|
High altitude stall recovery B737
We know that if a high altitude stall occurs for whatever reason, the initial action is to level the wings and lower the nose; thus reducing the angle of attack. Simultaneously thrust is applied. For 20,000 lbs thrust CFM 56 engines at 37,000 ft the thrust available is about one quarter of that at sea level and the pitch up from thrust application is easily countered. Different of course during a go around at sea level
Once recovery from the stall is made (almost instantaneously if the angle of attack if reduced sufficiently and promptly) the question arises what minimum safe airspeed should be attained before the pilot eventually aims to level out? To quick a level out after stall recovery at high altitude invites secondary stalls as evidenced by continued buffeting. The Boeing FCTM does not address this clearly, especially as the pilot may be tempted to prematurely initiate levelling out for minimum height loss. One clue may be found in the FCTM under Chapter 4, Climb, Cruise, Descent and Holding where in part, it states the following: "Recommended holding speeds can be approximated by using the following guidance until more accurate speeds are obtained from the QRH. Above FL250, use VREF 40 +100 knots to provide at least a 0.3 g margin to initial buffet (full maneuver ability)" This suggests that once the airspeed reaches approximately 230 knots (depending on weight of course) during the course of recovery after lowering the nose, it would be safe to stop further descent and review the situation from there. It follows that if the stall was the result of mishandling due to severe turbulence and this severe turbulence was still present during the recovery process, the recovery to level flight should be delayed until reaching severe turbulence optimum penetration speed. This recovery technique when used in the simulator results in a height loss of approximately 3500 feet. If using optimum severe turbulence penetration speed as the aiming point before recovery to level flight, then a height loss of approximately 5000 ft is needed. Constructive comments welcomed. |
This isn't 737-specific (never flown the type) but if it's an EFIS one with an AOA-based stall cue on the airspeed tape, it's a perfect time to use it for live info in a dynamic situation where everything that matters is changing: airspeed, stall speed, elevator authority, actual G, and stall G.
|
Forget the wings level bit first.
Disconnect ap Disconnect at Pitch down to -5 degrees Trim Full thrust Roll wings level Check soeedbrake lever down. Wait for vref 30 plus 100 then recover |
the initial action is to level the wings and lower the nose |
Airbus also recommends pinhead method.
|
Pin Head
Correct. Anyone who has been through basic flying training would know this. First, unstall the wings!! This of course is done by unloading the wings, by lowering the nose, not leveling the wings!! Ailerons central until the a/c is under control. A jet is a fixed wing aircraft. No different to a Cessna 150, for E.G. Don't get away from the basics. AF 447 confirmed that. To be fair, no one has deep stalled a A340 and recovered, to my limited knowledge. |
This from a highly experienced training captain? Gosh.] Hold control column firmly. Disconnect AP and AT. Smoothly apply nose down elevator to reduce the angle of attack until buffet or stick shaker stops. Nose down stabilizer trim may be needed. (See warning note re excessive use of pitch trim). Continue recovery (See note) Roll in the shortest direction to wings level if needed (see warning Note) Advance thrust levers as needed. Retract the speed brakes. The critics have a point. Just testing you:E The order should be lower the nose while actioning all the other parameters asap. |
This is the danger of pprune. Follow QRH / FCTM guidance. It is NOT full thrust initially due 737 thrust pitch characteristics and potential for secondary stall. Also swept wing jet aircraft have very different stall characteristics to a Cessna 150 exacerbated by high altitude / mach effect. AF447 changed the stall recovery philosophy and it is key that light aircraft and swept wing jets are different aerodynamically.
|
Pitch down as necessary. Don’t be an numbers guy and pitch to X, pitch to what is needed.
Recovery? Read the language - 0.3G protection .... that is the lightest of back pressure. How many guys only pull at 0.3G at the minimum recovery speed? Very few, which is when the secondary stall, now an accelerated stall, occurs. |
Never heard -5 degrees before, only pitch to unstall the wing. Depending on how deeply the stall was developed that could be a very significant nose down attitude.
