Among other things in his assertion that it would be the same in Boeing or Airbus airliners. The scenario he describes in particular does not apply to Airbus FBW airliners. In those, a go-around is not initiated by "clicking" a button but by pushing the thrust levers to the forward stop. This will always give full go-around thrust and enable the appropriate autopilot/flight director modes. The computers will never "ignore" the pilots commanding GA thrust, regardless of whether the airplane is in the air or on the ground.

New passenger's report posted on avherald:

On Aug 8th 2016 a passenger reported in the reader comments on AVH below, that the approach was normal, the landing gear was down. The aircraft touched down, however, the nose was not lowered onto the runway and the aircraft appeared to climb away again, the gear was retracted, however, the engines did not spool up. The aircraft made ground contact again, skidded along the runway with the right hand engine separating from the aircraft but still being dragged along with the aircraft until the aircraft came to a full stop.

Luckily it was posted here by a kind soul: http://www.pprune.org/rumours-news/5...ml#post9467384

Mr Bailey is wrong in his assertation that -



There is a newish trend to assume that the automatics are failsafe & to religiously trust them to the extent that complete control is given over to them. Some pilots follow this mantra so far as to fail to see the need to monitor when the aircraft automatics are engaged. A B777 check captain of the airline involved here stood in front of the recurrent class that I was in & asked "How many pilots does the B777 have?" The correct answer according to him was "Three - the captain, the first officer & the autopilot".


I disagree with this position. The autopilot is not a pilot. It is a pilot aid. It is not there to cover for ineptitude or lack of sufficient, suitable training, despite what management may think or desire.


Simply stated in relation to what it appears may have happened here, you do not just hit TOGA in a go-around & then totally ignore the thrust lever position (B777) or the thrust indications on the upper DU. You are required to ensure that the automatics have given you what you have requested of them & if they haven't, DO IT MANUALLY! This is what is meant by 'Fly the aircraft'.


Over reliance on the automatics is a hot topic these days & has been done to death on various threads here & on other websites. However, it is becoming more prevalent in the industry as inadequately trained or incompetent or lazy pilots (or a combination of all three) become more widespread in the industry. Sounds harsh, but that's the truth of the matter that I see almost every time I go to work. A number of recent accidents also highlight this problem.



Maybe Mr Bailey's statement that 'pilots are rigorously trained to trust supposedly failsafe automatics' is at the very heart of the problem.

No computer programmer can cover all the real world "if then" situations.
What is needed is a big red button labelled "Tiger Moth Mode"

Ask yourself how many times have you uttered "Why is it doing that".
If you are in cruise you probably have time to sort it out. In a go-around from a bounce, time is not on your side.

The problem is, the 777 autothrottle was never designed or intended to be a 'failsafe' system. It was not certified as flight critical because the pilot can always override what it's doing simply by manually moving the thrust levers. Plus, unlike most flight critical systems, there is an "ON/OFF" switch. If pilots are being taught otherwise, you've just identified a big part of the problem.

Mr Bailey's analysis sounds correct, although he has simplified it for his audience. Hence his statement that pilots are "rigorously trained to trust failsafe automatics".
It would be more accurate to say repeated experience reinforces expectation of a certain result from a given action, and in an unusual and stressful event the inevitable intuitive reaction may end up with a situation nobody expected.

We can bang on about pushing the thrust levers up as an associated action, and fair enough, but strictly speaking thats not how the system is designed, and not how we use TOGA on a day to day basis. Off the top of my head the only procedures formally trained that way are Windshear and GPWS. I have never heard of go round off a bounce being trained for either.

And who knows, maybe the thrust levers were shoved forward but not far enough? In fact that sounds quite feasable: an armful of power lever and expecting the autothrottle to finish it off would disguise the situation.

It has become common practice to brief the FCTM procedure for bounced landing in the last few years. I have never heard any body say "oh, and by the way: Autothrottle TOGA mode won't work". That would be because nobody has thought of it before. Same as prior to the Asiana crash nobody had thought about the effect of indirect mode change. And half of all pilots still don't understand it.

This thread will probably spin off into a manual skills blag but that would be missing the point. And the point is the Boeing FCTM, FCOM and training.

Of course all this is still just speculation in the absence of facts.

Not necessarily true
 

"It has become common practice to brief the FCTM procedure for bounced landing in the last few years. I have never heard any body say "oh, and by the way: Autothrottle TOGA mode won't work". That would be because nobody has thought of it before. Same as prior to the Asiana crash nobody had thought about the effect of indirect mode change. And half of all pilots still don't understand it."

