My book "Angle of Attack" is out!

[seagull]

Excellent points. Would love to see more training on go arounds, alternate/direct law, compound failures etc. But how does one go about this? Companies are profit driven, and we all know we can't show up to the training department on our days off and ask for some sim sessions. I venture to guess more pilots probably would if given a choice.

You are correct that it is a challenge to get the companies to change. We actually need the entire framework to change, which occurs at the industry/regulator level. That is a big reason that started this book project. It is also why I am working on a number of projects that are outside the formal framework that will hopefully create the pressure to change things.

 

[Cherokee_Cruiser]

Come on now, it sounds like you are absolving the crew of blame. As I said, pitots froze up on a NWA A330 as well. They lost all speeds. They didn't end up like AF447.



From 0:57 to 1:36 he did nothing but chase the FD bars. At 1:36 he doesn't have aileron authority to follow the FD bars. At 1:42 he goes full aft and full left, which is the last position of what the FD bars were showing him. This is the epitome of what a low time ab-initio pilot would do in a panic situation which is above his experience level. Not to mention the sheer stupidity of any Airbus pilot physically going above 10 degrees on the attitude indicator in the Flight Levels and holding it there!

We can beat around the bushes all day. Yes they flew into weather they shouldn't have. But all that weather did was ice up their pitot tubes and then the subsequent lack of airspeed indications and degradation to alternate law. IIRC the first pitot tube cleared itself and correct data came back in less than 60 seconds, and followed by the next two pitot tubes.

 

[Kingairer]

Haven't seen that yet.

 

[trafficinsight]

Come on now, it sounds like you are absolving the crew of blame. As I said, pitots froze up on a NWA A330 as well. They lost all speeds. They didn't end up like AF447.



From 0:57 to 1:36 he did nothing but chase the FD bars. At 1:36 he doesn't have aileron authority to follow the FD bars. At 1:42 he goes full aft and full left, which is the last position of what the FD bars were showing him. This is the epitome of what a low time ab-initio pilot would do in a panic situation which is above his experience level. Not to mention the sheer stupidity of any Airbus pilot physically going above 10 degrees on the attitude indicator in the Flight Levels and holding it there!

We can beat around the bushes all day. Yes they flew into weather they shouldn't have. But all that weather did was ice up their pitot tubes and then the subsequent lack of airspeed indications and degradation to alternate law. IIRC the first pitot tube cleared itself and correct data came back in less than 60 seconds, and followed by the next two pitot tubes.

I ask myself "self? What would cause someone to hold the stick full aft while the stall warning is screaming at them?"

What's your answer to that question Airbus driver? Can you expand your mind for a moment and instead of say "I would never do that." Ask yourself "what freakish combination of events would cause me to go down the same road?"

It's not that anyone is absolving anyone of error it's that saying "so and so blew it," while true, isn't at all helpful.

Sent from my XT1650 using Tapatalk

 

[seagull]

Come on now, it sounds like you are absolving the crew of blame. As I said, pitots froze up on a NWA A330 as well. They lost all speeds. They didn't end up like AF447.



From 0:57 to 1:36 he did nothing but chase the FD bars. At 1:36 he doesn't have aileron authority to follow the FD bars. At 1:42 he goes full aft and full left, which is the last position of what the FD bars were showing him. This is the epitome of what a low time ab-initio pilot would do in a panic situation which is above his experience level. Not to mention the sheer stupidity of any Airbus pilot physically going above 10 degrees on the attitude indicator in the Flight Levels and holding it there!

We can beat around the bushes all day. Yes they flew into weather they shouldn't have. But all that weather did was ice up their pitot tubes and then the subsequent lack of airspeed indications and degradation to alternate law. IIRC the first pitot tube cleared itself and correct data came back in less than 60 seconds, and followed by the next two pitot tubes.

It is also what a high time pilot does when faced with high uncertainty. Do not forget the windshear procedure of alpha-floor. Also, you are missing what it is like to be in a thunderstorm at 1/2 a g for an extended period of time with no apparent reliable instruments. It is the people that want to find that simple solution so they can think "I would never do that" that are the ones more likely to not seek the skills necessary to prevent an event in the future.