I’m sure someone will correct me but I have heard the A350 in one of its stall tests required almost 30 degrees nose down pitch to recover, in part due to some CofG issues, but still shows the point! |
At high altitude, with mach effect and less dense air coupled with swept wing aerodynamics the pitch attitude to unstall the wing is markedly below the horizon and will result in a considerable loss of height. This is one of the conclusions of the AF447 catastrophe, the fact that a stall condition was not recognised by an apparently well trained crew. The traditional 'cessna' style recoveries with minimum height loss are not applicable to jet transports. The height loss is a factor when related to terrain proximity which is part of general SA. QRH / FCTM applies and beware tips / techniques from unverifiable sources!
|
Originally Posted by misd-agin
(Post 10053919)
Pitch down as necessary. Don’t be an numbers guy and pitch to X, pitch to what is needed.
|
On a serious note, a big pet peeve of mine about this topic is when people call it a "high altitude, low energy" situation. It is not low energy, from 30,000 feet you have gobs of energy below you (namely, 1491 knots' worth) to use. That is, as long as you have the understanding and fortitude to use it.
Of course, I know they mean "low kinetic energy" and not "low total energy," but in that case they could have just said "low airspeed" with no loss of information. But no, they have to muddy the waters by using the fancy high tech sounding term. Unqualified "energy" is supposed to mean "total energy." |
In my high altitude stall sessions, which notably just began a little over a decade ago during recurrent training, the recovery has never been immediate, particularly when the setup included plenty of NU trim by George prior to the stall.
Centaurus is quite right regarding the underslung engine pitch considerations at altitude - the onset of thrust is far slower than when down low and the sim modeling clearly shows this. Speaking of sim modeling, according to the experience of NG test pilots, the buffet during the approach to a high altitude stall is so pronounced, one would have to be in a coma to miss it. This is an area where the simulator lacks fidelity. |
Originally Posted by RHS
(Post 10054053)
I’m sure someone will correct me but I have heard the A350 in one of its stall tests required almost 30 degrees nose down pitch to recover, in part due to some CofG issues, but still shows the point!
|
I must admit our -800 sim gives a very significant amount of buffet. Perhaps no where near that of an aircraft but I can't imagine not noticing.. Replicating the Turkish Airlines B737 Amsterdam accident where the thrust levers closed due to a defective radio altimeter while flying the ILS is also most instructive. There, the continuous back trimming of the stabiliser as the autopilot tries to maintain the aircraft on ILS glideslope at idle thrust means when the stall recovery is attempted when the stick shaker actuates while disconnecting the autopilot, considerable forward stab trim is needed for several seconds, coupled with a delicate balancing act in terms of elevator pitch control by the pilot to prevent critical height loss during the recovery phase from VREF minus 25 knots. This is where manual flying skills are absolutely vital; especially if the event happens in IMC or night. |
Speaking of sim modeling, according to the experience of NG test pilots, the buffet during the approach to a high altitude stall is so pronounced, one would have to be in a coma to miss it. This is an area where the simulator lacks fidelity |
Low level recovery I target 5degrees NU with a mid thrust settings.
As mentioned target the pitch attitude, hold then possibly like Turkish, lots and lots of trim. |
What about deep stall recovery? I ask, as if the initial stall recovery is not completed correctly and a secondary stall developes and then a deep stall. I.E., stabiliser also stalled. You need to regonise and recover from that very quickly, as the RofD increases at an alarming rate.
In any fixed wing aircraft, the wings must be unstalled whether it be a C150 or an A380. Apply all the power or trim you like. You are going nowhere, except down unless those wings begin to produce sufficent lift. |
Simulators are programmed to behave according to the data package which is given by the manufacturer based on test flight results and wind tunnel data. The buffet will not happen unless it is programmed. The device limitation may produce less but not more. Also about deep stall etc. Simulator behaviour is not authentic in the region outside the envelope where test pilot hasn't gone and no data is provided. So such exercises are misrepresentative of actual aircraft behaviour which no one knows.
|
Exactly!
Why havent they carried out these tests? Surely if the deep stall tests could be carried out during test. If the situation was unrecoverable, then a recovery parachute in the tail could be deployed. That should at least get the nose pointing down. I thought test pilots were of the "right stuff". Seems i maybe wrong. Are crews now, part of the test flight program? Has Health and Safety hamstrung flight testing also? Yes i know we shouldnt get into these situations but we do, humans being humans. |
Dan, most manufacturers explore both the stall boundaries and after effects.
