The report, unfortunately, failed to factually state how the airplane got airborne and climbed to 85 feet with the engines at idle.
Unless I am missing something, by rearranging the time sequence one can notice the thrust being at idle since approximately time 08:37:20. The TL were actioned at time 08:37:35 and took two seconds to start spooling up the engines. it would suggest that those 85 feet were reached through a bounce.
Regarding automation, it's all very neat to have it. Yet, airlines are spending lots of money in such "awareness" equipment and forcing pilots to abide by those calls and leave airmanship aside. In the old days (I am old), accidents were caused by reasons which were supposed to have been mitigated, or eliminated by automation. However, automation is becoming a new reason for accidents, different types of accidents.

It's called latent energy

When touchdown occurred, they were at Vref+15 (162kts), which is plenty fast enough to get airborne again with idle thrust if you rotate the aeroplane. They traded speed for height and got to 85’ and Vref-13 (134kts) before coming back down to earth again, which taking account of the drag from gear and flap and time on the ground seems reasonable. 162kts to 134kts is equivalent to c.360’ worth of height in a no-drag world, without adding in gains from an increasing headwind. No bounce required to explain the behaviour.

Nothing latent about it, it is just energy.

Looks like I've stumbled into airliners.net by mistake.

It was taught to me way back during training as being valid when it was actually a landing that was long and outside the touchdown zone e.g. Not as a result of temporary displaced threshold, database error etc. I can also recall the TRE stating its the only case it's mandatory and not optional (mayyyte) to go around for a RAAS call.
Now applying common sense and airmanship, I would be prepared to go into the office and say I didn't go around as a captain because it wasn't valid on the basis I had 2.5km of runway even if I landed halfway down it.. but this is the place where trainers divert a 380 because of OM-A auto callouts and not cause of common sense.

It does seem to me as a bystander that the definition of the TDZ as the first 3000ft of runway is somewhat arbitrary. You could land a good 2000ft past the end of the TDZ at Dubai and still have more runway left than landing at the correct point on many other runways around the world.

Going around after missing the TDZ by 600ft when there is still over 8000ft ahead does seem like overkill!

Basic skill sets
 

There are some basic skills that you need to master in order to be called a professional pilot, for example, being able to take off and or carry out an RTO within the confines of the runway, also you need to be able to land an aircraft and or carry out simple missed approach.

What is being missed by all of the posts seemingly blaming the unnecessary auto call out for a GA, is simply this; regardless of whether or not the GA was done or not, the basic skill of being able to carry out a bog standard missed approach was for some reason beyond this crew. This is a very basic skill that should be in every pilots toolbox. To put it another way, if the crew had of ignored the call out to do a GA, and instead continued with the landing, the gap in their skill set would not have been exposed this time but would still have existed.

This loss of basic skills seems to be becoming a common theme as more and more crew have not had any experience on anything other than modern late generation airliners. By way of recent examples, AF447, Asiana @ SFO, Air Asia X off Sumatra, and the ATR72 in Taiwan.

There is a very strong emphasis in our company for crew to use automatics in all phases of flight to the maximum extent, and manual flying is actively discouraged. This, I believe is a very flawed approach to both training and normal operations. The net result is, I believe, a very substantial eroding of basic hand flying skills in our company, and it would appear from recent industry accident history, that this insidious global reduction in basic skills is a very extensive problem.

Unfortunately the blame for this accident will be sheeted home to the crew, and the real culprits, the middle management of airlines around the world responsible for running training departments, and setting SOP's will come out unscathed.

What gazumped said!:ok:

Yes, it seems to be yet another case of 'automation complacency' and a lack of understanding about what the automatics will or won't do.

I wonder. The preliminary report includes a section on 'Go-around procedure' and also includes extracts from Emirates 777 FCOM and FCTM. Those documents are obviously in the investigators' sights.

The FCOM and FCTM are vague about the procedures required for a go-around after touchdown. The FCOM mentions that the TOGA switches are inhibited after touchdown, but the FCTM only states that 'an automatic go-around cannot be initiated after touchdown'. The FCTM also states that 'if a go-around is initiated after touchdown but before thrust reverser selection, continue with normal go-around procedures'. Normal procedures would imply pressing the TOGA switches, but in this case they are inhibited. It seems to me there's plenty of room for confusion if pilots don't fully understand the limitations of the system and nowhere does it state that the thrust levers must be manually advanced.

