaningless if all crew were found strapped in their appropriate seats.

[Lonewolf_50]

My speculation, and it is only speculation, is that turbulence was encountered while a seat swap was in progress and one of those involved in the swap put out a hand to steady themselves and pushed the throttles full forward. Speculation is based on lots of time in flight test where many seats swaps were observed. Who wouldn't have wanted to give the newbie a bit of stick time?
Of course that speculation is meaningless if all crew were found strapped in their appropriate seats. While I am not privy to Atlas' SOP, I'd be surprised if they did not dictate that such a crew swap would need to be done at altitude (FL whatever) before descent into the terminal area. We'll see.

[Banana Joe]

I am sure nobody would let a non-qualified pilot take commands. Definitely not at an established operator like Atlas. That's more than speculation. No offence.

[Old Boeing Driver]

What if they had a instrument problem that caused the aircraft to present bad attitude information (nose high). Next the crew adds max thrust and pushes forward on the control column to lower AOA /the perceived unusual attitude towards “level.” Then they are trying to determine whether the attitude information is wrong or if they had a pitot static issue. Once the break out of the clouds they recognise the severity of the situation and start to pull up, but given the high load factor from doing nearly 400KIAS they can only recover to -20 degrees pitch.

Very possible, and hopefully we will know more soon.
However, I would think that 3 sets of eyes would see the VSI pegged, Airspeed pegged, and the clacker sounding, all as warnings that something was not right.

[Banana Joe]

What if they had a instrument problem that caused the aircraft to present bad attitude information (nose high). Next the crew adds max thrust and pushes forward on the control column to lower AOA /the perceived unusual attitude towards “level.” Then they are trying to determine whether the attitude information is wrong or if they had a pitot static issue. Once the break out of the clouds they recognise the severity of the situation and start to pull up, but given the high load factor from doing nearly 400KIAS they can only recover to -20 degrees pitch.
It is possible and this is what happened to West Atlantic's CRJ that crashed in Norway (or was it Sweden) not long ago, however, the nose high recovery procedure does not call for maximum thrust. You would reduce thrust and increase bank to increase nose down pitch rate. From the video online it seems the aircraft was not in any banked attitude.

[Airbubba]

My speculation, and it is only speculation, is that turbulence was encountered while a seat swap was in progress and one of those involved in the swap put out a hand to steady themselves and pushed the throttles full forward. Speculation is based on lots of time in flight test where many seats swaps were observed. Who wouldn't have wanted to give the newbie a bit of stick time?

While I am not privy to Atlas' SOP, I'd be surprised if they did not dictate that such a crew swap would need to be done at altitude (FL whatever) before descent into the terminal area. We'll see.

A seat swap with a jumpseat rider from another carrier?

Even on a freighter that doesn't seem likely to me. At least not in 2019.

It was the case at some places that unauthorized seat swaps 'to give the engineer a leg' were common on ferry and freight flights as recently as the 1980's. I've also heard of giving a fed a leg without authorization.

For example, the seat swap with the flight engineer had happened in the crash of United 2885, a DC-8 freighter that went down at DTW in 1983:

https://en.wikipedia.org/wiki/United_Airlines_Flight_2885

In 1988 a Pan Am captain was fired for allegedly letting a flight attendant fly a 747 full of pax but he filed a union grievance and got his job back:

On September 25, 1988, Gay was employed by Pan Am as the captain of Pan Am Flight No. 81, a Boeing 747 passenger aircraft enroute from New York to Los Angeles. During the course of the flight, First Officer Dennis Brooks temporarily vacated the co-pilot's seat and Flight Attendant Naomi Kaneda sat in his place. It was reported that Ms. Kaneda put her hands on the control yoke, but it never was clearly established that she actually manipulated the controls of the aircraft. The testimony of certain witnesses before the National Transportation Safety Board did indicate that Ms. Kaneda actually hand-flew the aircraft and that the aircraft deviated from its assigned altitude at that time.

An investigation of the incident was conducted by the Regional Chief Pilot of Pan Am, and he concluded that Gay was guilty of misconduct in the incident. The Chief Pilot found that Gay had allowed an unqualified person to manipulate the controls of the aircraft, in violation of Pan Am's Flight Operations Manual and Federal Aviation Regulations. As a consequence of these findings, Gay was discharged by Pan Am on December 31, 1988. Gay then filed a grievance over his discharge as provided in the collective bargaining agreement entered into between Pan Am and its pilots. The grievance proceeded to arbitration before the System Board of Adjustment, the body designated to conduct arbitrations under the collective bargaining agreement. The Board found that Gay was deprived of a full and fair investigation of the charges lodged against him and ordered him reinstated with full back pay and seniority rights. The Board made no finding on whether Gay allowed a flight attendant to fly the aircraft.

https://law.justia.com/cases/federal/appellate-courts/F3/60/83/565776/

I haven't heard of these seat swap shenanigans for years though. Sometimes a current and qualified (and sober) deadheading pilot from the same company might offer to take a turn in the seat for a while as a courtesy to the operating crew but even that was explicitly forbidden in the manual about a decade ago where I worked.

[cappt]

The seat swap scenario doesn't sound likely. These three were all rated and experienced pilots, the lure of sitting in a seat just for the fun of it is not there. The Atlas pilots wouldn't even think of offering it up to a guy who just wants to get home/work. I've done a lot of jumpseating the last twenty years and it's never been brought up.

[MSVirginia]

Regarding the “why” on max thrust suddenly: if maneuvering around weather in Heading Select, it’s certainly possible to crank the Speed Select knob instead of the heading bug when intending to make a turn, commanding a higher airspeed inadvertently and no change in heading. The sudden thrust could surprise any crew, especially when tired and distracted by weather or ATC commands.

[jugofpropwash]

My speculation, and it is only speculation, is that turbulence was encountered while a seat swap was in progress and one of those involved in the swap put out a hand to steady themselves and pushed the throttles full forward. Speculation is based on lots of time in flight test where many seats swaps were observed. Who wouldn't have wanted to give the newbie a bit of stick time?

Of course that speculation is meaningless if all crew were found strapped in their appropriate seats.

How about the jump seat passenger simply standing up, perhaps to lean over the pilot's shoulder to get a better look at something? Or coming back from the loo? Or even seated in the jump seat, but perhaps leaning forward with belts off. And they hit turbulence, and he's pitched forward into the controls. Maybe sticks a hand out instinctively, trying to catch himself. Between the startle factor, passenger in the way, trying to regain his feet....

[TowerDog]

How about the jump seat passenger simply standing up, perhaps to lean over the pilot's shoulder to get a better look at something? Or coming back from the loo? Or even seated in the jump seat, but perhaps leaning forward with belts off. And they hit turbulence, and he's pitched forward into the controls. Maybe sticks a hand out instinctively, trying to catch himself. Between the startle factor, passenger in the way, trying to regain his feet....

Not likely, they were below 10,000’, everybody seated, strapped in and shoulder harness on.
seat swap even more unlikely, Captain could lose his job and his license for doing that.
The NTSB probably know what went wrong, but keeping quiet..

[extreme P]

Here is the message that has just been posted:
***************
Regarding the “why” on max thrust suddenly: if maneuvering around weather in Heading Select, it’s certainly possible to crank the Speed Select knob instead of the heading bug when intending to make a turn, commanding a higher airspeed inadvertently and no change in heading. The sudden thrust could surprise any crew, especially when tired and distracted by weather or ATC commands.

But the message never showed up?

One might notice an increase in engine noise, lack of turn, and max 399 in the speed window?

[Airbubba]

Here is the message that has just been posted:
***************


Any idea where this was just posted? :confused:

[jaysky]

Something along the lines of, upthread: [post 629]

Regarding the “why” on max thrust suddenly: if maneuvering around weather in Heading Select, it’s certainly possible to crank the Speed Select knob instead of the heading bug when intending to make a turn, commanding a higher airspeed inadvertently and no change in heading. The sudden thrust could surprise any crew, especially when tired and distracted by weather or ATC commands.