Actually, my company teaches press TOGA, push throttles and keep pressing TOGA until it engages. We teach this precisely because GA will not engage if the squat switches are on.

The point is that they don't have to understand it - as long as they do not give full control to it & continue to fly the aircraft. Once you abdicate your responsibility as a pilot to fly the aircraft & give it to the automatics, you are starting down a very slippery slope.


Fully understanding the automatic systems on an aircraft is desirable & everyone should spend time on this throughout their career, but with the complexity of systems these days & the lack of currency with some of the more unique & subtle areas of automation, no one can know or remember everything. That is why you should never give full control to the automatics. You should always be 'flying' the aircraft. The Asiana pilots should have been keeping airspeed & thrust indications in their scan, regardless of whether or not they thought the A/T was engaged & would control speed. If they had done that, the accident probably wouldn't have occurred. The Turkish B737 & the Colgan Dash-8 accidents also come to mind.


Attitude, altitude, airspeed & thrust need to be monitored at all times, regardless of whether or not the automatics are engaged. This is basic aviation as far as I'm concerned.

Doh! Pipeliner, got me there! Exageration for effect��

I've certainly been warned. Another place you need to push the throttles up manually is with windshear on takeoff after V1.

So you believe the "Pilots just flew into the trees because they didn't notice them" explanation for AF298 then?

Hehehe, don't shoot the messenger, personally know Capt Bailey from B727 days in Emirates and in a way he is right but again whatever was and will be said either in the press or social networks, general public not always understand and don't need to understand therefore let this be part of future training sessions after investigators give more details.

Many on this forum speculate too much, I believe the cockpit crew was trained within same standards as was cabin crew in their part of saving lives.

And for those who continue to debate issue of TOGA been applied or not, I have googled report of French BEA, B777 Air France (maybe it was mentioned here before)...saying clearly:
====================
quote:
On approach, descending on the flight path, pushing the TOGA switches leads to selection of go-around thrust guaranteeing a vertical speed of at least 2,000 ft/ min, automatic disengagement of all of the previously selected AFDS modes, and automatic engagement of the go-around roll and pitch modes. Pressing a second time leads to selection of maximum thrust.
and conclusion:
This serious incident was due to the inadequate monitoring of flight parameters by the flight crew (and this is the part where I believe the 521 crew failed).


The following factors contributed to this:

• Partial execution of the go-around procedure;
• Inadequate management of the automatic systems during execution;
• The conflict of plans of action between respecting the operators instruction and continuing the landing, which seemed to be safely possible according to the manufacturer.

=========================
unquote

Heheh that was long time ago, when there was no internet and mobile phones either...I have witnessed during my training in Toulouse the transition from the old to the new concept of flying, when experienced Capt, flying B737-400's threw his hat after few days in a A320 sim, saying:
I will retire in 2 yrs time and I don't need this...:{

Agree 100%, another example proves B777 is capable flying safely, despite 2 tail strikes and 3 bounces....
China Cargo Airlines, CPH 2011
ASN Aircraft accident Boeing 777-F6N B-2078 København-Kastrup Airport (CPH)

Nope, not true. This 'problem' was recognised many years ago. We have a 'Rejected Landing' procedure in our 777 FCOM that covers this scenario. It clearly states "TO/GA switches are inhibited" and "Autothrottle is not available". We regularly train for it in the simulator. If Emirates does not have the same procedure and/or doesn't train for it, then questions should be asked.

the Bailey opinion piece is rubbish.i flew the 320 and the 747-400. nobody ever taught me to trust and/or depend on anything---ever.

unless times have changed, which i doubt--- its been awhile.

Costal pilot,
Not sure I agree with you there,

So much emphasis is placed on teaching and understanding auto flight and protections today, when will they work, when won't they. IMHO, they are there to save my a$$ when I'm having a very bad day. I don't fly the airplane depending on them.

If the TOGA switch/es were pushed during the (suspected) bounce (i.e. in the air), would the A/T and A/P produce the desired result?

Buzzbox:

'rejected landing' and 'bounced landing recovery' are not necessarily the same thing in this context. Boeing manuals don't have a direct reference, unless your airline has customised them. I don't work for EK by the way.


Boeing FCTM:

"When a high, hard bounce occurs, initiate a go-around. Apply go-around thrust and use normal go-around procedures. Do not retract the landing gear until a positive rate of climb is established because a second touchdown may occur during the go-around"

Boeing FCOM:

"The TO/GA switches are inhibited when on the ground and enabled again when in the air for a go–around or touch and go."


Actually, something doesn't add up with this theory: during a bounce you'd only be on the ground for an instant so presumably TOGA would've been selected airborne and should work.

