Continental TurboProp crash inbound for Buffalo
Join Date: Jul 2004
Location: Found in Toronto
Posts: 615
Likes: 0
Received 0 Likes
on
0 Posts
No, I believe the stick shaker/stick push system is fine the way it is.
It appears as though someone used force to over ride the system. If that is true then, aside from the low airspeed, that is what caused this crash.
My sense at this time is that we should concentrate on the factors that initiated the first oscillations of the aircraft be that wing stall or tail plain stall, etc.
I added the etc. because we really haven't an analysis of the DFDR so we must rely on inference at this point which tends to exclude others.
The consideration of ice is directly related to the crews own comments. However in my view the realtionship of such comments to the use of descriptive words "heavy" or "severe" needs to be examined in light of the expected crew response by their training.
Then there is the issue of the apparently low aircraft speed (valid or not) again we must rely on the DFDR and perhaps radar traces for confirmation. However its consideration as a significant contributor needs to also be examined since it seems to coincide with the timeline of flap selection which preceeded the initial oscilations.
Many arguments are expected regarding the crew performance, but these can not be judged in context unless they are compared against their training.
I find the comments on this board about the use of a wing light and observations of the boots shedding ice in other events to be intriguing at first. It draws my attention to crew workload.
1) If the preventive actions are to be based on a SOP directly related to the identification of "severe icing" ?
and
2)that definition tied to the certification of the aircraft?
then
3) is it not so that an observation of the ice boots showing that they are unable to clear the wing of ice over long periods of time, than the aircraft operation must presume "severe" icing encounter and avoidance selected by the crew?
4) is it possible that the crew was distracted by need to make a judgemental call by observing the wing resulting in them not following their airspeed?
I'm not a fan of having a crew have to make judgemental calls about the transitioning of "heavy" wing ice to "severe" wing icing by taking their eyes repeatedly off their instrument scans.
I'll wait and see which way my question marks lean in the points above
I added the etc. because we really haven't an analysis of the DFDR so we must rely on inference at this point which tends to exclude others.
The consideration of ice is directly related to the crews own comments. However in my view the realtionship of such comments to the use of descriptive words "heavy" or "severe" needs to be examined in light of the expected crew response by their training.
Then there is the issue of the apparently low aircraft speed (valid or not) again we must rely on the DFDR and perhaps radar traces for confirmation. However its consideration as a significant contributor needs to also be examined since it seems to coincide with the timeline of flap selection which preceeded the initial oscilations.
Many arguments are expected regarding the crew performance, but these can not be judged in context unless they are compared against their training.
I find the comments on this board about the use of a wing light and observations of the boots shedding ice in other events to be intriguing at first. It draws my attention to crew workload.
1) If the preventive actions are to be based on a SOP directly related to the identification of "severe icing" ?
and
2)that definition tied to the certification of the aircraft?
then
3) is it not so that an observation of the ice boots showing that they are unable to clear the wing of ice over long periods of time, than the aircraft operation must presume "severe" icing encounter and avoidance selected by the crew?
4) is it possible that the crew was distracted by need to make a judgemental call by observing the wing resulting in them not following their airspeed?
I'm not a fan of having a crew have to make judgemental calls about the transitioning of "heavy" wing ice to "severe" wing icing by taking their eyes repeatedly off their instrument scans.
I'll wait and see which way my question marks lean in the points above
LIS -- I will need to see more data before I can accept that the flying surfaces were clean enough that flying qualities were unaffected. Same applies to manual override of the stick push. Not to say that you are incorrect. The data will tell the story.
Guest
Posts: n/a
Chealander
Is exploring the pilot's possible attempt at defeating the StickPusher.
If PF did try to overcome the SP, it is counterintuitive and counter to training. Why? At 1600 AGL, nose down is also counterintuitive. Full Power and "nose level" might be a "reasoned" compromise under fear producing cues. If PF pulled, where are the feet? On the Pedals? What of previous training? Is "defeat the SS/SP" trained? In trying to pull "smoothly", do pitch excursions result? Who wants to "do nothing" while the SP points us at the deck? All guesswork, indeed.