[Airbubba]

Something along the lines of, upthread: [post 629]

Thanks :ok:

But the message never showed up?

Sometimes you have to refresh the page to get the latest post, I've had the same issue.

[GlueBall]

V1rhot8 says: "What if they had a instrument problem that caused the aircraft to present bad attitude information (nose high). Next the crew adds max thrust and pushes forward on the control column to lower AOA /the perceived unusual attitude towards “level.” Then they are trying to determine whether the attitude information is wrong or if they had a pitot static issue."

In my years of flying the SAI has always been my friend, prominently in view just a few centimeters from the main instrument panel, easily incorporated in my scan. The idea of ever chasing a faulty attitude indicator into a 49 degrees dive is impractical reality.

[Australopithecus]

Is the pitot or “probe” heat on a 767 auto or manually selected? Does faulty probe heat trigger a caution? Just wondering if a blocked pitot on descent yielded an alarmingly low airspeed indication prompting an overreaction with thrust and attitude?

[V1rhot8]

V1rhot8 says: "What if they had a instrument problem that caused the aircraft to present bad attitude information (nose high). Next the crew adds max thrust and pushes forward on the control column to lower AOA /the perceived unusual attitude towards “level.” Then they are trying to determine whether the attitude information is wrong or if they had a pitot static issue."

In my years of flying the SAI has always been my friend, prominently in view just a few centimeters from the main instrument panel, easily incorporated in my scan. The idea of ever chasing a faulty attitude indicator into a 49 degrees dive is impractical reality.

Well I am just trying to formulate a theory of why a pilot might put in nose down control column input. The fact that the NTSB stated it initially combined with things I have been hearing leads me to believe that they did indeed push it over... but why? Also if the FO was PF, the SAI is not a few cm away.

The second component is the FBI has stepped down from the investigation. This in my mind rules out suicide, because if they were going that direction you'd need to investigate why someone might feel that way. You'd want an agency experienced with interviewing people.

[Sailvi767]

Regarding the “why” on max thrust suddenly: if maneuvering around weather in Heading Select, it’s certainly possible to crank the Speed Select knob instead of the heading bug when intending to make a turn, commanding a higher airspeed inadvertently and no change in heading. The sudden thrust could surprise any crew, especially when tired and distracted by weather or ATC commands.

Its certainly possible and I have had that happen. The change in aircraft pitch is barely noticeable while handflying at 230 knots. Certainly nothing that would cause issues.

[Murexway]

My speculation, and it is only speculation, is that turbulence was encountered while a seat swap was in progress and one of those involved in the swap put out a hand to steady themselves and pushed the throttles full forward. Speculation is based on lots of time in flight test where many seats swaps were observed. Who wouldn't have wanted to give the newbie a bit of stick time?

Of course that speculation is meaningless if all crew were found strapped in their appropriate seats.

Well, never say Never, but it would have been unusual, to say the least, in the terminal area.

[aterpster]

My speculation, and it is only speculation, is that turbulence was encountered while a seat swap was in progress and one of those involved in the swap put out a hand to steady themselves and pushed the throttles full forward. Speculation is based on lots of time in flight test where many seats swaps were observed. Who wouldn't have wanted to give the newbie a bit of stick time?

Of course that speculation is meaningless if all crew were found strapped in their appropriate seats.

6,000 feet, MSL, and within 30 miles of destination airport would be a highly inappropriate place to do a seat swap. And, a seat swap with the jump seat rider who was not authorized or qualified to fly the 767?

[B2N2]

My speculation, and it is only speculation, is that turbulence was encountered while a seat swap was in progress and one of those involved in the swap put out a hand to steady themselves and pushed the throttles full forward. Speculation is based on lots of time in flight test where many seats swaps were observed. Who wouldn't have wanted to give the newbie a bit of stick time?

Of course that speculation is meaningless if all crew were found strapped in their appropriate seats.

You’ve posted previously you were involved in something related to the MD-80.
No idea what that was but you don’t just do “seat swaps” for a jolly and certainly not at that stage of the flight.
Seat swaps with relief crew occur at cruise and the “landing” crew is seated 1-1.5 hrs out.
350-500 miles.

[tdracer]

Is the pitot or “probe” heat on a 767 auto or manually selected? Does faulty probe heat trigger a caution? Just wondering if a blocked pitot on descent yielded an alarmingly low airspeed indication prompting an overreaction with thrust and attitude?




Probe heat is automatic on the 767 (all heated probes). Any probe heat failure generates an EICAS message.

[Murexway]

You’ve posted previously you were involved in something related to the MD-80.
No idea what that was but you don’t just do “seat swaps” for a jolly and certainly not at that stage of the flight.
Seat swaps with relief crew occur at cruise and the “landing” crew is seated 1-1.5 hrs out.
350-500 miles.
Not to answer for EXDAC, but I flew Capt on the Mad Dog for 12 years and swapped legs if there was another qualified Capt on the crew list (rare), but never swapped seats. I did go to the lav a few times :) but only in cruise and nobody sat in my seat. If the sequence was over 8 hrs flying or 14 hrs duty (very rare) we had a relief crew, but the seat exchange was done at the gate.

[B2N2]

Exactly so this whole seat swap lost your balance or the jumpseater wanted to take a picture and fell onto the controls is just utter jibberish.
Frankely is astounding people even post this and are serious.

[bsieker]

Well, the first part of the name is "Professional Pilots" . Wish there was a way to enforce that....

This has been done to death and back. With that you'd throw away all the valuable contributions of engineers, ATC personnel, psychologists, sociologists, human-machine interface specialists, safety practitioners, and even *gasp* regulators and managers, etc., and end up a pilots-only echo chamber. I guess you can have that if you want, but that is not what this place is intended to be.

Bernd
(aerospace safety engineer, accident analyst, private pilot)

[Meester proach]

This has been done to death and back. With that you'd throw away all the valuable contributions of engineers, ATC personnel, psychologists, sociologists, human-machine interface specialists, safety practitioners, and even *gasp* regulators and managers, etc., and end up a pilots-only echo chamber. I guess you can have that if you want, but that is not what this place is intended to be.

Bernd
(aerospace safety engineer, accident analyst, private pilot)


OK, make it the “ professional people in aviation “ forum. I’m sure no one wants to discourage our colleagues in all the areas, but maybe try and stifle the spotters, nutters and those who wear tin foil hats

[cats_five]

OK, make it the “ professional people in aviation “ forum. I’m sure no one wants to discourage our colleagues in all the areas, but maybe try and stifle the spotters, nutters and those who wear tin foil hats

There are aviation professionals who are nutters, and who wear tin foil hats. And there are non-aviation professionals who have a great deal to contribute.

[The Ancient Geek]

Do you want to exclude those of us with no medical due to silver hair ?

[FCeng84]

Do you want to exclude those of us with no medical due to silver hair ?

No way!!! Too many of us a bit wetter behind the ears are so busy staring at the single tree we have been assigned to manage that we can miss the shape of the forest. All of us certainly need the wisdom of those of you who have been around long enough to see how we got where we are to have some sense of how to move forward. It's hard to know how to get out of the woods if you were born in the thick of it and grew up not knowing any better. As one a little past half way between newbie and certified grey beard I implore all of you who have been around the block a few times to keep sharing your insights with the rest of us. Lord knows we need every bit of help we can collect!

As gums might say, FCeng defers ...

[aterpster]

Not to answer for EXDAC, but I flew Capt on the Mad Dog for 12 years and swapped legs if there was another qualified Capt on the crew list (rare), but never swapped seats. I did go to the lav a few times :) but only in cruise and nobody sat in my seat. If the sequence was over 8 hrs flying or 14 hrs duty (very rare) we had a relief crew, but the seat exchange was done at the gate.
Seats have to be swapped in cruise for a long flight that requires a relief pilot.

[stilton]

Seats have to be swapped in cruise for a long flight that requires a relief pilot.