Agreed gazumped
The logical outcome of this will be the continuation of such accidents until something changes. Ultimately the industry managers will have to come to the realisation that in order to gain another step down in the accident rate, they must acknowledge the problem & there must be training in both automation AND manual skills.

Buzzbox
 

Nowhere is it stated on my bedroom door that I must open it before I attempt to enter. I can assure you quite emphatically that if you attempt to enter your bedroom without opening the door you will bump your head.

Likewise if you attempt a GA without pushing up the power levers you will hit the ground, I don't care what is written in what book, FCTM, OM etc, basic physics.

Also if you retract the gear without having POSITIVE rate, and you neglect to push up the power levers you will make headlines. Basic, basic stuff.

We as in industry, have been let down by very very poor training. The lack of skill displayed by this crew is not isolated to this crew, but, in my opinion very widespread, and I say again, the responsibility should sheeted home to training departments, not only in this airline but worldwide. Unfortunately this will not happen, the crew will be sacrificed, and the lapses in global crew skills will continue.

Our company has now incorporated into the next cyclic a GA from on the ground, just prior to reverse selection, talk about experts in shutting the door after the horse has bolted!

It's invalid if annunciated after touchdown, as per current Company Notam.

Am I the only one here who is troubled by the statements who claim that the pilots are entirely to blame and it was simply their incompetence that led to this accident. "Poor training, automation dependency, children of the magenta line etc etc".

Never heard of the Swiss cheese model? As for bog standard go around well if you touchdown PRIOR to the go around then it ain't a go around! Also, it's one thing to fail to fly the aircraft to within set limits through poor flying skills but another to be caught out, perhaps, in an unusual situation. Rejected a landing after touchdown isn't something I've practiced regularly but I suspect it will be part of the next cycle of sim scenarios.

I don't know out it's SOP at this airline to leave autothrottle engaged on a manual landing but if it is not then that MIGHT explain why a crew could be surprised when TOGA doesn't engage. On a similar type it works on a time/radalt combination and a colleague related to me how it surprised him when he rejected a landing during the flare. Of course when he realised what had happened he manually selected an appropriate thrust setting. In all honesty I think I may have reacted in the same way.

So perhaps we can give the crew the benefit of the doubt and not proclaim their guilt just yet. Just a thought.

BBK

The report doesn't mention whether actions were commanded and perhaps a prelim report doesn't do that anyway.

But from the timings in the report:

0837:19 RAAS aural message
0837:23 Aircraft becomes airborne
0837:27 Flap lever moved
0837:28 ATC issue clearance, which was also read back correctly
0837:29 Gear lever selected up
0837:35 Both TL's moved from idle

In a 3 second period:

The flaps were moved, presumably after being commanded
A clearance was given and read back
The gear handle was moved, again presumably after being commanded

Perhaps the readback could have been made after the gear selection. That's not clear but either way, it's a busy 3 second period.

gazumped
 

Thanks very much for the physics lesson. Really.

You might have noticed the term 'automation complacency' at the beginning of my previous post. Far too many pilots have become far too reliant on the automation instead of using it as a tool to help them fly the aircraft. The procedures and training at most airlines are geared towards the maximum use of automation and there is precious little opportunity for most pilots to practice their 'basic skills'. It's no wonder that such skills are being eroded.

In my opinion, the problem is exacerbated by poorly documented procedures and poor training. Many pilots simply do not understand the limitations of the automatic systems they have become reliant upon. When something goes wrong and the automatics don't operate as expected, those same pilots lack the currency (or training) to fall back on their basic skills.

The Final Report into EK521 will no doubt highlight many different factors that contributed to this accident. I'm betting that poor documentation of the TOGA inhibit function and its implications for a go-around after touchdown will be one of those factors.

BBK
 

My posts are not blaming the crew, but the training department of this and every other airline, mine included.

It is a sad fact that the middle management of this airline will sacrifice this crew rather that address their own inadequacies.

It did not miss my attention that just when 100% of effort should have been on flying the aircraft and monitoring the PM made a radio transmission.

Possibly whilst moving the flap lever and raising the gear?

