Go Back  PPRuNe Forums > Flight Deck Forums > Tech Log
Reload this Page >

The vital importance of high altitude stall recovery training in simulators

Wikiposts
Search
Tech Log The very best in practical technical discussion on the web

The vital importance of high altitude stall recovery training in simulators

Thread Tools
 
Search this Thread
 
Old 6th Oct 2014, 10:01
  #21 (permalink)  
 
Join Date: Jun 2009
Location: Germany
Age: 71
Posts: 776
Received 3 Likes on 1 Post
centaurus
From simulator experience it takes very little nose down attitude below the horizon before it is obvious you are out of the stall and the next thing is to decide when to level out.
Thank you for your well placed answer. Let me add to the quoted point one more comment: The amount how much the pitch has to change depends on the AOA when beginning pitch down. If recovery is initiated just after sounding of the stall warning, then your statement is correct. But actually you might not be in a stall already, as the system sounds the horn prior to stall, it warns in advance of an imminent stall. At that point the reduction of AOA by few degrees is obviously sufficient. In a fully developped stall the difference between pitch attitude and flightpath, which is called AOA, will be much greater, AF447 had AOA in excess of 60°. After reducing the pitch by 5, 10, or even 20° would still leave a high enough AOA to keep the jet stalled. As AOA is not readily displayed in most aircraft to stick to some kind of degrees of pitch nose down is dangerous and will prolong recovery or even prevent recovery like AF447 showed.

Edit: Another point to note would be that different aircraft behave different, and upsets might happen in some degraded modes of aircraft systems or cause those degradation during the upset (degradation of flight LAW's, degradation of autoflight systems like autothrust, autotrim). The effects of thrust changes remain screened by such fully functional systems, the effect therefore unknown until those systems like autotrim or dampening system fails.

Stall recovery training and stall recovery procedures have to take acount of those possible degradations and should therefore be oriented on the basic aircraft handling.

Last edited by RetiredF4; 6th Oct 2014 at 11:23.
RetiredF4 is offline  
Old 6th Oct 2014, 10:24
  #22 (permalink)  
 
Join Date: Apr 2005
Location: Australia
Posts: 1,414
Likes: 0
Received 0 Likes on 0 Posts
After reducing the pitch by 5, 10, or even 20° would still leave a high enough AOA to keep the jet stalled
Having studied the posts by the OP, he made the point several times that it was necessary to get the nose below the horizon which means a descent in order to increase speed which is the same as lowering the angle of attack in my book. You seemed to have mis-understood his point judging by the highlighted text. It was never implied that lowering the nose only a few degrees from an extremely high angle of attack (as you appear to state) to a still stalled angle of attack was sufficient to recover from a stall.
A37575 is offline  
Old 6th Oct 2014, 10:34
  #23 (permalink)  
 
Join Date: Apr 2005
Location: Australia
Posts: 1,414
Likes: 0
Received 0 Likes on 0 Posts
And unbelievably it was flown by a TRE with 10,000 hours on type. Presumably he had flown approaches to stalls in the simulator,
You would be dismayed at the number of airline pilots who have not carried out or being trained for high altitude (30,000 to 41,000ft) stall recovery techniques in the simulator. This may be because 15,000ft is considered by some simulator instructors as sufficient for high altitude.
A37575 is offline  
Old 6th Oct 2014, 11:13
  #24 (permalink)  
 
Join Date: Jun 2009
Location: Germany
Age: 71
Posts: 776
Received 3 Likes on 1 Post
A37575
Having studied the posts by the OP, he made the point several times that it was necessary to get the nose below the horizon which means a descent in order to increase speed which is the same as lowering the angle of attack in my book. You seemed to have mis-understood his point judging by the highlighted text.
Thank you for taking part in the discussion.

The problem of the misunderstanding is imho that the title of the thread is

"The vital importance of high altitude stall recovery training in simulators"

and the discussion centers around the topic of an "approach to stall recovery"

In the situation the thread title suggests the AOA might be anything from AOA being slightly greater than stall AOA to 60° degrees above stall AOA, whereas the statements i'm commenting on cover only the area just prior and just above stall AOA.