HPSOV L:

The Rejected Landing procedure in our 777 FCOM states the following:

The problem with a bounced landing recovery is that the TOGA switches do not become active again until a short time (2 seconds??) after the aircraft becomes airborne. Pressing them will do nothing until after the inhibit ends. The Rejected Landing procedure in our FCOM covers both cases.

I understand the Rejected Landing procedure is not included in the 777 AFM, but I also understand that some airlines, including my employer, have chosen to include it in their FCOM, presumably with Boeing's blessing. Why hasn't Boeing seen fit to include it in the AFM for use by all operators?

Thanks Buzzbox,
Good info, yes you have to wonder why we don't all get the same stuff.
I suppose the information is all there in the basic manuals but you have to work to extract it and tie it all together.

Regarding the above procedure and TOGA being inhibited:
I have watched a pilot off a 777 flying a simulator landing in a 737, he did a nice job of the entire session ( obviously a competent pilot in my opinion) and when he reached the flare he flared nicely but didn't reduce the thrust, even when we scooted along at 20ft for quite some time. Eventually I told him to close the thrust levers. This guy was a good pilot, but when his capacity was being used he had no motor skill memory for the thrust levers. It had been eroded from years of automation in the flare which was designed and introduced to increase flight safety .
I'm not suggesting this played a role in this accident as I know nothing about it but it is relevant to the discussion of automation in this case.

This is a bit scary. There are some very good posts here by pilots of B777. What is scary for a non-pilot is, there appears to be a variation in their understanding of the system. There is also apparently a variation in standards of training and write ups in FCTM between airlines.
Another difference that is very evident. We have Byron Bailey type of pilots (A few pilots left), who rose from earlier moderately automatic aircraft, and those who have flown only the modern jets. I have seen in ME airlines cadet pilots who start on the modern jet like 777 and are now sitting on left seat. I think it will be this lot who are most vulnerable when automatic system misbehaves (or used wrongly), and you have seconds to understand and react. And when you have some one to retract the gear after a bounce, .... I need not say more. As one experience pilot once told me, "we are paid these days for the first 5 min and the last 10 min of a flight". That was on a A310 when FMS was introduced. We have gone several steps ahead with automation after that. The challenge is how to train and remember all the logic behind automation at critical times. It is a bigger challenge when one switches from Airbus to Boeing or vice versa as Asiana pilot in SFO incident testified. The discussion on this thread has become very interesting and revealing at the same time.

I've heard and read enough to form an opinion as to the most likely cause of the accident. I'm going to keep that opinion to myself to respect the wishes of the company but more importantly, to respect the pilots involved. I wasn't there, I don't know for sure.

The bounced landing is not untrained as some have said, but some airlines do not give it much time and more importantly, do not really have a formally laid out SOP for it. Automation reliance is a problem, the 320 through the trees was due to someone thinking he understood the system well, but tragically failing to appreciate a small but vital piece of knowledge. The 777 into SFO certainly had contributing factors relating to auto flight that influenced the outcome. How an individual or individuals react or recover from the startle effect is variable depending on how the atoms are put together when they were created, unfortunately some will recover from startle quicker than others. The ones you don't hear about are a combination of sheer luck and the attributes / abilities of the crew on the day. ( there for the grace of your god etc )

In another life more Orange than this one, this very subject became the main focus of a recurrent sim after various incidents around the world involving bounced landings in airbus types.. It was discovered that the programming of the spoiler extension, could under certain circumstances, result in an aircraft being high ( off a decent bounce ) with the thrust off and the spoilers at full extension. When replicated in the Sim this resulted in the aircraft dropping like a tool box onto the runway....it was an eye opener. The solution was for airbus to reprogram the spoilers to only go to half extension unless more parameters were satisfied. This stopped the aircraft plummeting quite so hard back onto the runway. When combined with a program to teach pilots how to fly away from a big bounce or land off what we described as a "skip", the hard landing rate reduced significantly.

Lessons will be learnt and Boeing may even reprogram the automation in recognition of the potential threat in situations like this. The fact remains, as people have said in previous posts, we have to keep flying the aircraft when things don't go to plan. I think it was Bob Hoover who said "keep flying through the crash".. sound advice, but how many of us really know how we would react under the same circumstances. At the end of the day we are a community and should stick together, our jobs are stressful enough with the myriad of tests, medicals and hoops that we constantly have to jump through. If we drop the ball, shouldn't we be able to rely on the one group of people who know what its like to do what we do?

BS

Well said and since you have mentioned A310, not an easy plane to handle in GA, how many of us here discussing did base training on an a/c and not in FFS. Touch and go's with basic rules, no reverse, flaps 1 notch up and trim for T/O and TOGA and I can tell feels much different and that is exactly what new gen pilots need, get out of the box and use your common sense....