After four years flying the MD80 I can assure
you it doesn’t have the legs that would require an inflight seat swap for a relief pilot

[Skyborne Flyer]

Apparently Wall Street and Business Insider (https://www.businessinsider.de/pilot-error-amazon-air-atlas-air-crash-ntsb2019-3/?r=US&IR=T) reporting NTSB stating pilot error as the cause of the crash. In the final minutes, pilots increase thrust passing through wx, overreact to nose up pitch due to thrust increase, end up 49deg nose down. NTSB saying reaction very strange for experienced aviators, as is obvious. Very strange

[gums]

Salute!

Couldn't resist.....
by ater'pester"
Seats have to be swapped in cruise for a long flight that requires a relief pilot.

Well, I still consider myself a "professional". And had a little bit of time as a government-paid triggerman that allowed me to gain a degree of experience with aerodynamics and avionics and.... All in planes that I could not stand up in or walk to the loo ( deference to the Poms) or grab a fresh cup of coffee. So for a good test try sitting in a small chair with oxygen mask on and maybe a relief tube to help, or a box lunch so the crumbs clogged up your valve in the mask, for 13 or 14 hours.

Just my "whine of the day" at the pester!

Gums remembers....

Jez thot we needed a bit of light commentary and this thing is very sad.

[jimtx]

After four years flying the MD80 I can assure
you it doesn’t have the legs that would require an inflight seat swap for a relief pilot

It is very strange. The only thing that comes to mind is a DC-8 crash in the 80s or 90s where fatigue might have made the pilots “punch drunk”. So we would have to know what the pilots’ duty day was like. If a tag leg after an all nighter with delays in between flights they would be zombies. NTSB will let you know someday. Nobody on atlas dare chime in if that’s the case.

[runner1021]

It is very strange. The only thing that comes to mind is a DC-8 crash in the 80s or 90s where fatigue might have made the pilots “punch drunk”. So we would have to know what the pilots’ duty day was like. If a tag leg after an all nighter with delays in between flights they would be zombies. NTSB will let you know someday. Nobody on atlas dare chime in if that’s the case.
Looks like the aircraft operated KONT/KMIA/KIAH. Assuming this crew flew it out of KONT they'd be looking at about a 10 hour duty day.

[Weapons Grade]

After four years flying the MD80 I can assure
you it doesn’t have the legs that would require an inflight seat swap for a relief pilot
I am a little confused: I thought the airplane was a B767, not a MD80, in which case you are correct - the MD80 doesn't have the legs for a relief pilot. BUT...the crashed airplane was a B767, cest na pas?
As for the chatter about changing crew positions:
a. How many crew were assigned to the flight, and how many were type-rated and current on type (I may have missed that somewhere in the posts)?
b. Was the jump-seater (assumption) rated and current on type with the operator?
I opine that it would be a very "courageous" captain to permit a non-type rated company pilot to occupy either CM1 or CM2 seat at any stage of the flight.
I wait to be enlightened.

[jimtx]

Looks like the aircraft operated KONT/KMIA/KIAH. Assuming this crew flew it out of KONT they'd be looking at about a 10 hour duty day.
Thats what the aircraft did but we don’t know what the crew did. All nighter from somewhere,sit till they get their 11am departure ready for you, I would be a zombie. I only had to experience zombie land once to know that you are far from 100% and it differs with crews. But all guesswork. NTSB will have the story.

[DaveReidUK]

Apparently Wall Street and Business Insider (https://www.businessinsider.de/pilot-error-amazon-air-atlas-air-crash-ntsb2019-3/?r=US&IR=T) reporting NTSB stating pilot error as the cause of the crash.

You must be reading a different article.

From the one that you linked to:

"Robert Sumwalt, the NTSB's chairman, predicted it would take more than a year of work to determine the cause of the crash"

Which do you think is more likely to be something that the NTSB would actually say ?

[valvanuz]

You must be reading a different article.

From the one that you linked to:



Which do you think is more likely to be something that the NTSB would actually say ?

More than a year for the final report, probably. But by the preliminary report which should be out in a couple of weeks, there will be some pretty strong hints.

[aterpster]

More than a year for the final report, probably. But by the preliminary report which should be out in a couple of weeks, there will be some pretty strong hints.

Because it's an air carrier accident, sometime prior to the final report the NTSB will open a docket. The docket will have a wealth of factual reports including the CVR transcript.

[ABusDrivr]

Got this info from my airlines unofficial forum. Unofficial of course .

The initial bobble is from turbulence at 6200’. When the FO called for flaps 1, the captain accidentally hit the toga button. Toga didn’t engage until after flaps were set to 1, which then brought engine power to full, and started the initial pitch of 10 degrees nose up. The FO was startled, and shoved the nose forward... The CVR is startling, and baffling. The CA was pulling so hard against the FO that he sheared the pins on the stick and at that point had no control. They were IMC at the time. When they broke out into VMC, the FO said oh schit and started to pull. That was the round out you see. I won’t get into anything more until everything comes out. The records, the CVR, and what happened in the flight deck is truly shocking. They hit a negative 4 G dive initialy on the FOs push. All you hear is stuff hitting the ceiling and at one point a loud thud. They think the thud may have been the JS hitting the ceiling and maybe not wearing the shoulder harness. Like I said, I won’t get into anything more about the background of how it all happened. This is the accident in a nutshell. The facts that will come out are shocking.

[flyingchanges]

TOGA not even armed until flaps are not up.

[flyingchanges]

Nor does GA mode give you full power...

[jimtx]

TOGA not even armed until flaps are not up.

But you can envision continuous contact with the switches with the arm while moving the flap handle to 1. Would the thrust levers continue to advance because the rate of climb was not being acheived?

[EDML]

And the FO never learned to use the attitude indicator? With -49° pitch everything will be brown - easy to see.

[Meester proach]

Would you have to keep yourself pressing TOGA though, to get the power, as it was first pressed before “ flaps not down “.

would have to be the ultimate startle factor to go from a 2000 FPM climb to 49 degrees nose down

[Murexway]

Got this info from my airlines unofficial forum. Unofficial of course .

The initial bobble is from turbulence at 6200’. When the FO called for flaps 1, the captain accidentally hit the toga button. Toga didn’t engage until after flaps were set to 1, which then brought engine power to full, and started the initial pitch of 10 degrees nose up. The FO was startled, and shoved the nose forward... The CVR is startling, and baffling. The CA was pulling so hard against the FO that he sheared the pins on the stick and at that point had no control. They were IMC at the time. When they broke out into VMC, the FO said oh schit and started to pull. That was the round out you see. I won’t get into anything more until everything comes out. The records, the CVR, and what happened in the flight deck is truly shocking. They hit a negative 4 G dive initialy on the FOs push. All you hear is stuff hitting the ceiling and at one point a loud thud. They think the thud may have been the JS hitting the ceiling and maybe not wearing the shoulder harness. Like I said, I won’t get into anything more about the background of how it all happened. This is the accident in a nutshell. The facts that will come out are shocking.Just got the same sequence of events from an old classmate. Only thing additional was that the FO apparently entered the weather and turbulence with the speed brakes fully deployed. I also got the extra stuff about the FO.

[jimtx]

Just got the same sequence of events from an old classmate. Only thing additional was that the FO apparently entered the weather and turbulence with the speed brakes fully deployed. I also got the extra stuff about the FO.
So this might be relevant: https://www.businessinsider.com/amazon-air-pilots-labor-issues-2019-2?fbclid=IwAR3CShBNhhG_29Qe2cIRd8kTc8o71_NEu4r7RWzj9mWfOexWY zCom2nO24E

[V1rhot8]

Remember in the first page of this thread when someone filtered the ATC audio (background noise from foreground) and the last transmission was Ok. Well that pull you hear at the end is supposedly the jumpseater telling the crew to pull up.

We'll be able to confirm once the transcripts come out, but I've heard the same narrative as above.

Sounds like the PF got disoriented by the pitch up from TOGA being engaged, the buffeting from speedbrakes made him think they were stalling so he pushes it over (way more aggressively than you'd expect) freezes and then when they break out he comes to but by then recovery is impossible.