That said, there is absolutely a difference between those pilots who have had the pleasure of a lot of hand flying up in the 30s and higher, back prior to RVSM (where they could and did dispatch us across the Pacific with no autopilot) and those that have "come of age" since RVSM and a number of other constraints. Part of the reason that the industry does not recognize this as a problem is that we are still able to leverage the skill set of the generation of pilots that came before. With newer pilots never having the opportunity to experience high altitude hand-flying or a wide variety of control responses, as opposed the the predictability of a FBW control system, we are losing resilience.

 

[mrivc211]

At the very least, if I were pulling back the ENTIRE TIME and it WASN"T WORKING, I would try and do the opposite to see what happens. Geez you have 30,000 feet to play with. Full aft the ENTIRE time? I'm sorry. You can do all the research you want, these guys blew it. Either you have it or you don't. I've done high altitude stall training several times, the airplane is not responsive at all but the yoke/stick still does what its supposed to do.

 

[seagull]

At the very least, if I were pulling back the ENTIRE TIME and it WASN"T WORKING, I would try and do the opposite to see what happens. Geez you have 30,000 feet to play with. Full aft the ENTIRE time? I'm sorry. You can do all the research you want, these guys blew it. Either you have it or you don't. I've done high altitude stall training several times, the airplane is not responsive at all but the yoke/stick still does what its supposed to do.

This is a total of 3 minutes. They DID try the opposite, it did not appear to work. The reason is that for it to work would take at least 20 seconds, which is a large fraction of the time. You are pushing to zero g or less during that time. Full forward stick. If you have done high altitude stall in training, who's simulators did you use? Were they modeled for it? Most are not. You are over simplifying the situation here.

 

[mrivc211]

This is a total of 3 minutes. They DID try the opposite, it did not appear to work. The reason is that for it to work would take at least 20 seconds, which is a large fraction of the time. You are pushing to zero g or less during that time. Full forward stick. If you have done high altitude stall in training, who's simulators did you use? Were they modeled for it? Most are not. You are over simplifying the situation here.

I know it takes more than a few seconds. It takes a "trend" of fwd stick. If you push too far forward, you are going to porpoise and just makes the stall worse in that your peaks and valleys will be greater apart. I used Flight Safety's sims.

 

[MikeD]

That is part of the point. Their "basic airmanship" was actually quite good based on all available evidence. That is part of what makes this accident a good case study.

The basic airmanship that the crew needed to revert to in order to identify what was going on, but to which they never reverted to? Even the BEA cited in their report that the crew never IDd the problem or the status of the aircraft because they never did revert to any basic airmanship. They never understood the initial failure they had, and never listened to what the airplane was telling them. The sustained stall, shown by heavy buffeting and a stall warning sounding, was never understood to be what it was, and any sort of useful stall recovery maneuver was consequently never applied.

That the crew did not understand the aerodynamics, did not possibly understand meteorology, did not understand what the airplane was telling them when it was talking to them both aerodynamically as well as with systems, may have had training deficiencies, and did not revert to nor apply basic airmanship; does not absolve the crew of errors in the actions they did take, as well as actions they failed to take. These are the "what" of the accident, followed by the "how", that are the primary causal factors of why that jet ended up in the ocean. The analysis items....the "why" items.....are all secondary contributing factors to the what. And are very real and serious contributing factors, but contributing nonetheless.

That said, I fully agree with the changes recommended that need to be made regarding stall training, academic training and systems training deficiencies

 

[MikeD]

It's not that anyone is absolving anyone of error it's that saying "so and so blew it," while true, isn't at all helpful.

Sent from my XT1650 using Tapatalk

That the thing. The crew did blow it, through their actions and inactions, AND, there is good analysis as to why they did. Doesnt make the crew a bunch of bad guys, as they never intended or wanted that plane to end up where it did nor were they doing anything reckless; it just makes them humans who made a series of errors through actions and inactions that put the plane into the ocean (the what and how's). Errors that trace back to an array of contributing factors, from design ergonomics to training to airmanship, etc (the why's). But the primary causal factor of why that jet ended up in the drink, was their failure to recognize a minor issue, an improper reaction to it with manual aircraft control, which when that only worsened the situation with the aircraft entering a stall, a stall whose aerodynamic and aural indications and warnings were neither recognized nor heeded, all of whch contributed to a breakdown of CRM and teamwork and a rapid increase in confusion, and control inputs being made and not getting time to take effect, as the aircraft settled 40,000' to the ocean. Lots of lessons learned and lots of takeaways. However these need to be placed in their proper slots of primary, secondary and tertiary; as well as causal vs conttibutory.