I have to say a.net put up their decent behavior on this topic, on the other hand ft showed its dark side. Of course pprune always encourages freedom of speech more than any other forum.

For those questioning why did they pull the gear? Well the aircraft went from 0 to 85' so I would assume that qualifies as positive rate.

I agree. I think it is high time that aircraft manuals were rewritten to actually educate pilots on how and why things work, rather than just the legal minimum.

This crash might also point the fingers at a few safety 'committees' who have turned every runway in the world into something that is less than 2000m long with their whole 'touchdown zone' mantra. This combined with the way that company's seem to treat you if you land outside the touchdown zone has created a grey area of fear in where you could easily land and stop comfortably after touching down at 1050m but get a bollocking from the company or you go around and attempt a difficult maneuver which is rarely practiced with automation changing modes and a high cognitive workload.

The question for safety committees is why are you going around when you have 3000m left to stop in? On long runways the touch down zone is merely arbitrary. In reality it should be calculated on the actual physical length of available runway you are landing on. So on a short runway it will be tight on a long runway you have much more flexibility.

Chaps, Ladies

Those of us who are of some age and from the UK may remember the Schedule 8 & schedule 10 test format for the renewal of an AR/IR and the company OPC. During the schedule 8 check, the man from the ministry would come to ensure that skill retention was above the line, and that skill fade in a limited amount of items if any had not dropped below the line and affected safety.

The schedule 8 items were very limited as the skill fade, which concerned the man from the ministry were the handling of an Engine failure at V1, demonstrating bank and pitch control prior to flap retraction, transitioning to a higher bank after flap retraction, while following the needles outbound on the SID and then entering the hold correctly and not forgetting to time it.

We then had to Perform an ILS approach RAW data and at minimums perform a go-around. Finally we also needed to demonstrate a non precision approach and a full stop landing using reverse thrust with an outboard engine inoperative.

Those of you who believe that skills are eroding have no real basis to compare skill retention data. There is no scientific evidence that pilots of the years past where better than us today in skill based exercises. Actually the contrary may be true due to better engineered jets.

The mishandling of planes during critical phases of flight has not contributed to any significant accidents in the last 30 years since automation became more ubiquitous.

None of these items checked diligently on the skill retention lists of years gone by have been a problem recently in any of the accidents on the Air Asia 320, Air france 330 or on the emirates 777.

How would you be a better pilot or for that mater a better rested pilot if you hand flew any modern jet around the globe at Flight level 290 out of RVSM airspace?

One does not need to be an expert in human factors to see that wasting time flying manoeuvres that are not relevant in a particular skill set does not improve the overall skill set for a particular task as a go-around. Spending hours flying around does not make you better at performing go-arounds it makes you more tired.

The definition of skill is the ability to do something well, it's expertise. In order to become proficient in a skill one needs to be trained to do a particular task again and again.

Competence rather than skill is the ability to do something successfully, by using skills, knowledge, attitude, and it depends on ability.

As we have moved to more complex aircraft skill training is not enough. What is required is a competency based training environment, which looks at competency rather than skill.

I am not in Emirates therefore I cannot comment on the training and continuous Pilot Development these two chaps received, or for that matter what the rest of the Emirates Pilots receive.

Scientifically proven, pilots when reacting to stimulus from the environment, we don't decide on a course of action based on a classical decision making model as we don't have time. We react to what we perceive is happening and we match our reaction to what we have done in the past. We do not optimise we are satisficing, as do fire chiefs, trauma doctors and nurses and for that matter army officers on the battle field.

Our brains are woefully underpowered for what we want them to do in time critical events thus we use heuristics.

On the day the two chaps had to process within seconds, inputs from the environment, from the aircraft, combine that with knowledge and rules in the OMA and in the FCOM and then make a decision under critical time pressure and react by matching what they have done in the past with this event.

It's funny that some have criticised their reaction as unskilful. The contrary is evident. They were very skilful. Years of training worked and they retracted the FLAPS to 20 as required, they pressed the TOGA buttons, they pitched to up, they retracted the Gear as required and the PM even had the presence of mind to call Speed as he is required to do.

No one teaches or requires the PM to look at engine thrust during a go-around and good luck to all of us if we think that the solution is to include one more item in the process during a high task load event. What is required is competency training.