Depending on aircraft type, stall characteristics can vary widely. Small differences in build quality or maintenance, structural flexing, weight variation, control rigging, ... all add variability to many uncalculable aerodynamics effects. See the NASA report relating to 737 stalling, including high altitude. There is reference to a simple stall model suitable for the task, but first define the task; awareness, recognition, avoidance, or recovery. Then provide the funding according to means and need, considering the ever present variability of the human. You could train everyone, but never be assured of the behaviour in operation. And how many accidents have their been from unrecovered stall encounters in situations where recovery was feasible, i.e. not very low altitude, impaired controls, etc. “All airline pilots agreed or strongly agreed that they were surprised by the surprise stall scenario. In that scenario, less than one quarter of the airline pilots strictly followed the proper stall recovery procedure on which they had been briefed. Less than half maintained a nose-down input until the stall warning stopped. It may not matter what stall model you use if pilots are going to react inappropriately; better to avoid the situation altogether. https://www.faa.gov/pilots/training/...r_Training.pdf |
This thread has gone from a simple discussion to the absurd! Deep stall discussion is similar to how low can you go. Simulators are toys programed to behave and react in certain ways that are predetermined. Stall should be recovered from with the least amount of altitude lost even if the buffet continues for several seconds longer. A buffet can occur at cruise when operating at almost maximum altitude and a bank of something around 20 degrees. Apply some thrust and minimum roll and the buffet is gone. I remember the days of training in light aircraft. Idle thrust hold altitude and wait of the buffet/horn. Keep the nose up and wings level in falls through. Don't change a thing, the nose falls, speed rebuild followed by another stall, again and again.
The discussion about don't roll wings level is really more airframe specific. Some use a spoilers for roll, that brings about a whole new topic. Keep it simple and you will not kill yourself. |
Originally Posted by mustangsally
(Post 10055102)
Stall should be recovered from with the least amount of altitude lost even if the buffet continues for several seconds longer.
The discussion about don't roll wings level is really more airframe specific. Some use a spoilers for roll, that brings about a whole new topic. If there's no need for a hurry to roll, better to wait longer rather than shorter to make sure the wing is completely unstalled. |
Vessbot...agreed..I thought for good reason powering out of stalls had gone out of fashion, certainly on anything with underslung engines....
I certainly can’t speak for the 737 but FWIW after the well known high profile stalling/ incidents/accidents worldwide a few years back our Boeing big twin procedure was amended by Boeing to this: Initiate the recovery: Smoothly apply nose down elevator to reduce the angle of attack until buffet or stick shaker stops .... Continue the recovery: Roll in the shortest direction to wings level if needed* Advance thrust levers as needed..... As currently taught (and as in the QRH and FCTM) the emphasis is on doing it very much sequentially i.e. you must get rid of the buffet before rolling or applying thrust..... |
Originally Posted by Pin Head
(Post 10053775)
Forget the wings level bit first.
Disconnect ap Disconnect at Pitch down to -5 degrees Trim Full thrust Roll wings level Check soeedbrake lever down. Wait for vref 30 plus 100 then recover |
Some really bad information here. 5 dog nose down might not break the stall. And waiting for x airspeed to pull?How hard are you going to pull? How heavy are you? How high are you? Or does any of that aerodynamics and physics stuff matter?
Simulators don’t have post stall AOA fidelity. That’s the change. A and B have agreed to a generic n/b post stall fidelity model. You can’t compare current, or especially older, ‘stalls’ in a simulator and pretend they’re the real thing. They’re not. I’m not sure how you can look at this video and write “apparently well trained” - |
Attempting to answer Centaurs’s opening question logically;
Emergency manoeuvres - EGPWS Pull Up, Windshear - recommend speeds as low as ‘respecting the stall warning speed’, normally 1.1 stall speed. However, this assumes that the aircraft has a stall warning system which corresponds with the general certification requirements - surprising if not. Furthermore, that the high level stalling characteristics are similarly warned, and that the stall margin (Mach effects) enable reasonably safe flight at this low speed - possibly not. Given that altitude loss is not an issue at high altitude - no not even considering risk of collision etc, etc; focus on the primary safety issue. Then starting a gentle pull up after loss of stick shake, aiming to level / climb at a higher speed (min manoeuvring?) would appear to be safe. But further problems are the sensitivity of the speed display at altitude (flap speeds, etc, not available) and the use of Mach for reference. These similarly can be overcome by trading altitude for speed - turb speed, which might be required anyway for turbulence/ manoeuvre, i.e. addressing factors which could have contributed to the stall situation Always test assumptions and logic - comments? |
Originally Posted by mustangsally
(Post 10055102)
Stall should be recovered from with the least amount of altitude lost even if the buffet continues for several seconds longer.