 

[seagull]

I know it takes more than a few seconds. It takes a "trend" of fwd stick. If you push too far forward, you are going to porpoise and just makes the stall worse in that your peaks and valleys will be greater apart. I used Flight Safety's sims.

Flight Safety's sims are not yet programmed to handle stall beyond the point of stall warning. How far back into the stall did you pull it?

Also, remember your response is based on what we call hindsight bias. It is very easy to see what someone did wrong,and when, with hindsight. Accidents looks like "chains of events" in hindsight. None of that is true when things are actually happening.

 

[seagull]

That the thing. The crew did blow it, through their actions and inactions, AND, there is good analysis as to why they did. Doesnt make the crew a bunch of bad guys, as they never intended or wanted that plane to end up where it did nor were they doing anything reckless; it just makes them humans who made a series of errors through actions and inactions that put the plane into the ocean (the what and how's). Errors that trace back to an array of contributing factors, from design ergonomics to training to airmanship, etc (the why's). But the primary causal factor of why that jet ended up in the drink, was their failure to recognize a minor issue, an improper reaction to it with manual aircraft control, which when that only worsened the situation with the aircraft entering a stall, a stall whose aerodynamic and aural indications and warnings were neither recognized nor heeded, all of whch contributed to a breakdown of CRM and teamwork and a rapid increase in confusion, and control inputs being made and not getting time to take effect, as the aircraft settled 40,000' to the ocean. Lots of lessons learned and lots of takeaways. However these need to be placed in their proper slots of primary, secondary and tertiary; as well as causal vs conttibutory.

These determinations still require a heavy dose of hindsight bias. All of this is clear ONLY in hindsight. Again, this sounds to me that if someone has poisoned your drink and you did not know it it is still your fault for drinking it! And, while what you write is true on the face of it, it is also pretty much useless, like HFACS. Also, I would disagree that there was a breakdown of CRM at all, why do you think that? I would also challenge even a test pilot to make the control inputs long enough to have an effect under the conditions they were in.

What is really amazing to me is that even with the benefit of hindsight, a great many pilots would still fall into the exact same trap. They STILL do not understand the systems or what high altitude stall are like well enough to avoid a similar situation, and we STILL know that most pilots do not know how to work their radar or meteorology enough to avoid the scenario in the first place. I am talking about numbers north of 90% of airline pilots. They THINK they have the knowledge, but they don't. They, instead have ad-hoc and incorrect mental models of what occurred, and the reasons it did. These false narratives help them to feel better but have done nothing to prevent a future similar event.

 

[Tworotor7]

Flight Safety's sims are not yet programmed to handle stall beyond the point of stall warning. How far back into the stall did you pull it?

Also, remember your response is based on what we call hindsight bias. It is very easy to see what someone did wrong,and when, with hindsight. Accidents looks like "chains of events" in hindsight. None of that is true when things are actually happening.

I recently did upset recovery in a G550, Flight Safety sim. It was modified to fly past the normal envelope. The program has been out for 2+ years, I believe.

 

[seagull]

I recently did upset recovery in a G550, Flight Safety sim. It was modified to fly past the normal envelope. The program has been out for 2+ years, I believe.

Interesting, although the modeling for the smaller jets is a bit simpler due to lack of fuselage flexure, the last I had checked they had not done so, but perhaps they were discussing the larger transports? How much do you know about the modeling? Who did it? The part 60 rule has only been out for a year now and it was not clear what would be required 2 years ago.

 

[Tworotor7]

Interesting, although the modeling for the smaller jets is a bit simpler due to lack of fuselage flexure, the last I had checked they had not done so, but perhaps they were discussing the larger transports? How much do you know about the modeling? Who did it? The part 60 rule has only been out for a year now and it was not clear what would be required 2 years ago.