If PF did try to overcome the SP, it is counterintuitive and counter to training. Why? At 1600 AGL, nose down is also counterintuitive. Full Power and "nose level" might be a "reasoned" compromise under fear producing cues. If PF pulled, where are the feet? On the Pedals? What of previous training? Is "defeat the SS/SP" trained? In trying to pull "smoothly", do pitch excursions result? Who wants to "do nothing" while the SP points us at the deck? All guesswork, indeed.
Supercharged PPRuNer
Join Date: Nov 2000
Location: Doon the watter, a million miles from the sandpit.
Posts: 1,183
Received 0 Likes
on
0 Posts
Flying the Q400 myself, Otto's story makes very interesting reading. There but for the grace of God . . .
SOP at my operator is that once the 'one to go' call is made, PF's hand goes to the power levers and stays there until altitude capture - for two very good reasons. Firstly to stop him fiddling with the pitch wheel (a well known 'gotcha') and secondly as a reminder to do something with the power once we level off.
A two axis autopilot and manual throttles can make for an interesting combination.
SOP at my operator is that once the 'one to go' call is made, PF's hand goes to the power levers and stays there until altitude capture - for two very good reasons. Firstly to stop him fiddling with the pitch wheel (a well known 'gotcha') and secondly as a reminder to do something with the power once we level off.
A two axis autopilot and manual throttles can make for an interesting combination.
Join Date: Jan 2008
Location: Denmark
Age: 79
Posts: 158
Likes: 0
Received 0 Likes
on
0 Posts
RatherBe Flying,
You wrote:
"Does the Q400 simply level off and start bleeding airspeed without any indications? Is setting this trap something that should be happening during high workload?"
I have not flown the Q400 - would have liked to though!
But anyway, I am sure that the answer to your first question is a yes.
I flew 3 types of commercial aircraft with autopilot without autothrottle (Caravelle III, F-27 and DC-9) for many years - of course you had to remember to add power at level off after a descent, but what is the problem? You have to remember a lot of things when flying - for example, you MUST remember to set local QNH before approach and landing - forgetting this could be far more dangerous than forgetting to add power.
As always when flying:
COCKPIT DISCIPLINE IS VITAL.
brgds
"Does the Q400 simply level off and start bleeding airspeed without any indications? Is setting this trap something that should be happening during high workload?"
I have not flown the Q400 - would have liked to though!
But anyway, I am sure that the answer to your first question is a yes.
I flew 3 types of commercial aircraft with autopilot without autothrottle (Caravelle III, F-27 and DC-9) for many years - of course you had to remember to add power at level off after a descent, but what is the problem? You have to remember a lot of things when flying - for example, you MUST remember to set local QNH before approach and landing - forgetting this could be far more dangerous than forgetting to add power.
As always when flying:
COCKPIT DISCIPLINE IS VITAL.
brgds
Last edited by grebllaw123d; 18th Feb 2009 at 15:53.
Join Date: Jul 2007
Location: Oxenfforrdde
Posts: 168
Likes: 0
Received 0 Likes
on
0 Posts
New video
Buffalo plane crash: amazing new pictures - MSN Video
Hope this hasn't already been posted.
video taken by teenager directly after crash
Hope this hasn't already been posted.
video taken by teenager directly after crash
Join Date: Jul 2007
Location: New York
Posts: 4
Likes: 0
Received 0 Likes
on
0 Posts
From the NY times, Feb 18, 2009:
Buffalo Crash Inquiry Sees Signs of Crew Error
By MATTHEW L. WALD
A re-creation of the last moments of the plane that crashed in Buffalo Thursday night, based on data from the “black boxes,” shows that the crew may have overreacted to an automatic system that was trying to protect the aircraft from flying too slowly and crashing from an aerodynamic stall.
By using the data to create a computer animation of the flight’s final seconds, investigators have developed a theory that after the automatic system pointed down the nose of the plane to generate speed, the crew may have overreacted by yanking back on the yoke and pointing the nose too high, according to a person familiar with the animation. The nose then plunged, and the airplane rolled and crashed into a house outside Buffalo, killing all 49 people on board and one man on the ground. The Wall Street Journal reported on its Web site on Tuesday evening that the investigators were looking at crew action as a possible cause of the crash.