Now if the title of the thread is misplaced and the discussion likes to concentrate on the approach to stall situation and thus the avoidance of a stall situation, then the title should be changed or posters should make their point clear when discussing only the latter.

A37575
It was never implied that lowering the nose only a few degrees from an extremely high angle of attack (as you appear to state) to a still stalled angle of attack was sufficient to recover from a stall.
I might have a language barrier there, but here is the original wording, bolding by me.

Quote:
centaurus
From simulator experience it takes very little nose down attitude below the horizon before it is obvious you are out of the stall and the next thing is to decide when to level out.
That is true as a stall avoidance maneuver or a stall with very little higher AOA (just after stall warning sounded), but not to recover from a developped stall.

Most students were good at approach to stall recoveries, because they choose to initiate recovery the earliest moment possible (which they are supposed to do, military jet) with an automated response stick forward, firewall the engines, stick back), but when they had to wait a bit longer they lowered the nose not long enough, pulled back on the stick too early and went into a secondary stall. All at safe altitude, so no danger.

There is only one way to do it when sufficient altitude is available, lower the nose not a little bit, lower it not to a specified amount of degrees below the horizon, but lower it until stall warning ceases = AOA below stall warning AOA (hopefully not like AF447) and speed has been built up enough for recovery maneuver. That is the point im trying to make.

Last edited by RetiredF4; 6th Oct 2014 at 22:46.
RetiredF4 is offline  
Old 6th Oct 2014, 12:32
  #25 (permalink)  
 
Join Date: Oct 2007
Location: Here and there
Posts: 386
Likes: 0
Received 1 Like on 1 Post
Latest report on unusual attitude scary recovery. It really makes you wonder about the paucity of training today's inexperienced first officers are given in simulators. Accent on automation is wrong. Accent on pure flying skills are needed well before automation is brain-washed into their heads forever:


An All Nippon Airways Co. co-pilot who accidentally put his plane into a violent dive and roll in 2011 came closer to losing control of the Boeing Co. 737 than previously believed, according to data released by investigators.

The Japan Transport Safety Board's final report about the serious incident over the Pacific Ocean reveals there were multiple warnings of an impending aerodynamic stall, while the plane carrying 117 people exceeded its maximum operating speed a number of times. The report said the jet also exceeded its structural-load limit, or the most stress the aircraft is expected to experience in service.

After the co-pilot mistakenly operated a rudder-control switch at 41,000 feet, instead of a different switch that unlatched the cockpit door, the jetliner plummeted 1.2 miles in slightly more than 30 seconds and briefly flew nearly upside down. The recently released report provides some new details about the extent of the danger.

Two flight attendants were pinned to the cabin floor and were slightly injured. One attendant told investigators that after she felt "earthquake-like vertical shaking," she slumped to the floor on her knees from downward forces so strong she couldn't raise her arm.

The report is likely to refocus industrywide interest in high-altitude upsets, or incidents in which jetliners slow dramatically or end up with the nose or wings at unusually steep upward or downward angles.

The 100-page document also highlights the 38-year-old co-pilot's delayed and confused response, which investigators attributed to gaps in training, undue reliance on automation and seeming anxiety about quickly letting the captain back into the cockpit. According to the report, "excessive dependence on autopilot" exacerbated "lack of full awareness about the need to monitor" flight controls. The co-pilot couldn't recall the stick shaker's activation, it said.

The report said the co-pilot failed to recognize there was a problem for 17 seconds, and then alternately pushed forward and pulled back on the controls. The captain, returning from a bathroom break, was locked out of the cockpit while the plane nose-dived and executed back-to-back rolls in opposite directions. The maneuvers lasted about 90 seconds, though passengers may not have fully realized what was happening because it was dark outside.

An earlier report disclosed many of the co-pilot's errors and the plane's excessive speed. But it suggested there was only a single activation of the "stick shaker," a last-ditch safety alert that a plane is about to lose lift and may be on the verge of a crash.

Investigators found that the plane's excessive-speed warning was also activated more than once during the episode.

All Nippon said it "took action following the incident to prevent any recurrence, and we are continuing to take additional measures in line with the report's recommendations."

The co-pilot, who hasn't been identified, remains an All Nippon employee but isn't currently assigned to flight duties, according to the carrier.