[AviatorDave]

So this might be relevant: https://www.businessinsider.com/amazon-air-pilots-labor-issues-2019-2?fbclid=IwAR3CShBNhhG_29Qe2cIRd8kTc8o71_NEu4r7RWzj9mWfOexWY zCom2nO24E

At least according to TT, I would not classify the crew as inexperienced in this case. However, if what was posted here as information off the CVR is true, it would indicate abysmal instrument flying skills of the FO, as well as possibly bad CRM if the CA cannot do better than ripping the stick out in an attempt to counteract an apparently disoriented FO who is pushing the nose down, and a FO who does not hand over the controls when in doubt about his abilities. Yes, events may have unfolded pretty fast, but still that‘s what crews are trained for.

Fatigue and being overworked might very well turn out to be key factors in letting the crew‘s performance drop below acceptable standards at the time of the accident.

[Murexway]

So this might be relevant: https://www.businessinsider.com/amazon-air-pilots-labor-issues-2019-2?fbclid=IwAR3CShBNhhG_29Qe2cIRd8kTc8o71_NEu4r7RWzj9mWfOexWY zCom2nO24EYes, but even at the major pax carriers life isn't perfect. Back of the clock flying and contract violations on the part of crew schedulers (who are being pushed) are part of life. But I'm sure it's much worse at the freight outfits below FEDEX and UPS.

[Murexway]

Only additional stuff I saw was that the airplane crashed with the autothrottles and A/P engaged and the elevators were split due to cross inputs on the control columns. It was only 18 seconds from 6,300 to the ground.

[SteinarN]

Even a well-trained and able crew will eventually screw up when fatigue and being overworked comes into play.

I agree completely. Even the best person can screw up on a bad day, even more so when tired.

[Twiglet1]

Even a well-trained and able crew will eventually screw up when fatigue and being overworked comes into play.

The flight was during the day. If the crew had been up all night then yes fatigue may/will have come into it. If not then I hope it wasn't self inflicted e.g. crew commuting
Only the final report will confirm. May they RIP

[EternalNY1]

Got this info from my airlines unofficial forum. Unofficial of course .

The initial bobble is from turbulence at 6200’. When the FO called for flaps 1, the captain accidentally hit the toga button. Toga didn’t engage until after flaps were set to 1, which then brought engine power to full, and started the initial pitch of 10 degrees nose up. The FO was startled, and shoved the nose forward... The CVR is startling, and baffling. The CA was pulling so hard against the FO that he sheared the pins on the stick and at that point had no control. They were IMC at the time. When they broke out into VMC, the FO said oh schit and started to pull. That was the round out you see. I won’t get into anything more until everything comes out. The records, the CVR, and what happened in the flight deck is truly shocking. They hit a negative 4 G dive initialy on the FOs push. All you hear is stuff hitting the ceiling and at one point a loud thud. They think the thud may have been the JS hitting the ceiling and maybe not wearing the shoulder harness. Like I said, I won’t get into anything more about the background of how it all happened. This is the accident in a nutshell. The facts that will come out are shocking.

Wow. Just ... wow.

I understand the pitch up illusion (and in this case, reality, at least to 10 degrees) you get when going TOGA in the clouds.

But how can you possibly pitch down to 49 degrees in response??

I don't care if you were trained only on Cessna's before this flight ... you don't even do this in a Cessna!

If this is at all true, and if the FO didn't do that intentionally, it boggles the mind.

[tdracer]

When the FO called for flaps 1, the captain accidentally hit the toga button. Toga didn’t engage until after flaps were set to 1, which then brought engine power to full, and started the initial pitch of 10 degrees nose up.

I suspect the stated sequence is a bit off - i.e. the captain selected flaps 1, then bumped TOGA. TOGA is armed as soon as 'flaps not stowed' is true which would happen almost immediately after flaps 1 selected.
Sounds like spacial disorientation strikes again :(.

[Bleve]

Somatogravic Illusion (http://aviationknowledge.wikidot.com/aviation:somatogravic-illusion)
I’ve experienced it once at night and I can tell you it is almost totally overwhelming. I can easily imagine the FO (due to the large acceleration) believing he was pitching very high nose up and ‘instinctively’ pushing forward. That increases acceleration which makes the illusion even worse, so you want to push forward even more. By this stage he would have been in cognitive overload and totally unresponsive to the CA’s desperate attempts at recovery.

[AviatorDave]

Read the link posted by jimtx.
Crews seem to be pushed way over the limits, so I don‘t think that anything can be ruled out by the fact that the accident happened during daytime.

[COflyer]

I suspect the stated sequence is a bit off - i.e. the captain selected flaps 1, then bumped TOGA. TOGA is armed as soon as 'flaps not stowed' is true which would happen almost immediately after flaps 1 selected.
Sounds like spacial disorientation strikes again :(.

Would you need to bump just one TOGA sw, or both?

[Herod]

IIRC, on all aircraft I've flown that had TOGA switches, the buttons are inset, to prevent accidental activation. That means a digit has to be inserted into the cup to select TOGA, not just bumping against the switch.

[V1rhot8]

Would you need to bump just one TOGA sw, or both?


Just one

IIRC, on all aircraft I've flown that had TOGA switches, the buttons are inset, to prevent accidental activation. That means a digit has to be inserted into the cup to select TOGA, not just bumping against the switch.

It's a switch on the back of the thrust lever that sticks out

https://i0.wp.com/aerosavvy.com/wp-content/uploads/2017/08/GoAround-Switches-1200.jpg?ssl=1

[747-8driver]

Weird, on the B747 those are the autothrottle disconnect switches...

[Old Boeing Driver]

Got this info from my airlines unofficial forum. Unofficial of course .

The initial bobble is from turbulence at 6200’. When the FO called for flaps 1, the captain accidentally hit the toga button. Toga didn’t engage until after flaps were set to 1, which then brought engine power to full, and started the initial pitch of 10 degrees nose up. The FO was startled, and shoved the nose forward... The CVR is startling, and baffling. The CA was pulling so hard against the FO that he sheared the pins on the stick and at that point had no control. They were IMC at the time. When they broke out into VMC, the FO said oh schit and started to pull. That was the round out you see. I won’t get into anything more until everything comes out. The records, the CVR, and what happened in the flight deck is truly shocking. They hit a negative 4 G dive initialy on the FOs push. All you hear is stuff hitting the ceiling and at one point a loud thud. They think the thud may have been the JS hitting the ceiling and maybe not wearing the shoulder harness. Like I said, I won’t get into anything more about the background of how it all happened. This is the accident in a nutshell. The facts that will come out are shocking.

Just a couple of questions....... This is the first post from this person........
The info came from an unofficial airline forum
Does TOGA command full power?
If one control column "pins" are sheared, does that make the other useless?
Wasn't this a bit far out to start putting flaps out?
Were they actually IMC when this started?
Another post said they had the boards up. Would they be calling for flaps with the boards up?

Thanks for any comments.

OBD

Update (per an instructor): TOGA tries to maintain 2K ROC...They were IMC. It is OK to use flaps/boards in some circumstances, and they were close enough to use both in order to expidite descent. Unknown about the column question.

[742]

Got this info from my airlines unofficial forum. Unofficial of course .

The initial bobble is from turbulence at 6200’. When the FO called for flaps 1, the captain accidentally hit the toga button. Toga didn’t engage until after flaps were set to 1, which then brought engine power to full, and started the initial pitch of 10 degrees nose up. The FO was startled, and shoved the nose forward... The CVR is startling, and baffling. The CA was pulling so hard against the FO that he sheared the pins on the stick and at that point had no control. They were IMC at the time. When they broke out into VMC, the FO said oh schit and started to pull. That was the round out you see. I won’t get into anything more until everything comes out. The records, the CVR, and what happened in the flight deck is truly shocking. They hit a negative 4 G dive initialy on the FOs push. All you hear is stuff hitting the ceiling and at one point a loud thud. They think the thud may have been the JS hitting the ceiling and maybe not wearing the shoulder harness. Like I said, I won’t get into anything more about the background of how it all happened. This is the accident in a nutshell. The facts that will come out are shocking.