The plane, Continental Connection Flight 3407, from Newark to Buffalo, was flying on auto-pilot. In that aircraft, a Bombardier Dash 8 Q400, that device can control altitude and course, but not the throttle, according to aviation experts. According to the National Transportation Safety Board, when the “stick-pusher,” which takes control of the plane and points the nose down, activated and the autothrottle kicked off, the crew tried to increase power. Apparently there was not enough altitude or time to recover control, however.
The stick-pusher may have activated at a speed higher than normal because it added a margin of safety to account for icing conditions, investigators say.
Investigators believe that the pilot at the controls was the captain, Marvin Renslow, who had begun flying the Dash 8 only in December. He previously flew a smaller turboprop, the Saab 340. That plane is subject to a problem in bad weather called tail plane icing, in which the airplane’s tail suddenly ceases to function in its role of applying downward pressure at the back to the plane, and holding the nose up. He may have been alert to the possibility of the nose dropping in icing for that reason. But Safety Board personnel at the crash scene said that the Dash 8 is not susceptible to the tail plane problem.
Buffalo Crash Inquiry Sees Signs of Crew Error
By MATTHEW L. WALD
A re-creation of the last moments of the plane that crashed in Buffalo Thursday night, based on data from the “black boxes,” shows that the crew may have overreacted to an automatic system that was trying to protect the aircraft from flying too slowly and crashing from an aerodynamic stall.
By using the data to create a computer animation of the flight’s final seconds, investigators have developed a theory that after the automatic system pointed down the nose of the plane to generate speed, the crew may have overreacted by yanking back on the yoke and pointing the nose too high, according to a person familiar with the animation. The nose then plunged, and the airplane rolled and crashed into a house outside Buffalo, killing all 49 people on board and one man on the ground. The Wall Street Journal reported on its Web site on Tuesday evening that the investigators were looking at crew action as a possible cause of the crash.
The plane, Continental Connection Flight 3407, from Newark to Buffalo, was flying on auto-pilot. In that aircraft, a Bombardier Dash 8 Q400, that device can control altitude and course, but not the throttle, according to aviation experts. According to the National Transportation Safety Board, when the “stick-pusher,” which takes control of the plane and points the nose down, activated and the autothrottle kicked off, the crew tried to increase power. Apparently there was not enough altitude or time to recover control, however.
The stick-pusher may have activated at a speed higher than normal because it added a margin of safety to account for icing conditions, investigators say.
Investigators believe that the pilot at the controls was the captain, Marvin Renslow, who had begun flying the Dash 8 only in December. He previously flew a smaller turboprop, the Saab 340. That plane is subject to a problem in bad weather called tail plane icing, in which the airplane’s tail suddenly ceases to function in its role of applying downward pressure at the back to the plane, and holding the nose up. He may have been alert to the possibility of the nose dropping in icing for that reason. But Safety Board personnel at the crash scene said that the Dash 8 is not susceptible to the tail plane problem.
Join Date: Jan 2008
Location: Sonoma, CA, USA
Age: 79
Posts: 143
Likes: 0
Received 0 Likes
on
0 Posts
Re 613
Grizzled.
Yes, my comment was a tad sarcastic having spent many hours in unforecast conditions.
Had to shut down an engine once because of a Butterfly strike, though.
Yes, my comment was a tad sarcastic having spent many hours in unforecast conditions.
Had to shut down an engine once because of a Butterfly strike, though.
Join Date: Sep 2006
Location: East Yorkshire
Posts: 453
Likes: 0
Received 0 Likes
on
0 Posts
Am I right in thinking that the landing gear was not down when flying clean at 137knts.
The reason i ask is that the a/c i fly J41 the low speed horn sounds if the speed goes below 150knts. That really gets your attention.
Surely they must have had some warning !
The reason i ask is that the a/c i fly J41 the low speed horn sounds if the speed goes below 150knts. That really gets your attention.
Surely they must have had some warning !
Join Date: Nov 2006
Location: SoCalif
Posts: 896
Likes: 0
Received 0 Likes
on
0 Posts
PTH: "The DC9, which can get into a deep stall, which is why vortillions or chin strakes are fitted DOES NOT HAVE A PUSHER. (there is a unique hydraulic ram to ensure ''nose down'' elevator input in a deep stall, but it is not in any way a stick pusher)"
"It is T TAIL SWEPT WING JET."