Japan's safety board urged the carrier to ensure compliance with single-pilot operation protocols and enhance training to cope with high-altitude stall warnings.

The report should get a lot of attention in the industry since it details "a real poster-child event about complacency" and an inappropriate response, said Rory Kay, an ex-737 captain and former senior pilot-union safety official who now works as a training captain on Boeing jets. Pilots should "treat high-altitude stalls totally differently than those at low altitude," he said. Thinner air up high means jet engines take longer to rev up power, and a sudden upward nose command is more likely to create control problems.

Once the plane was back on the correct course and altitude, neither pilot disclosed the extent of the problem to air-traffic controllers, who in turn never pushed for answers. The jet, en route to Tokyo's Haneda airport from Okinawa, made a normal landing.

The crew's failure to promptly alert mechanics about the specifics of the event, according to the report, meant the plane continued to carry passengers for four more flights before all required inspections for possible structural damage were conducted. The checks didn't detect any problems.

The nose of the ANA jetliner was pointed 35 degrees down from level flight, a much steeper angle than passengers typically experience. The plane was subjected to forces nearly 2.7 times the force of gravity.

The report determined that the co-pilot's training didn't include dealing with high-altitude flight upsets or any "upset recovery training accompanied with a stall warning."

Before the upset, the ANA co-pilot failed to follow company procedures requiring him to put on his oxygen mask when he was left alone at the controls.

Investigators found that the co-pilot's initial preoccupation with opening the cockpit door prevented using "calm judgment" to continue monitoring controls and then to fly the plane manually. Upon returning to the cockpit, according to the report, the captain took over the controls because he found the co-pilot's condition "fairly unsettled."

The ANA event occurred two years after the crash of Air France AF.FR -1.35% Flight 447, a widebody Airbus A330 that stalled at high altitude. That crew's failure to understand and counteract a rapid descent ended in the death of all 228 people aboard. The crash was a seminal event that sparked world-wide emphasis on stall-recovery training, particularly at cruise altitudes.

Many airlines quickly revised simulator training to include lessons learned from the Air France accident.

There have been more recent instances of high-altitude upsets. An American Airlines Boeing 757, cruising at 35,000 feet over Venezuela on August 30, lost some 7,500 feet in about one minute. There were no injuries and the flight continued on to Brazil. An American spokeswoman couldn't immediately be reached for comment.

Japanese investigators said the ANA co-pilot, who was relatively inexperienced, had less than 2,800 flight hours in another 737 version and only 197 hours in the Boeing 737-700 involved in the incident. As previously reported, the cockpit-door unlock switch on his earlier aircraft was similar to the location, size and shape of the rudder switch on the 737-700 model.
Judd is offline  
Old 6th Oct 2014, 12:42
  #26 (permalink)  
 
Join Date: Jun 2006
Location: Australia
Posts: 1,186
Likes: 0
Received 0 Likes on 0 Posts
Japanese investigators said the ANA co-pilot, who was relatively inexperienced, had less than 2,800 flight hours in another 737 version and only 197 hours in the Boeing 737-700 involved in the incident

The hours do not indicate a "relatively inexperienced" pilot although it is highly probable that he was hired initially with the usual 200 cadet flying hours straight from flying school then fed a diet of automation from the time he first entered a 737 simulator.
Tee Emm is offline  
Old 6th Oct 2014, 13:43
  #27 (permalink)  
 
Join Date: Feb 2000
Location: UK
Posts: 1,683
Likes: 0
Received 137 Likes on 87 Posts
Years ago a well presented AA training video on unusual attitude recovery pointed out that AOA was a standard parameter in FDRs, but not on the Flight Deck ... so why NOT available as standard instrumentation? ... it still isn't , so why NOT?
Cornish Jack is offline  
Old 6th Oct 2014, 14:29
  #28 (permalink)  
 
Join Date: May 2013
Location: London
Posts: 121
Likes: 0
Received 0 Likes on 0 Posts
I would. Air Algerie MD80 apparently stalled at high altitude in cloud during storm penetration. The accident appears to have developed as a loss of airspeed awareness leading to a fully developed stall and roll off departure to the port side with post stall gyrations. The nose then became planted way nose down, left wing down as the stall broke, but they never recovered from the nose down-wing down attitude
Sounds more like failing to fly an attitude rather than an altitude. Not really what I mean. Any phase of flight runs the risk of stall of the crew are ignorant of the correct technique or where monitoring is lacking.