So the stab is trimmed for 240 knots. FO pushes full forward. Captain pulls back and splits the yokes (as designed). So you have a stab trimmed for 240 knots, one elevator full down and one elevator full up. And a lot of power has been added, which of course is "nose up".

That does not result in what the airplane did.

[Old Boeing Driver]

So the stab is trimmed for 240 knots. FO pushes full forward. Captain pulls back and splits the yokes (as designed). So you have a stab trimmed for 240 knots, one elevator full down and one elevator full up. And a lot of power has been added, which of course is "nose up".

That does not result in what the airplane did.

When one column over rides they other, full control is given to the operable control column. the control surfaces are not split.

Again, this would be the case in older Boeings. Do not know about this one.

[Old Boeing Driver]

Here is a diagram to show the "Break out" box. When one side overrides the other, control is not lost. Just transferred to the other column. Diagram is from a 737NG

https://cimg5.ibsrv.net/gimg/pprune.org-vbulletin/640x420/controls_5d9aaa7068113856f9f8b117c6d9cafe8be292b0.png

[TRey]

Weird, on the B747 those are the autothrottle disconnect switches...

And is quite the adjustment when coming from flying a 747 to the 75/76 I fly now.

As to the other discussions:
TOGA on Boeing...it is correct that 2000 FPM is commanded on the FIRST push of the TOGA switch...on some Boeing 74/75/76 (customer option) the SECOND push gets you MAX thrust. So the scenario that these were mistakenly pushed once and possibly twice might be a long shot given the system design and physical placement of the TOGA switches. NTSB said MAX thrust was recorded on the FDR. So one push doesn't make sense to me given what we know about the system.

[Pilot DAR]

When one column over rides they other, full control is given to the operable control column. the control surfaces are not split.

Question (not assertion), I recall training which lead me to understand that in the case of a jammed control surface, and control columns being disconnected from each other, the two halves of the flight control systems were now independent of each other (split), and the two pilots were to then determine who's controls were more effective in maintaining control. I don't recall a mechanical system which gave full control to one control column over another. This training to me was in respect of a DC-8-63. Is my recollection in error?

The Jetstream 4100 has a cockpit release to split the two elevator circuits from each other, so the left control wheel controls the left elevator, and right controls right. In the Dash 8, there is a release in the ailerons, so if one jams, it is released, and split from the other, and the cockpit controls now operate only the free aileron. This is unrelated to control columns being disconnected from each other at the cockpit.

Doesn't the "breakout box" depicted for the 737NG, simply allow the two control columns to act independently of each other to control their respective elevator?

[niteflyr]

When one column over rides they other, full control is given to the operable control column. the control surfaces are not split.

Again, this would be the case in older Boeings. Do not know about this one.

No. They would split.

[Old Boeing Driver]

Here is the 767 diagram. Image is not as clear as the other. It does show there is some type of override between the 2 columns. I think it has something to do with the transducers. The systems are split, left and right, but in the override scenario, the operable one still controls all surfaces.

https://cimg1.ibsrv.net/gimg/pprune.org-vbulletin/296x170/767_9e46333dcbf82ea57d961ff0cb8ec47f89e6e41b.jpg

[Airbubba]

TOGA on Boeing...it is correct that 2000 FPM is commanded on the FIRST push of the TOGA switch...on some Boeing 74/75/76 (customer option) the SECOND push gets you MAX thrust. So the scenario that these were mistakenly pushed once and possibly twice might be a long shot given the system design and physical placement of the TOGA switches. NTSB said MAX thrust was recorded on the FDR. So one push doesn't make sense to me given what we know about the system.

Actually, max thrust is indeed commanded initially with the 767 TOGA switches and then thrust is reduced when a 2000 fpm climb is achieved. So, the thrust will not throttle back if you don't allow the nose to rise and you don't pull back or disarm the autothrottles.

[MLHeliwrench]

Actually, max thrust is indeed commanded initially with the 767 TOGA switches and then thrust is reduced when a 2000 fpm climb is achieved. So, the thrust will not throttle back if you don't allow the nose to rise and you don't pull back or disarm the autothrottles.

Does pulling back on the throttles disengage autothrottle? Is there an announciation when TOGA selected. If yes - visual and/or aural? If throttles are pulled back does TOGA pitch up cancel?

I guess what I’m getting at is how quickly would the thrust and climb be recognized as an inadvertent activation of TOGA.

The possible delay between possible switch activation and flaps 1 being configured and the resultant TOGA would make it hard to figure out immediately.

[extreme P]

" When pushing either GA switch:

• autothrottle increases thrust with adjustment, as necessary, to maintain a climb rate of 2000 fpm"

FCOM -300ER reference.

[SpamCanDriver]

Been a long rime since I've flown the 767.
But this doesn't make sense, the TOGA switches only work when armed.
So accidentally pressing them, then selecting flaps 1 would result in nothing happening????

[Flightcap]

I know nothing about the 76, so can't comment on the mechanics of the TOGA switch. But if the above narrative is correct, the physiological factors alone may hampered the recovery. A -4.0 G load is something very few of us have ever experienced or are prepared to be surprised by. It is possible the CA and FO were physically unable to correct the dive once started until the aircraft was established in a more neutral G load. Further, the resulting vision impairment would have in itself been greatly disorienting and may have caused the FO to lose track of his artificial horizon.

[Livesinafield]

Take that CVR "report" with a big pinch of salt, I remember on the germanwings accident there was a "report" about how on the Cvr they heard the window go and then all rushing air as the plane came apart....

[TRey]

Actually, max thrust is indeed commanded initially with the 767 TOGA switches and then thrust is reduced when a 2000 fpm climb is achieved. So, the thrust will not throttle back if you don't allow the nose to rise and you don't pull back or disarm the autothrottles.

Hey Bubba...thanks for additional info.
Unfortunately I have never witnessed that particular behavior in the 757 or 767 in real life or in the sim world. Often times the power only comes up 1/2 to 2/3 of the G/A thrust limit, depending on weight and other factors. In my experience Boeing has a lot of options on these airplanes so that’s one I haven’t heard yet...or it could be I just don’t pay close enough attention...just ask any FO who flies with me, they’ll tell ya:)

[V1rhot8]

Just a thought, what if bumping TOGA with F1 was only the cause of the pitch change, and since the PF thought it was going to stall he pushed the yoke forward then gave it max thrust?

[dusk2dawn]

Just a thought, what if bumping TOGA with F1 was only the cause of the pitch change, and since the PF thought it was going to stall he pushed the yoke forward then gave it max thrust?
And then forgot about it for the next 6000’???

[atakacs]

Complete ignorant here...

Would selecting TOGA move the thrust leveler accordingly?

[FlightDetent]

Does the 767 have accelerated trim rate with flaps extended?

My brutally condensed account of the Rostov (737) accident is that pushover WITH trim (and flap relief + speedtape confusion) put the AC nose heavy into dive from which the crew then did not have the right combination of alertness, skills and altitude to recover.

If the push came with a sustained nose-down trim, how long for it to become impossible to overpower by the other pilots pull?

[Organfreak]

Does the 767 have accelerated trim rate with flaps extended?

My brutally condensed account of the Rostov (737) accident is that pushover WITH trim (and flap relief + speedtape confusion) put the AC nose heavy into dive from which the crew then did not have the right combination of alertness, skills and altitude to recover.
If the push came with a sustained nose-down trim, how long for it to become impossible to overpower by the other pilots pull?

Is it online somewhere?

[V1rhot8]

And then forgot about it for the next 6000’???
I know that's the part that doesn't make any sense.

[Feathers McGraw]

I know that's the part that doesn't make any sense.

Suspect that the -4g pushover would be a significant factor.

[EternalNY1]

Suspect that the -4g pushover would be a significant factor.

How is that even possible?

I've been through this illusion many times in IMC and while very convincing, nothing that would cause me to shove a 767 control column to the stops.