|
Why havent they carried out these tests? |
Super trio of Tube's by D Davis' days test flying. I thought I'd pop in for a moment, but two hours later I was still hanging on every word.
I'd owned his book since 1970, though seen my crew's later copies, however, I'd never heard him speak before. A bit of Winkle Brown in his delivery. One thing in the later book was an impassioned plea to enable crews to be able to get hands on a real aircraft for occasional real flying. Probably gave bean-counters nightmares. I was lucky, for years I was able to take the real aircraft into the stall, (Viscount) or push, (BAC 1-11) It's true to say that while I enjoyed that era, I was always rather perplexed by the lack of trainer's knowledge about pulling out of the (substantial) dive. I recall doing two or three in quick succession from 30,000 and getting a good view of the ocean before feeling any significant g. I had NO guidance on the recovery commencement. Great fun. Of course, getting the load off the wings was done for me by the nitrogen ram. The Claxton made quite sure one was not nodding off. |
If you want to hear what D P Davies had to say about stall testing and other aspects of certificating aircraft, go to this thread. These are the ones I think Loose rivets was referring to above.
https://www.pprune.org/tech-log/6029...-aircraft.html |
Why havent they carried out these tests? Report: Fedex DC10 near Raymond on Jun 14th 2008, aerodynamic stall while in holding http://www.fss.aero/accident-reports...9-11-11-US.pdf |
Vesspot. WHY? Because you might be near the ground and die.Stupid statement.
|
Check Airman.Because you might be near the ground and die! Again a stupid statement from you.
|
At high altitude its not important how much altitude you lose but hear the ground YOU MIGHT CRASH BURN DIE
|
[QUOTE=misd-agin;10055151]Some really bad information here. 5 dog nose down might not break the stall. And waiting for x airspeed to pull?How hard are you going to pull? How heavy are you? How high are you? Or does any of that aerodynamics and physics stuff matter?
Simulators don’t have post stall AOA fidelity. That’s the change. A and B have agreed to a generic n/b post stall fidelity model. You can’t compare current, or especially older, ‘stalls’ in a simulator and pretend they’re the real thing. They’re not. /QUOTE] I’m not sure I understand what you are stating here.0 As of now (most) regulations only require datapacks with stall models up to onset of initial buffet or thereabouts. On March 30, 2016, the FAA published changes to the 14 CFR Part 60 Qualification Performance Standards (QPS) that define updated general, subjective, and objective testing requirements for high angle of attack modeling and the qualification of full stall maneuvers on Level C and Level D FSTDs. EASA published NPA 2017-13 addressing many changes including advance stall. Some of the founding members of AUPTRA have conveniently merged UPRT with advanced stall to produce ridiculously high cost aero model add ons for FSTDs Most changes will be in regulations from 2019? |
Originally Posted by vilas
(Post 10055639)
Can the commercial aircraft be tested like an aerobatic aeroplane? The design and purpose is to fly them within certain boundaries. If for whatever reason a line pilot ventures beyond that then the uncertainty of the outcome must be accepted. China airline 747 survived a barrel roll but wasn't designed or can't be tested for that.
|
The point is test flights are done to meet certification requirements. Also commercial aircraft cannot be built like combat aircraft. Then only thing they will carry is fuel. Besides since there are no ejection seats test pilots cannot risk there lives unnecessarily. It will be safer to ensure through better training that pilot will remain within the envelope. Stay clear of 447 otherwise there will be another 10000 replies.
|
Having just done an exercise in a 330 sim which was in some ways to replicate the AF447 incident, the amount of pitch down and altitude loss to unstall the wing, and then recover without inducing a secondary stall, was quite memorable. A few hands worth of forward trim helped.
|
| All times are GMT. The time now is 09:20. |
Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.