Where the crew are exercising good judgement, technique and awareness and still get horribly close to the wrong end of the amber bar. If that is the case, just imagine the lacklustre crew in the same situation.
Kefuddle is offline  
Old 6th Oct 2014, 16:06
  #29 (permalink)  
 
Join Date: Jun 2010
Location: Canada
Age: 37
Posts: 382
Likes: 0
Received 0 Likes on 0 Posts
...or there are experienced pilots out there with woefully inadequate knowledge of the correct stall avoidance and recovery techniques.
I think this is part of the problem.

Regulators and airlines need to mandate more theory and more practical training with regard to upset/stall experiences. There needs to be more aerodynamics training and more actual flying (airplane or simulator) experience.

It doesn't help that there are airlines in Canada who still teach their pilot to power out of the approach to stall exercise, disregarding an AC that has been out since 2005, and has since been replace with a more detailed AC, that says to treat any approach to stall as an actual stall and use elevator to reduce the AOA. It states that multiple times and it also states multiple times that altitude loss is secondary to stall recovery. It couldn't be clearer - powering out of stalls with zero altitude loss is NOT the way it should be taught but that's the way you get a 4 (highest mark) on the flight test for that exercise.

For those interested, here is the current AC: https://www.tc.gc.ca/eng/civilaviati...-700-1793.html

Japanese investigators said the ANA co-pilot, who was relatively inexperienced, had less than 2,800 flight hours in another 737 version and only 197 hours in the Boeing 737-700 involved in the incident
I've found throughout my relatively short aviation career that hours mean virtually nothing. You could take a group of 2000 hour pilots and they'd all perform wildly different from each other. You could take a group of 5000 hour pilots and do the same with the same results. And so on. The only thing you'll see is that with experience, you'll notice an improvement in skill and decision making but when looking at a specific pilot, it's impossible to judge a pilot's skill and decision making solely based on their flight hours.

but when they had to wait a bit longer they lowered the nose not long enough, pulled back on the stick too early and went into a secondary stall.
This is an issue with the way they train now. They'll set the airplane up at about 10,000' to 14,000' (most of the time in VMC with smooth flying conditions), in level flight, then slow down at a constant rate while trimming to about 20 knots above the stall speed, then with hands already on the control column and power levers, they will slam the power levers forward as soon as the stall horn sounds and then push a bit on the control column so that they don't climb immediately (because it will if you don't push enough after adding full power). The airplane on these training flights is at the lowest weight that you'll ever encounter while flying operationally and at it's forward CG limit.

That is the absolute easiest possible recovery you could have for a stall event! And that's all that's required... and they're still teaching it wrong. It's frustrating when discussing this with pilots at the airline because they dismiss the AC... what experience and training do they have to say that what is in that AC is not correct?

There is only one way to do it when sufficient altitude is available, lower the nose not a little bit, lower it not to a specified amount of degrees below the horizon, but lower it until stall warning ceases
Before you add power.

Teaching to power out of a stall event (stall horn, stick shaker/pusher, or actual aerodynamic stall) is the opposite of what you should do, especially at high altitude. Rapid power advances may cause asymmetric thrust which, if you are stalled, can put you into a spin.

I think it should be taught to lower the nose until the horn goes off, then smoothly add power while keeping the nose at the same attitude. By the time you've added full power (~2-3 seconds) you should be at a speed to start smoothly coming out of the dive. Adding power while there is still an indication of a stall shouldn't be done, especially at high altitude where longitudinal stability is reduced. This is a generic recovery procedure and each airplane will handle a bit differently but the basics are all the same.

Power is only used to reduce the altitude loss after a stall. How do you think a glider recovers from a stall?