And I'm disregarding the mentions of the captain pulling opposite.

[Euclideanplane]

With so many hacks in effect anyway, would it make sense to program Hal to not allow any maneuvers that would subject inhabitants of the flight deck to more than +/- 3g of vertical push force? I realize that it could be necessary towards the very end to have leeway. But it seems that in an otherwise normal regime, it should not be possible to achieve it somehow by mistake, as what seems the case here.

[safelife]

Makes sense. Standard on Airbus FBW, since 1986.

[henra]

The systems are split, left and right, but in the override scenario, the operable one still controls all surfaces.

Stupid question: How does one end up in an override scenario compared to a split?
And second: What determines which yoke is overriding which?

[rhanson]

I'm having a real hard time with the -4g load, especially in a transport category aircraft designed for -1.0 + safety factor. Even when I hit the -2g limits in an aerobatic, it's really significant.

Edit: in fact does the 767 even have sufficient elevator/stab authority to get itself into that situation, relatively instantaneously?

[Pilot DAR]

Suspect that the -4g pushover would be a significant factor.

I suspect that a -4G pushover would result in a lot of bent 767, a lot of cargo pinned against the ceiling in the hold, and pilots being forced out of their seats, with everything loose in the cockpit pinned against the cockpit ceiling. If that airplane experienced that event, I opine it was hardly flying, or being flown under control, afterward.

[EternalNY1]

I suspect that a -4G pushover would result in a lot of bent 767, a lot of cargo pinned against the ceiling in the hold, and pilots being forced out of their seats, with everything loose in the cockpit pinned against the cockpit ceiling. If that airplane experienced that event, I opine it was hardly flying, or being flown under control, afterward.

Agreed. All the valid talk about inexperienced FO's aside, a -4G push, even as a quick reaction, is insanity.

You would think that would quickly be rectified.

[suninmyeyes]

What I find unlikely is the comment that the Captain pulled back so hard he sheared the pins on his control column. I have always been taught on Boeings that you cannot apply excess force to the control column that can result in damage. If the controls are jammed due ice you can pull or push as hard as you physically can. If the Captain and copilot oppose each other there will eventually be a break out and then then the left elevator will follow the Captain’s input and the right elevator will reflect the copilot’s input. This happened on the Egypt Air 767 crash in the atlantic.

[megan]

Wonder if there are any similarities with Northwest Boeing 720 accident.

ASN Aircraft accident Boeing 720-051B N724US Everglades, FL (http://aviation-safety.net/database/record.php?id=19630212-0)

The final report on the crash determined the cause of the accident to be the unfavourable interaction of severe vertical air drafts and large longitudinal control displacements, resulting in a longitudinal upset from which a successful recovery was not made.

[Educated Airman]

This latest discussion supports my theory of the stab actuator jackscrew fuselage attach bolt falling out. As the bolt worked its way out to the point where only the threads were engaged with the jackscrew, there would be enough slack to allow an aerodynamically preloaded stab to move slightly stab nose down resulting in a slight pitch up. The PF would instinctively push forward on the column aerodynamically unloading the stab rotational force and allow the bolt to fall out and the stab to rotate stab nose up past its design limit. When the stab moved pat the design limits, contact with the stab structure and the elevator control cables would have compromised the control cable integrity rendering the elevator controls useless. Due to the sudden unexpected negative Gs, the PF would instinctively think they were in a wind shear stall ant push forward on the column and throttles. The other pilot would respond to the precipitous nose down attitude by pulling aft on the column. Being in a sudden loss of control situation, and both pilots equally convinced that their recovery attempt was correct would result in a column disconnect with no resulting elevator movement as the control cable integrity was compromised. When the sudden and severe pitch down occurred, the pitch momentum would allow the stab to stall out. As the pitch down momentum continued, the greater flat square area of the stab forward of the hinge point would push the stab to a stab nose down causing the final pitch up. Look at the picture of the jackscrew and notice no blot or compromise of the attachment lug.

[Feathers McGraw]

Agreed. All the valid talk about inexperienced FO's aside, a -4G push, even as a quick reaction, is insanity.

You would think that would quickly be rectified.

Just a thought, any g applied is effectively due to a pitch change around the centre of lift/gravity. I think that means a magnified g load at the front of the aircraft, so what is felt on the flight deck is effectively more than the average g felt by the airframe.

[EternalNY1]

Just a thought, any g applied is effectively due to a pitch change around the centre of lift/gravity. I think that means a magnified g load at the front of the aircraft, so what is felt on the flight deck is effectively more than the average g felt by the airframe.

Oh no doubt about that, a -4G push is putting everyone hard against the restraints and possibly the jumpseater against the ceiling, depending on if they are strapped in.

[runner1021]

Oh no doubt about that, a -4G push is putting everyone hard against the restraints and possibly the jumpseater against the ceiling, depending on if they are strapped in.
Against the ceiling until the negative Gs ended, and then forward into the throttle quadrant? A frightening thought.

[NSEU]

FO pushes full forward. Captain pulls back and splits the yokes (as designed). So you have a stab trimmed for 240 knots, one elevator full down and one elevator full up. And a lot of power has been added, which of course is "nose up".

That does not result in what the airplane did.

Note that the split elevator system on the 767-300 has a limit of 20 degrees between each side. There is a torque tube in the tail tying the two sides together, but there is an override mechanism on this torque tube which allows independent left/right side movement up to 20 degrees. The column force required to activate this override is 25 lbs at breakout, increasing to 41 lbs at 20° of elevator movement. This is related to the tail stuff. The override at the front (between the columns) is also 25lbs. With no air load, the elevator can move a maximum of 28.5° up and 20.5° down with a full forward and aft movement of the control column. If the autopilot was engaged, they would be fighting the camout springs in the elevator servos (until a camout was detected and the A/P disconnected). On top of all this, you have elevator feel.

You'd need to be a rocket scientist to figure out the cumulative effect of all these devices.

Shear pins? Que?

[henra]

This latest discussion supports my theory of the stab actuator jackscrew fuselage attach bolt falling out.

Do you really think this scenario would look identical on the FDR traces as a full ND Elevator Control column input?
Cue: NTSB are no idiots.

[hooke.s_law]

Sorry, I cannon post the link. There is an article on Quora "What-caused-an-Atlas-Air-767-to-crash". For updates and details click on (more)... Among the latest updates regarding the NTSB investigation, the article includes a picture of the recovered jackscrew and its position at impact.

[Airbubba]

I wonder if the report of a 4 g push over is accurate. The single poster ABusDrivr also said that the nose pitched up 10 degrees with the power increase but the NTSB said the pitch up was about 4 degrees in its March 12, 2019 update.

Under NTSB CVR protocols only a few people will ever hear the CVR recording. Has someone been authorized to release a gist of the contents through Atlas or the Teamsters?

From the NTSB CVR Handbook cited previously, here is the non-disclosure agreement:

Cockpit Voice Recorder Handbook for Aviation Accident Investigation, February 2016

CVR NONDISCLOSURE AGREEMENT
ACCIDENT ID:_________________________________

Each of the undersigned acknowledges that he/she has read National Transportation Safety Board (NTSB) regulations in Title 49 Code of Federal Regulations 831.11 and 831.13, which are printed on the reverse side of this document, and agrees to comply with those rules.

Each undersigned further agrees not to disclose or cause to be disclosed any information from the cockpit voice recorder (CVR) recording or any transcript thereof except as expressly authorized by the NTSB. Approval from the NTSB Investigator-In-Charge must be obtained before disclosure of any information. Disclosure of information to his/her respective organization shall include only that information which is directly related to safety and at no time shall non-pertinent remarks, comments, or conversations be disclosed to any person, party, or organization.