I'm not sure about larger transport, as I've been in the corporate world for the past few years...
I'm told it was a group effort between Gulfstream flight test data, their test pilots, and flight safety. It was a great program and they were able to simulate all aspects of the flight envelope. Pretty cool.


https://news.flightsafety.com/gulfstreamarticle/upset-prevention-and-recovery/

 

[Low&Slow]

Can I add letters after my name too, it makes me look more legit.

Wheelsup, PoGoSTK

Just add "esq." There aren't any real, actual requirements to use it in the USA.
Anyone can use it.

 

[mrivc211]

Flight Safety's sims are not yet programmed to handle stall beyond the point of stall warning. How far back into the stall did you pull it?

Also, remember your response is based on what we call hindsight bias. It is very easy to see what someone did wrong,and when, with hindsight. Accidents looks like "chains of events" in hindsight. None of that is true when things are actually happening.

It was high altitude stall training approved by the FAA in those sims because a bunch of idiots were reading newspapers in speed mode and letting the plane stall. The FAA signed off on it so not sure how the sims couldn't be approved for it, but it doesn't surprise me that that would happen. We never pulled the stick to stall it. We let the plane get slow, stall around mid 30's, then try to recover. In the real world, guys were trying to recover by going full power and keeping the nose level to minimize altitude loss, then calling center and asking for lower immediately to try and hide what they had done. All this did was worsen the recovery and add the amount of altitude you were going to lose before you recovered. The proper recovery method in that particular case was to keep the nose down and find the "sweet spot" in terms of pitch down attitude until you've gained enough airspeed back then level off, or you were going to porpoise down 10-15,000'.

 

[seagull]

The fact that the FAA "signed off on it" means absolutely nothing, as you indicated. The rule 60 compliance is not required for another couple of years, still, so they approve it based on _current_ requirements. Also, you did not indicate how far into the stall you got into before the recovery, although absent a correctly modeled simulator (and there are still only a handful on the planet for transport airplanes) is really is all practically negative training at best.

 

[jonny]

There is an art and a science involved in flying jets. If you feel it in your ass, it's the art side; the science side is responding to the indications you are getting in the cockpit. But if you feel it in your ass, and the instruments aren't backing up what you're feeling, set a pitch and power that will make you fly and figure out if your ass is wrong, or the instruments are wrong. I tell FO's that are playing the thrust like a trombone that the Lord giveth, the Lord taketh away... So a plus 10kts can become a minus 20kts shortly later... Set a thrust and adjust if you're going to exceed a limitation. The feeling it thing has been sent by the wayside in a lot of places and that is sad. Congrats on getting published. It is not an easy thing!

 

[Cherokee_Cruiser]

Actually, that is not what you had in the front seats. Both of them actually came from flying families, had glider experience, aerobatic experience and flew on the side. You are missing many key aspects, sounds like you read a Vanity Fair article for much of that, which was mostly wrong. This article might help a bit: https://airlinesafety.wordpress.com/2015/04/07/high-altitude-stalls-how-well-do-you-understand-them/

Their experience level was in the final report. The main instigator Bonin had about 2900 hrs and most of that was Airbus sidestick time. I think he had almost 1000 hrs on the 330 fleet type at AF, which means he handflew ~ 10 minutes tops every 10 hr flight. How much actual handflying skill did he really have? Very minimal.

That is part of the point. Their "basic airmanship" was actually quite good based on all available evidence. That is part of what makes this accident a good case study.

Really? Can you name one thing or things they got right as part of "basic airmanship" ??

I'd say their initial correct identification of the problem (looks like we have bad speed indications) was about the only thing they got right. Everything else they got wrong. Procedure says once you have the problem identified you follow your company policy to fix it. Robert (left seat FO) went straight to ECAM and read that to try and take steps to fix it. But the unreliable airspeed relies on the pilot doing other stuff first to get the airplane under control. AP off, AT off, FD off, select climb power (already there), initial pitch 5 degrees, clean up flaps/gear, level off. Had these steps been taken, and ignore *everything* on ECAM for about 90 seconds, they would have own straight out of it and speeds would have come back as ice melted/broke off pitot probes.