Thoughts on that, RetiredF4?
italia458 is offline  
Old 6th Oct 2014, 17:45
  #30 (permalink)  
 
Join Date: Jun 2009
Location: Germany
Age: 71
Posts: 776
Received 3 Likes on 1 Post
italia 458
Thoughts on that, RetiredF4?
Thank you for your excellent post, i completely agree and have nothing to add.
RetiredF4 is offline  
Old 6th Oct 2014, 20:51
  #31 (permalink)  
 
Join Date: Dec 2003
Location: Tring, UK
Posts: 1,838
Received 2 Likes on 2 Posts
There is only one way to do it when sufficient altitude is available, lower the nose not a little bit, lower it not to a specified amount of degrees below the horizon, but lower it until stall warning ceases = AOA below stall warning AOA (hopefully not like AF447) and speed has been built up enough for recovery maneuver. That is the point im trying to make.
Amen to that, bro’.

Having done some sim high-altitude exercises fairly recently, I think the absolute fidelity of the simulator was secondary to the emphasis on the above.

Hopefully, none of us will ever be in a situation where we have to recover from a low-speed LoC event but exposure to even generic training shows that a) you need to reduce the AoA in a positive and determined way and b) you’re going to lose a *lot* of altitude in the process but this is absolutely required to survive. Whether this is 4,000' or 14,000’ is less relevant - it takes what it takes...

As a someone who came into power flying from gliding, I always found the “old” method of stall recovery rather weird and difficult to get to grips with: now it’s just back to what I always did!
FullWings is online now  
Old 6th Oct 2014, 22:22
  #32 (permalink)  
 
Join Date: Aug 2007
Location: The Netherlands
Age: 67
Posts: 288
Likes: 0
Received 0 Likes on 0 Posts
AOA Gauge

Please be aware that although most civvie (big) jets do not have a real AOA gauge on the flight deck, they do have an indication of stall AOA in the form of the red bar(ber pole) on the low end side of the speed scale.

Yes, it may seem that that barberpole is only an indication of minimum speed, but that is not true: its position on the speed scale is driven by AOA - pull g and the red bar comes up (showes a higher minimum speed), unload and the red barberpole moves down to indicate a lower minimum speed. In other words, unload untill your actual speed is outside of the barberpole region and during pillout, do not let the barberpole grow up into your actual speed.

And about simulator fidelity - perhaps the motions are not exactly how they would be in the real aircraft, but manipulating the controls to make the SIM do what you want it to do is still a better learning experience than to have no maneuvering experience at all - real stall or stick shaker situations often show that pilots (without maneuvering experience) are too hesitant to make necessary control movements if those required movements are unusualy large or require unusuallly large forces, i.e. are outside the scope of daily routine tiny well trimmed out stick and or rudder applications. And before someone brings it up - the control movements that brought down the AA A-300 out of JFK were not "required control applications", they were excessive and totally not needed during the wake encounter.
EMIT is offline  
Old 7th Oct 2014, 01:07
  #33 (permalink)  
 
Join Date: Jun 2006
Location: Australia
Posts: 1,186
Likes: 0
Received 0 Likes on 0 Posts
Appreciate if anyone could provide the direct link to the Japan Transport Safety Board Final Report (English version) on the All Nippon Airlines B737-700 incident on 7 September 2011 described in this thread. Have tried Google but no Final Report. Thanks in advance
Tee Emm is offline  
Old 7th Oct 2014, 02:02
  #34 (permalink)  
 
Join Date: Dec 2002
Location: Where the Quaboag River flows, USA
Age: 71
Posts: 3,411
Received 3 Likes on 3 Posts
Nothing wrong with this generation if pilots a thousand hours solo in a "steam gauge" single seat fighter wouldn't fix. You learn about AOA, trim, low and high speed aerodynamics there, all the while scaring the hell out of you.
galaxy flyer is offline  
Old 8th Oct 2014, 13:14
  #35 (permalink)  
 
Join Date: Jan 2006
Location: US
Posts: 2,205
Likes: 0
Received 0 Likes on 0 Posts
ZFT - ultimately we agree - pilots as end users don't have accurate post stall simulator modeling. It doesn't matter that it could be done as an end user all I need to know is that the simulator doesn't accurately modeling how the airplane will respond.
misd-agin is offline  
Old 8th Oct 2014, 13:36
  #36 (permalink)  
 
Join Date: Jan 2006
Location: US
Posts: 2,205
Likes: 0
Received 0 Likes on 0 Posts
DL and AA 737's and 777's have AOA gauges.