By placing his/her signature hereon, the undersigned acknowledges that the unauthorized release of CVR information shall be grounds for immediate dismissal from the investigation, and may result in further legal sanction.
I HAVE READ AND UNDERSTAND THE ABOVE CONDITIONS.
SIGNATURE NAME (PRINTED) PARTY DATE

____________________ ________________________ ______________ _________

Further guidance for CVR contents disclosure in the interest of safety:

8.3. CVR disclosure policy does not prohibit a party from implementing safety-of-flight related changes within its organization as a result of its participation in a CVR group, however, the NTSB requires approval by the IIC and the directors of the Offices of Research and Engineering and Aviation Safety of the subject matter prior to any disclosure of CVR information to the group member’s organization. It is the CVR group member’s responsibility to notify the CVR group chairman prior to disclosure. Disclosure of information to a group member’s respective organization shall include only information that is directly related to safety. At no time shall CVR remarks, comments, or conversations be disclosed to any person, party, or organization.

Is the version of events from ABusDrivr based on a briefing from the Atlas union or airline safety folks? Or, is it crew van speculation based on what little information has been publicly released?

It could literally be months before the CVR transcript is released.

Did ABusDrivr give us a hot tip on what to expect? Or, did he drop off the Beale Ciphers and depart, never to be heard from again?

[MartinAOA]

Just got the same sequence of events from an old classmate. Only thing additional was that the FO apparently entered the weather and turbulence with the speed brakes fully deployed. I also got the extra stuff about the FO.

Only additional stuff I saw was that the airplane crashed with the autothrottles and A/P engaged and the elevators were split due to cross inputs on the control columns. It was only 18 seconds from...

Don't these comments by Murexway somehow validate ABusDrivr's post?

[ABusDrivr]

1 post and gone?
I have tried to respond a few times but for some unknown reason my responses are not being posted. The sequence of events I posted came from a non-Atlas pilot only forum from a US airline. I'm guessing, as well as one other poster here, someone has gotten a rough sequence of events from what happened from someone "in the know." I have been a long time lurker on the forum here and just wanted to contribute info that I hadn't yet seen on here. I am an Airbus 320 series CA for a US airline .

[Airbubba]

Don't these comments by Murexway somehow validate ABusDrivr's post?

Or at least, maybe the version given originates from the same source whether or not it is accurate.

I also got the extra stuff about the FO.

There have been posts made on other forums about the FO's training history at Mesa and Atlas but they seem to be quickly removed.

[Dave Therhino]

I wonder if the report of a 4 g push over is accurate. ?

A "4g pushover" (-4g, or even -3g) is well beyond the design envelope. They were in a 4g pull up for the last several seconds - maybe that's what you read?

[DaveReidUK]

This is in danger of becoming apochryphal. Can anyone point to a verifiable reference to this supposed 4g pushover?

If not, I suggest we move on ...

[fox niner]

This accident happened on feb 23. Almost 30 days ago, so a preliminary report is almost due.
Should contain info about a 4g negative dive (with or without a mig 28)

[DJ77]

4g or 49 ?

[Rated De]

This is in danger of becoming apochryphal. Can anyone point to a verifiable reference to this supposed 4g pushover?

If not, I suggest we move on ...

Yes.
There are three families looking for answers.
There is media feeding from posted comments.

[Airbubba]

A "4g pushover" (-4g, or even -3g) is well beyond the design envelope. They were in a 4g pull up for the last several seconds - maybe that's what you read?

From ABusDrivr's post above:

They hit a negative 4 G dive initialy on the FOs push. All you hear is stuff hitting the ceiling and at one point a loud thud. They think the thud may have been the JS hitting the ceiling and maybe not wearing the shoulder harness.

4g or 49 ?

Certainly a good possibility that a typo could be the source of this claim.

I have tried to respond a few times but for some unknown reason my responses are not being posted. The sequence of events I posted came from a non-Atlas pilot only forum from a US airline. I'm guessing, as well as one other poster here, someone has gotten a rough sequence of events from what happened from someone "in the know."

Thanks for the background on your earlier post.

[Old Boeing Driver]

Regarding this GA button issue. From what I have read here, the button is only armed when flaps are not "0"

I assume that selection of flaps 1 causes the LED's to extend.

If that is so, surely they would have been blown off the airplane by the time it reached 425 knots going down.

However the NTSB said they found a very small debris field and there has neen no mention of parts and pieces being found elsewhere.

[B2N2]

A -49 degree pushover
B -4g degree pushover
C -4g pushover

Im going with option A and calling it a typo.
The -49 degrees has been established by the NTSB.
A minus 4G has not.

[Sailvi767]

The aircraft would not have survived a -4G push. -1G is the normal airliner design goal for no damage and -1.5 for failure. In addition it’s doubtful to impossible that the airfoil on the 767 would be capable of producing -4 G’s at 210 knots. It’s not a symmetrical aerobatic airfoil and it CL would be very poor under negative G. It would exceed the stalling AOA well before 4 G’s.

[Euclideanplane]

The aircraft would not have survived a -4G push. -1G is the normal airliner design goal for no damage and -1.5 for failure. In addition it’s doubtful to impossible that the airfoil on the 767 would be capable of producing -4 G’s at 210 knots. It’s not a symmetrical aerobatic airfoil and it CL would be very poor under negative G. It would exceed the stalling AOA well before 4 G’s.

Exceeding stalling AoA is incompatible with the video which does show a nose dive and not a stall, is there agreement about that?
If there is physical damage to the aircraft in this scenario, would there be bits and pieces departing from it, and which ones?

[henra]

Exceeding stalling AoA is incompatible with the video which does show a nose dive and not a stall, is there agreement about that?
If there is physical damage to the aircraft in this scenario, would there be bits and pieces departing from it, and which ones?
What he is referring to is that -4g (besides very likely shedding the wings downward at -4g) would be prevented by a negative accelerated stall much below -4g at those speeds since the Cl max for an Airliner wing in negative direction will be much worse than in positive Cl. And at 230 kts even in positive direction max g (at Cl max) will be somewhere around 2,5. Above the AoA for that Cl max you will get an accelerated stall and no further increase in g load.
So we can pretty safely rule out -4g on aerodynamic/physics reasons alone.

[Sailvi767]

Exceeding stalling AoA is incompatible with the video which does show a nose dive and not a stall, is there agreement about that?
If there is physical damage to the aircraft in this scenario, would there be bits and pieces departing from it, and which ones?

You can stall a aircraft in positive or negative G. I am referring to stalling the wing under negative G.

[B2N2]

Aerobatic performances do not reach that amount of negative G-force nor does a fighter plane.
At -4G with a transport category aircraft the structure will fail.

This is NOT a legitimate 767 Vg diagram but borrowed from another thread on here for the purposes of explaining.


https://cimg2.ibsrv.net/gimg/pprune.org-vbulletin/727x484/image_44137132bf1d76acf3f88e237e3ba3a6cb2d4050.jpeg

You can’t just pull or push unlimited.
You need to stay within the operating envelope both in positive and negative G.
Again this is NOT a 767 diagram so ignore the speeds it’s about the aerodynamic principles.
The “normal” stall speed is the speed at which the aircraft stalls under unaccelerated flight at 1G.
Above this speed we call it an “accelerated stall” as we our now stalling the aircraft at a higher speed and higher then 1G.
The limit for this is Va or maneuvering speed.
Above this speed we will overstress the airplane before it stalls. Could be temporary or permanent deformations of structure depending on the speed.
This requires inspections and possibly repairs.
We have something similar below the 1-G line although the area in the graph is much smaller.
Keep in mind that we float (weightless) at 0G and even -1G is a pretty radical maneuver let alone -2G.

Design and certification criteria mandate the aircraft is much stronger on the “+” side then on the “-“ side of the graph.
Just as an example, hard landings are “-” and even extreme turbulence may not go much past -1G if at all unless we fly into a massive thunderstorm.
Keep in mind again, 0 is weightless and -1 we’re being accelerated out of our seats.

Here is a rare example of extreme turbulence that sent a galley cart into the ceiling. That will already happen at -.01G during which it will float up and at -0.3 it will probably smack the ceiling.


https://cimg5.ibsrv.net/gimg/pprune.org-vbulletin/619x793/image_e074feec0a9de7cbe8275bbcb5076f3eb9d53c67.jpeg

With all respect but I think the ones that advocate the -4G theory have never experienced even -1G.