It is my understanding that the first DC-9 to get chin strakes was the -50, the first stretch, but were for stability in normal flight regimes. The DC9-80, aka MD-80, got a stick pusher fitted only after deep stall testing, in which they had to deploy a chute to recover on one occasion.
BTW, what's the difference between a stick pusher and a ram?
Wasn't a BAC1-11 the first T-tail jet to get in deep stall and crash?
GB
"It is T TAIL SWEPT WING JET."
It is my understanding that the first DC-9 to get chin strakes was the -50, the first stretch, but were for stability in normal flight regimes. The DC9-80, aka MD-80, got a stick pusher fitted only after deep stall testing, in which they had to deploy a chute to recover on one occasion.
BTW, what's the difference between a stick pusher and a ram?
Wasn't a BAC1-11 the first T-tail jet to get in deep stall and crash?
GB
Join Date: Jan 2008
Location: Denmark
Age: 79
Posts: 158
Likes: 0
Received 0 Likes
on
0 Posts
Wellington Bomber,
From info given in link in post #520:
"One minute prior to the end of the recording the airplane was travelling at a calibrated airspeed of 134 knots, the gear was selected down, 20 seconds later the flaps were selected to 15 degrees."
An existence of a "low speed horn" has not been mentioned before in this thread.
"One minute prior to the end of the recording the airplane was travelling at a calibrated airspeed of 134 knots, the gear was selected down, 20 seconds later the flaps were selected to 15 degrees."
An existence of a "low speed horn" has not been mentioned before in this thread.
Join Date: Jan 2001
Location: Dallas, TX USA
Posts: 739
Likes: 0
Received 0 Likes
on
0 Posts
Another possible scenario....
I'm still not clear on the exact sequence of events in that final minute, so this is just speculation on my part.
First, the 190kts speed in icing is probably required to maintain low AOA, to keep ice from forming on the bottom of the wing, a common practice in icing conditions.
The NTSB has stated that the normal Vref speed at the aircraft's weight should have been 119kts. Add the +20kt ice switch, and this is taken to 139kts. The very fact that this switch exists, suggests that ice accumulation on the wings was considered possible, thus the added speed margin for the stall.
Even though the NTSB has stated the calibrated airspeed was 134kts when the landing gear was deployed, they've stated they are still not too confident yet in this number.
When the landing gear was lowed 1 minute prior to impact, the stick shaker didn't activate until 34 seconds later, during the flap transit from 5 to 10 degrees. This suggests to me the pilots probably adjusted airspeed after the landing gear deployment.
During the flap transit the stick shaker and pusher activated. This may have been triggered by the airspeed briefly falling below the +20kt added margin (just prior to another expected power adjustment by the pilots), rather than an actual aerodynamic stall. At least this seems possible based on what we know so far.
What happened next is a mystery. We know the stick pusher disconnected the autopilot, which was immediately followed by a 31 degree pitch up. At their airspeed at that moment, I think it was this first severe pitch up that sealed their fate, creating a real (and deadly serious) aerodynamic stall.
We know the stick pusher should have lowered the nose, but it didn't. Why? The great mystery (for me) is what could have raised the nose instead? Some possibilities are:
I'm still not clear on the exact sequence of events in that final minute, so this is just speculation on my part.
First, the 190kts speed in icing is probably required to maintain low AOA, to keep ice from forming on the bottom of the wing, a common practice in icing conditions.
The NTSB has stated that the normal Vref speed at the aircraft's weight should have been 119kts. Add the +20kt ice switch, and this is taken to 139kts. The very fact that this switch exists, suggests that ice accumulation on the wings was considered possible, thus the added speed margin for the stall.
Even though the NTSB has stated the calibrated airspeed was 134kts when the landing gear was deployed, they've stated they are still not too confident yet in this number.
When the landing gear was lowed 1 minute prior to impact, the stick shaker didn't activate until 34 seconds later, during the flap transit from 5 to 10 degrees. This suggests to me the pilots probably adjusted airspeed after the landing gear deployment.
During the flap transit the stick shaker and pusher activated. This may have been triggered by the airspeed briefly falling below the +20kt added margin (just prior to another expected power adjustment by the pilots), rather than an actual aerodynamic stall. At least this seems possible based on what we know so far.