Checklists have had verbiage using the AOA gauges removed because they were poorly written AND Boeing said the procedures hadn't been flight tested AND in certain circumstances the AOA gauges could be inoperative or give misleading information.


The generic low altitude loss of airspeed training had lead to bad habit patterns that did not transfer to high altitude/low energy flight regime.


So having generic "put the nose this low" that havn't been vetted by flight test is meaningless.


Example - doing loss of airspeed at altitude in the simulator. CKA, new to fleet, didn't know I'd been involved in the issue for over two years. So we do the high altitude upset and run the airplane totally out of airspeed. Very similar to AF 447 in that we were several thousand feet above MAX ALT in a nose high attitude. Ugly. CKA loved the FPV and had come up with a generic 'rule of thumb' that had worked for him up until then - "set the FPV 5(?) degrees nose low and you'll fly out". So we're at twice his target FPV attitude and the plane's barely flying - AOA barely below stick shaker. CKA - "no, no. Keep the FPV at X". Tiniest reduction in back pressure results in instantaneous stick shaker (anyone doing a high AOA knife fight know's what I'm talking about). Unload, go back to twice his target FPV attitude(double wasn't a goal, just what was required to make the plane fly), and we eventually recover.


CKA "huh, that's always worked before." Ugh.


During the event N1's were limited to 75% due to the nature of the failure(s). Didn't bother deselecting EEC's/ELC's because AOA, and not engine thrust, was the primary focus needed to fly the aircraft safely.


Debrief sent to stop teaching unofficial techniques that didn't cover all areas of the flight envelope.


Low altitude techniques, like the AB stall recovery training in Normal Law, can be dangerous in other flight regimes or non normal flight law.


Basic flying skills still are the #1 priority.
misd-agin is offline  
Old 8th Oct 2014, 19:48
  #37 (permalink)  
 
Join Date: Feb 2005
Location: Correr es mi destino por no llevar papel
Posts: 1,422
Received 1 Like on 1 Post
Originally Posted by Centaurus
Air Algerie MD80 apparently stalled at high altitude in cloud during storm penetration.
It did not penetrate active storm cell at any time before the stall.

Originally Posted by italia458
powering out of stalls with zero altitude loss is NOT the way it should be taught
It is not and was never taught this way as it is impossible to power out of stall for any aeroplane except for aerobatic ones of extreme performance (e.g. Su-31 and -35).


Originally Posted by misd-again
The issue is that the major airframe manufacturers are reluctant to either release or produce the required data for the TDMs to model.
Major issues being lack of a) need to provide such data at all b) funds c) adequately suicidal test pilots.

Originally Posted by Centaurus
A picture is worth a thousand words...
There is no aeroplane in this picture. Could you please provide us with some incidents/accidents where plane was stalled at high level following penetration of CB?

How and why does one teach stall recovery for pilots of stick-pusher equipped aeroplanes?
Clandestino is online now  
Old 8th Oct 2014, 23:06
  #38 (permalink)  
 
Join Date: Jan 2014
Location: glendale
Posts: 819
Likes: 0
Received 0 Likes on 0 Posts
clandestino, actually it has been taught for some time in FAA land. Now, I don't like it, don't think its right, but it has been taught that way for some time now.

Low altitude stall recovery is firewall power and accelerate. In underwing mounted engine aircraft one must be careful not to allow the nose to come up with engine power application.

I WAS NEVER A FAN OF THIS and think you should get out of the stall (push forward) while adding power.
glendalegoon is offline  
Old 8th Oct 2014, 23:25
  #39 (permalink)  
 
Join Date: Feb 2005
Location: Correr es mi destino por no llevar papel
Posts: 1,422
Received 1 Like on 1 Post
That's "Approach to stall recovery", not "stall recovery".
Clandestino is online now  
Old 9th Oct 2014, 00:27
  #40 (permalink)  
 
Join Date: Jan 2014
Location: glendale
Posts: 819
Likes: 0
Received 0 Likes on 0 Posts
clandestino

fine

then we have never practiced stall recovery in any us simulator session in any plane I have flown.
glendalegoon is offline  


Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.