Have a look at the diagram again.
In short the airplane would have come apart at -4G or at the very least lost the tail.
It did not, the videos show it with all major components attached.

[Sailvi767]

Aerobatic performances do not reach that amount of negative G-force nor does a fighter plane.
At -4G with a transport category aircraft the structure will fail.

This is NOT a legitimate 767 Vg diagram but borrowed from another thread on here for the purposes of explaining.


https://cimg2.ibsrv.net/gimg/pprune.org-vbulletin/727x484/image_44137132bf1d76acf3f88e237e3ba3a6cb2d4050.jpeg

You can’t just pull or push unlimited.
You need to stay within the operating envelope both in positive and negative G.
Again this is NOT a 767 diagram so ignore the speeds it’s about the aerodynamic principles.
The “normal” stall speed is the speed at which the aircraft stalls under unaccelerated flight at 1G.
Above this speed we call it an “accelerated stall” as we our now stalling the aircraft at a higher speed and higher then 1G.
The limit for this is Va or maneuvering speed.
Above this speed we will overstress the airplane before it stalls. Could be temporary or permanent deformations of structure depending on the speed.
This requires inspections and possibly repairs.
We have something similar below the 1-G line although the area in the graph is much smaller.
Keep in mind that we float (weightless) at 0G and even -1G is a pretty radical maneuver let alone -2G.

Design and certification criteria mandate the aircraft is much stronger on the “+” side then on the “-“ side of the graph.
Just as an example, hard landings are “+” and even extreme turbulence may not go much past -1G if at all unless we fly into a massive thunderstorm.
Keep in mind again, 0 is weightless and -1 we’re being accelerated out of our seats.

Here is a rare example of extreme turbulence that sent a galley cart into the ceiling. That will already happen at -.01G during which it will float up and at -0.3 it will probably smack the ceiling.


https://cimg5.ibsrv.net/gimg/pprune.org-vbulletin/619x793/image_e074feec0a9de7cbe8275bbcb5076f3eb9d53c67.jpeg

With all respect but I think the ones that advocate the -4G theory have never experienced even -1G.

Have a look at the diagram again.
In short the airplane would have come apart at -4G or at the very least lost the tail.
It did not, the videos show it with all major components attached.


Much better explained then how I posted it. -4G’s in this accident is a myth.

[B2N2]

Pitch -49 must have been misquoted somewhere as Pitch -4g which then took a life of its own.
Its not possible.

Hard landings are actually “-“ as far as the structure is concerned. Previous post corrected.

[megan]

-4G’s in this accident is a mythThe Northwest Boeing 720 accident I mentioned earlier they reached -2.8

[568]

Regarding this GA button issue. From what I have read here, the button is only armed when flaps are not "0"

I assume that selection of flaps 1 causes the LED's to extend.

If that is so, surely they would have been blown off the airplane by the time it reached 425 knots going down.

However the NTSB said they found a very small debris field and there has neen no mention of parts and pieces being found elsewhere.

RA <2500 Feet
Flaps out of "up"
Glide slope capture.

[Old Boeing Driver]

RA <2500 Feet
Flaps out of "up"
Glide slope capture.

If that is the case, then the GA button would have done nothing at 6,000 feet and no GS capture even if the flaps were not "0".

Thanks for the info.

[tdracer]

If that is the case, then the GA button would have done nothing at 6,000 feet and no GS capture even if the flaps were not "0".

Thanks for the info.

It's not particularly uncommon for RA to give temporary false readings due to reflections off of clouds.

[Old Boeing Driver]

It's not particularly uncommon for RA to give temporary false readings due to reflections off of clouds.

I'll take your word for the RA part, although I have never seen that.

But, what about the GS capture part?

[Airbubba]

The Northwest Boeing 720 accident I mentioned earlier they reached -2.8

Actually the CAB accident report said that the negative acceleration increased to a mean value of about - 2.8 g so the peak before the aircraft broke up may have been even higher.


https://cimg8.ibsrv.net/gimg/pprune.org-vbulletin/1778x226/nw_705_2_796c4cda35feb3fb6170b83ce72aa1f28585b92e.jpg

https://en.wikisource.org/w/index.php?title=File%3ACAB_Aircraft_Accident_Report%2C_North west_Airlines_Flight_705.pdf&page=4

Here are FDR plots from NW705 as published in the CAB report:


https://cimg9.ibsrv.net/gimg/pprune.org-vbulletin/2000x995/nw705_3a83b02dda428623037f24eec7bb225a43babcfc.jpg
Looking at the plots, maybe the -2.8 g number came from the average of the last two data points on the coarse foil technology FDR record.

[Sailvi767]

Their indicated airspeed to reach those negative G values were in excess of 400 knots. Not possible to get anywhere near those numbers at 210 to 230 knots. The first generation jets were also stronger and had higher VNE limits. We regularly cruised the 727 at 380 knots IAS at 16,000 feet on shuttle flights. VNE was 405. The A330 has a VNE around 330 knots and the 767 about 355.

[Airbubba]

Their indicated airspeed to reach those negative G values were in excess of 400 knots. Not possible to get anywhere near those numbers at 210 to 230 knots.

From the NTSB Investigative Update on March 12, 2019:

FDR, radar, and ADS-B data indicated that the airplane entered a rapid descent on a heading of 270°, reaching an airspeed of about 430 knots.

https://www.ntsb.gov/investigations/Pages/DCA19MA086.aspx (https://www.ntsb.gov/investigations/Pages/DCA19MA086.aspx)

[henra]

From the NTSB Investigative Update on March 12, 2019:
https://www.ntsb.gov/investigations/Pages/DCA19MA086.aspx (https://www.ntsb.gov/investigations/Pages/DCA19MA086.aspx)

Indeed. But that would be towards the end after they descended at steep angle and where they started pulling positive g's. I wouldn't be surprised to see >+4g somewhere in the readouts. The alleged -4g would have been at the beginning of the event. And that is where they were at ~230kts.

In the trace of the 720 accident you can see that at 220kts it produced ~ -2g, increasing to -2.8g @ ~280kts. The 720 had a ~20 - 25% lower wing loading than the 767, so should be able to produce ~25 - 30% more g at the same speed (especially in negative direction where no flaps/slats and tricked- out profiles will help).

When we extrapolate this data from the 720 accident we still end up with something around -2 to -2.5g maximum. We'll see the exact figures once the traces are published.
But I would expect something around -2g and +4 - 4.5g as maximum values.

[Airbubba]

Actually the CAB accident report said that the negative acceleration increased to a mean value of about - 2.8 g so the peak before the aircraft broke up may have been even higher.

Their indicated airspeed to reach those negative G values were in excess of 400 knots. Not possible to get anywhere near those numbers at 210 to 230 knots.

In the trace of the 720 accident you can see that at 220kts it produced ~ -2g, increasing to -2.8g @ ~280kts.

One of you may be right, I'm not so sure that wing loading at max gross weight is what determines the negative g capability of an airliner. The -4 g claim may indeed be from a typo in a forum posting.

It looks to me like the Atlas 767 must have been very light to climb to FL400 immediately over the Gulf of Mexico.

Freighters continue to crash at a significantly higher rate than passenger planes at U.S. carriers as we have often observed here over the years.

There are a lot of B-763's still flying, I would think that if there was an early indication of a mechanical failure in this Atlas mishap, the NTSB would say something by now.

[ironbutt57]

sudden updrafte, attempt to hold altitude..( vs attitude), hold that column forward...a bit longer than necessary

[NSEU]

If that is the case, then the GA button would have done nothing at 6,000 feet and no GS capture even if the flaps were not "0".

Thanks for the info.

The maintenance manual and my 767 engineering training notes show no reference to the radio altimeters. Also, the logic for flaps and G/S capture is "or", not "and".

Most references show "flaps out of up" (although I do have one reference which says flaps >1).