What happened next is a mystery. We know the stick pusher disconnected the autopilot, which was immediately followed by a 31 degree pitch up. At their airspeed at that moment, I think it was this first severe pitch up that sealed their fate, creating a real (and deadly serious) aerodynamic stall.
We know the stick pusher should have lowered the nose, but it didn't. Why? The great mystery (for me) is what could have raised the nose instead? Some possibilities are:
- The center of lift somehow moved forward (instead of backwards) during the flap transit, due to icing maybe?
- A rather serious stabilizer out of trim condition (due to ice) was suddenly manifested following autopilot disconnect.
- Ice accumulation causing some other unknown aerodynamic behavior?
- The pilot(s) hauled back on the yoke overriding the stick pusher for some reason. This seems very unlikely to me, given that their other actions following the upset (rapidly engines to full power, gear and flaps to retract) seemed appropriate.
Last edited by Flight Safety; 18th Feb 2009 at 17:28.
Otto
I have investigated two such incidents in USAF heavy jets that stalled and were recovered-once with just about no height left to lose! Both crews were experienced, one did have a fatigue issue. The fact that not much happens in the first 30 knots of airspeed loss-happens slowly, not much ambient noise change-means the loss gets serious before someone notices, if they do. Distractions are killers and my money is on their attention being focused on the icing and not on the flying. Slick, heavy jets are worse in this mode, takes time for the change to start and then very hard to arrest.
But, we have a long ways to go on the NTSB investigation.
GF
I have investigated two such incidents in USAF heavy jets that stalled and were recovered-once with just about no height left to lose! Both crews were experienced, one did have a fatigue issue. The fact that not much happens in the first 30 knots of airspeed loss-happens slowly, not much ambient noise change-means the loss gets serious before someone notices, if they do. Distractions are killers and my money is on their attention being focused on the icing and not on the flying. Slick, heavy jets are worse in this mode, takes time for the change to start and then very hard to arrest.
But, we have a long ways to go on the NTSB investigation.
GF
Join Date: Apr 2000
Location: FL, USA
Posts: 357
Likes: 0
Received 0 Likes
on
0 Posts
Flight Safety: '...gear and flaps to retract...upset...appropriate...' errr, not quite sure if you meant that the way it reads...I was always taught to leave the configuration changes until recovery complete, unless it is obvious that the selection of configuration caused the upset...( I once had an L-1011 try to start a rapid roll on a maint. test flt due to incorrect re-assembly of the gear microswitches...gnd spoilers deployed on one side only in flt...pulled the gear immediately back up...interesting...but plenty of altitude and VFR conditions).
Graybeard;
The DC9-30 series did not have nose-strakes/vortilons.
Yes, the BAC111 was the first deep-stall accident. That is the graphic still used in D.P.Davies' book to illustrate the phenomenon.
The DC9-30 series did not have nose-strakes/vortilons.
Yes, the BAC111 was the first deep-stall accident. That is the graphic still used in D.P.Davies' book to illustrate the phenomenon.
Join Date: Feb 2008
Location: Maine
Posts: 34
Likes: 0
Received 0 Likes
on
0 Posts
From Post 290 regarding the threshold for the Stick Shaker System:
The Stick Pusher System, Stick Shaker System and the flap auto pitch trim system.
The Stall protection system consists of 2 SPMs or Stall Protection Modules and uses the following parameters to calculate when the a/c is near a stall condition:
AoA data
Flap Position
EAS
Engine Torque
Icing Status
Both SPMs are used to calculate the Stick Pusher operating angle using the following parameters:
AoA
Flap Position
EAS
Power Lever Angle
Condition Lever Angle
Icing Status.
The SPMs calculate when to operate the Shaker & Pusher systems and sends a signal to the Automatic Flight Control System to disengage the autopilot, also for Stick Pusher activation it uses an average of the 2 AoA vane inputs.
Stick pusher operates under these conditions
CAS is less than 215kts
Altitude is +200ft AGL
Stick Pusher shutoff annunciator switch is not in the off position.
The Stick Pusher System, Stick Shaker System and the flap auto pitch trim system.
The Stall protection system consists of 2 SPMs or Stall Protection Modules and uses the following parameters to calculate when the a/c is near a stall condition:
AoA data
Flap Position
EAS
Engine Torque
Icing Status
Both SPMs are used to calculate the Stick Pusher operating angle using the following parameters:
AoA
Flap Position
EAS
Power Lever Angle
Condition Lever Angle
Icing Status.