The last time those contributing to this thread encountered bounces on landing, some more frequency than others, was probably during their first 20+ hours of basic flying training.

Bounce, decaying airspeed, select the appropriate attitude and apply power (and a touch of right rudder) ~ escape from danger, and climb away. Remember?

Automation in recent accidents has a lot to answer for, as it appears to wipe clean the memory of what you were taught (the basics) in the first place.

And as Cygnet46 (as was at PIK) would say of aircraft "it is just an aeroplane"

Swiss cheese..............

Cygnet66 (PIK)

Don't forget that there is also a variation in ticket prices. As a pilot it frustrates me when I see ticket prices that will barely cover the cost of the fuel for that sector. The effectiveness and size of the training department, the quality of candidates attracted by salaries, the number of tasks completed on Engineering checks, the number of pilots per airframe etc are all elements that affect the bottom line.
If Airline A has ten people in their training department, pays pilots the minimum they can get away with considering the market, runs the bare minimum tasks per Engineering check that the manufacturer allows and has pilots flying the maximum number of hours per month/ year that is considered legal, and offers cheap tickets, then Airline B who has a bigger more expensive training department ( ie better trained staff) , runs an extra 200 tasks for the same Engineering check, and has more pilots per airframe allowing well rested crews to operate at their best is stuffed. They have to compete on the price of the ticket, but their overheads are greater.
The "lean operators" drag the other operators down to their level. The only way to combat this is to have sensible minimum standards legislated. Unfortunately, the regulators don't often meet with pilot/ engineering groups ( those who understand the realities of safety) , they meet with the lawyers and accountants who run the Airlines and come up with rules like the recently introduced duty and flight time limitations that can see people who are not fit to operate a lawn mower in charge of a jet with 400 people on it.

EK pilots don't use A/P on for LDG's unless it is autoland LDG required to keep recency and knowing the wx conditions at the time of arr that was the case.
Personally I would also disengage A/T at the same time...anytime not happy with outcome of LDG phase go for GA and announce it so the PM knows your intentions, it looks like in case of B777 you need to press TOGA switches twice, first time for FD guidance and second time for your thrust in order to climb out


in case of 521, maybe PF accidentally disengaged A/T instead of selecting GA and here we go....no FD bars no thrust and bad decision of PM raising gear up on positive climb (induced by bounce)

what are the conditions for TOGA to be active again after 2sec?? :rolleyes:

There is a thread running in Tech Log called B777 logic - balked landing/GA after touch down

framer: "I see ticket prices that will barely cover the cost of the fuel"

You are right to be concerned about the quality of training, but I wouldn't focus too much on the lowest ticket prices. An airline like BA may well decide to compete with EZY on price, but only offer a handful of cheap seats. A lot of airlines will have ten different fare classes. O'Leary once said that his highest paying pax were those going to a funeral. I believe him.

I noticed this kind of comment appearing in the Fly Dubai Accident thread. Non pilots cannot believe how pilots don't understand x/y/z.

Well for the non pilots out there the reason for this is that Aircraft Manuals are written to tell you the absolute legal minimum to understand how to fly the aircraft. Sometimes this is useful, (ie knowing to much can hurt you) or it can lead to an accident when you get a weird failure or unusual circumstance.

That is why there are adverts for all these 'extra' manuals and technical APPs you can purchase. Bascially the real manual don't tell you enough. This compounded by some Check Airman who are frustrated Aeronautical Engineers.

Maybe it's time the manufacturers change their approach to manual writing?

I come back to an earlier posting (#495) where i suggested, that a "Pull-Up" procedure might save the day.

We have a rejected landing with a technique, a bounced landing with a technique, we might have a low level wind-shear with a technique and a threatening rapid loss of altitude at low level with a technique.

What do they have in common, what is their greatest threat?
- > Yes, an impending heavy ground contact/crash.
What would you do if a ground impact is looming?
- > Yes, a "Pull-Up" escape manoeuvre would probably be appropriate.

The perfidious thing here is, that a warning that would lead to such a manoeuvre is inhibited at this stage!

As we can read on these pages, there is much debate about when is what inhibited or when can what be engaged......
Simply imagine the average John Doe, with the average actual training and the average thin booklets we are given, and paint your own picture of how fast and accurately he can assess which of the above hazards he has actually encountered and which manoeuvre he should apply. All this during his initial stunning moments when rapid action is crucial.

I guess all Monday morning quarterbacks, especially the ones who have themselves never touched a -300) would pretend that this is easy, just know your aircraft and blah, blah ...