The SPMs calculate when to operate the Shaker & Pusher systems and sends a signal to the Automatic Flight Control System to disengage the autopilot, also for Stick Pusher activation it uses an average of the 2 AoA vane inputs.
Stick pusher operates under these conditions
CAS is less than 215kts
Altitude is +200ft AGL
Stick Pusher shutoff annunciator switch is not in the off position.
Join Date: Jan 2009
Location: alameda
Posts: 1,053
Likes: 0
Received 0 Likes
on
0 Posts
graybeard
the DC9 has a hydraulic ram to ensure nose down elevator because the elevators are aerodynamically powered ...not hydraulic, and there might be some scenarios where air flow would be insufficient.
IT IS NOT A stick pusher. If th pilot wants to activate this ram, you must manually push the yoke full forward...indeed the copilot checks it prior to each takeoff...you may notice if taxiing behind an MD80 or DC9 that the elevators are frequently flip flopped...one up, one down...hit the yoke forward and the ram puts them both down.
AS to strakes...yes on the later models...they replaced the vortilions on the earlier DC9's...these little things on the wing sort of looked like a tiny drop tank, but improved air flow to help avoid stall problems. I don't think the MD80 has a pusher.
I think the BAC111 and the Trident are examples of two T tail jets with stick pushers.
PJ2...the DC9 30 most certainly has vortillions...they look like little fuel drop tanks on the front part of each wing.
IT IS NOT A stick pusher. If th pilot wants to activate this ram, you must manually push the yoke full forward...indeed the copilot checks it prior to each takeoff...you may notice if taxiing behind an MD80 or DC9 that the elevators are frequently flip flopped...one up, one down...hit the yoke forward and the ram puts them both down.
AS to strakes...yes on the later models...they replaced the vortilions on the earlier DC9's...these little things on the wing sort of looked like a tiny drop tank, but improved air flow to help avoid stall problems. I don't think the MD80 has a pusher.
I think the BAC111 and the Trident are examples of two T tail jets with stick pushers.
PJ2...the DC9 30 most certainly has vortillions...they look like little fuel drop tanks on the front part of each wing.
Join Date: Nov 2002
Location: USA
Posts: 49
Likes: 0
Received 0 Likes
on
0 Posts
RBF
Apart from the 'one to go' beep there is no other audio warning of level off, visual cues are ALT* on the screen, but you have to be concentrating on it.
Speed bleeds off pretty quickly particularly on selecting gear and flap of course.
31 degrees nose up does appear, from the latest information, to be pilot input forceful enough to overpower the pusher - which we could guess was the response to what was believed to be a tail stall.
Interesting, winter training programmes have included lots about tail-stall (in the UK this year more than before, with the NASA vids forming part of our winter briefing) despite Bombardier's assertion from millions of hours of experience that the Q series is 'not susceptible to tail-stall'.
If the PF was indeed responding to his wing-stall circumstances using information about ice and tail-stall this may be another reminder that the "Diagnose" part of a crisis has to be absolutely objective however bad the moment might seem.
Apart from the 'one to go' beep there is no other audio warning of level off, visual cues are ALT* on the screen, but you have to be concentrating on it.
Speed bleeds off pretty quickly particularly on selecting gear and flap of course.
31 degrees nose up does appear, from the latest information, to be pilot input forceful enough to overpower the pusher - which we could guess was the response to what was believed to be a tail stall.
Interesting, winter training programmes have included lots about tail-stall (in the UK this year more than before, with the NASA vids forming part of our winter briefing) despite Bombardier's assertion from millions of hours of experience that the Q series is 'not susceptible to tail-stall'.
If the PF was indeed responding to his wing-stall circumstances using information about ice and tail-stall this may be another reminder that the "Diagnose" part of a crisis has to be absolutely objective however bad the moment might seem.
Join Date: Jan 2001
Location: Dallas, TX USA
Posts: 739
Likes: 0
Received 0 Likes
on
0 Posts
RRAMJET, you may be correct about the landing gear, but it would make sense in icing if you thought the flaps triggered the upset. For now, I don't think the landing gear had anything to do with the accident, unless it's related to overall airspeed decline.