I come back to my suggestion:
For such situations, so close to ground, with that much drag and at a certain stage with very low power, I strongly suggest that recommending ONE simple manoeuvre, that might not be the most elegant, but the one with the best chances to get out of impending trouble, could be the "Pull-Up" procedure, even without GPWS warning. At least it would leave no doubt and take you out of automation doubt.

Once out of trouble, have your PM switch on the AT, engage TOGA and then continue as per normal GA procedure.

(let me speculate on this accident, you might shoot me down: I guess the pilots went the other way around: First a normal GA with gear up, then when it turned ugly, went for a W/S escape procedure. It didn't work. Maybe the other way around, with pull-up first and GA after, it might have.)

Boeing don't publish a balked landing procedure. It's left up to certain operators and regulators to train their own procedures. However it is basic airmanship to ensure full thrust is applied. A pull up manoeuvre is a bad idea as you would need to pull the nose up to 20 degrees whereas a rejected landing requires maintaining the current attitude until climbing safely away when a normal go around can be followed (TOGA, retracting flaps to go-around setting, then gear).

From these valuable contributions by 777 pilots you get the sense we are getting close to the explanation for this crash. There were similar discussions around the Rostov accident. ( note, effectively same operator as 521). And there have been others. Presumably the majors, Boeing and Airbus get to read or hear about these discussions in an ad hoc way. Perhaps if a pilots body (IFALPA?) pulled the most relevant ones into a combined document and sent it to them, some weight would be added to the discussions which are presumably going on in the marbled halls. Meanwhile although the explanation must already be known by some in the know, we have to be patient. The clock is ticking to 3rd September when UAE GCAA are required to issue first report.

That is not how the sytem works.

In our baulked landing procedure we are told never to trust that the TOGA switches will function (inhibited if the ground air switches have functioned and the spoilers deployed) after a bounce or deep landing but instead to believe they haven't and advance max manual full thrust. (as someone stated earlier on the thread full arms length push!)

After take off, hit the TOGA switches, but keep the throttles forward in the case that the thrust levers will attempt to motor backwards. Once TOGA is engaged and the FMA's have been confirmed by the NHP (SOP FMA call out after take off) then you can relax on the throttles and monitor them.

The initial report will, hopefully, contain some actual facts for discussion.

There are several aspects in this thread, all of which might contribute to an accident, but perhaps none in isolation.
Speculation about accidents requires a wider view; not just a question of if pilots know how the guidance system works or not, or know and forget, or misjudge the situation, or just follow SOPs. Or if the system design, operation, or training are inconsistent, or overly complex in rare or unexpected circumstances.

There are assumptions behind all of these aspects; why the system was designed so, the expectations for use, consideration of human performance and limitations, likelihood of encountering situations.
As an example, there are sound arguments for using automation during the approach (AT, FD) and for standardising operations, but it is unlikely that these can accommodate every scenario in increasingly complex and high tempo operations. Thus if training focuses on always following SOPs, and the SOP is to select TOGA, then the policy must account for, or accept situations where the required actions will deviate from the standard. This implies that the policies, training programs, documents and practical training must similarly involve a common understanding of the system and operation.

There can be many viewpoints for solutions and conflict resolution, some may involve a specific scenarios, or training, or policy, but none (I venture) will be without assumption and compromise. Safety improvements will require that several aspects be addressed simultaneously; I.e. reconsider the design, the content or policy of SOPs. Undertake wider investigations of an apparent increase in bounced landings, hazards of tailwind operation, human-automation interaction, and changing norms and expectations of modern society re training and knowledge.
Don't expect to find answers in one accident, one view, one idea, or in one solution; look at all operations for vulnerabilities, assumptions, and the consequences of not knowing, forgetting, or by just following SOPs.

I hear what you are saying oldchina but the lowest ticket price wouldn't be quite so low if the regulators had sensible bottom lines when it comes to things like training and pilot flight/duty times. The Rostov accident mentioned above by Portmanteau is a case in point. The Captain had resigned siting fatigue, it was 4am, he was half way through an eleven hour duty, truck drivers in civilised countries are not allowed to do the kind of duty hours that crew were doing and are required to have more rest breaks out of the vehicle. Yet it was legal.
The Airlines will always compete to get their overheads as low as possible, it is the applied law that determines how low they go. The ticket prices are a reflection of this problem and that is why I focussed on them.
I'm not in any way suggesting fatigue was a factor in the 521 crash, just using the above example as a way of highlighting that the regs are the bottom line whether it be maintenance, training,or flight/duty limits and they need to be sensible.