What is also evident is that they lacked the competency not the skill to do the manoeuvre as clearly their understanding of the automation failed them on this occasion, which is probably because they have not been trained in such an event, and or fatigue and or the startle effect.

Remember it is proven scientifically that Subject Matter Experts (pilots) in highly risky environments react to stimuli with recognition primed decisions. These actions require competency not just skill. This requires proper competency training not just skill based tick boxing.

Combine the above with fatigue as we no longer fly sensible hours rather we fly to the limits and any one of us on the day could have ended up in the same situation, as having no thrust available on a go-around is something that we are not taught and we don't expect.

There but for the grace of God go I

about positive rate issues
 

Yes IF splitting hairs - it does BUT !:=

But it seems to me that in real life - what is needed is to define positive rate as a combination of two factors Increasing air speed AND increasing altitude as a minimum useful definition.

When inerita/energy available as a function of speed is traded for altitude gain, the speed will decrease.

Via wiki- consider a bicycle or car or plane going x miles per hour and as it starts UP a hill by coasting, with NO additional power /pedaling /thrust applied it will eventually stop- and absent sufficent pedaling or power input- it wil go back DOWN the hill

Ditto for an airplane with too late application of power.:ugh:

It's interesting to see how some were advocating the use of sophisticated anemometers, criticising the lack of real time data available to the pilots when all they had to do was move the thrust levers forward :ooh:.

They were doing a GA in windshear conditions.
The wind measurements, not from anemometers, but from profilers, would have identified the exact conditions, providing SA to the crew to avoid the entire situation.

..... and that would have made the PF shove up the thrust levers ..... not :ugh::ugh:

give it a rest, your bone has no flesh. It's not about wind readings.

Come on, stop using facts life isn't that easy. BTW Where is the tin foil ........ ;)

Either you have never sat at the controls of an airliner or you are one of those pilots who should have ventured into the academic sphere. Which one is it?

Gear should not normally be retracted until pilot(s) have a known positive rate of climb with throttles advanced and producing adequate thrust. Adequate altitude, sound/vibration from engines producing thrust and the human body feeling accelerations are some clues. Training for common sense?

On AB types in this type of senario, you get all sorts of hooters & warnings ( once power applied) is the 777 the same? Not mentioned in the report but quite a distraction if you are not expecting it. (Note was part of my brief when landing at GIB as it was never trained)

If they left the Gear down
 

What would have been to outcome, if they just left the landing gear down. Would it have been a lot worse than it was?. Been thinking about this, but not keeping me awake at night. :O

Seriously, even in the best conditions, how often is a GA (Miss) undertaken flawlessly? :=

Yes, you'll get a CONFIG warning when the thrust levers are advanced and the aircraft is still on the runway with landing flap selected, the same as Airbus.

Ask yourself this, if the crew in question had used manual thrust for the last 1000ft of the last 200 approaches they had done ( ie the thrust levers were part of their subconscious motor skills) , would the accident have happened?
Serious question. Take the time to think about it. It is the heart of this accident.

Framer, I agree. Boeing's policy of using autothrottle at all times on the 777 contradicts their policy on all their other types of disengaging autothrottle when hand flying. Airlines will follow that policy because of modern litigation. It is a bad policy that erodes basic handling skills and instrument scan and creates speed/thrust complacency. Furthermore, it is poor design if pressing TOGA after touchdown, however brief, gives TOGA modes on both axis of the FD but leaves the thrust at idle - deliberate mixed modes in unique circumstances with PFD indications that you have what you selected is a recipe for this sort of accident. It doesn't mean that the crew shouldn't have been aware that the thrust levers never advanced, but Boeing are largely responsible, just like they were for the same system behaving stupidly and similar crew over-reliance and system confusion in SFO.

Framer
 

Spot on, too much reliance on A/T, same as Asiana @ SFO, and Turkish Airways @ Schipol.

Lesson for airline training departments, A/T doesn't always do what you expect, when it stuffs up close to the ground if pilots don't actually 'pilot' it turns to s$&t remarkable quickly.

Why are we having this conversation, why aren't we just hand flying more often and learning from experience that, for example 29 degrees nose up at 38,000' with AF447 is inviting disaster?

Why aren't we hand flying all visual approaches, ingraining muscle memory with exactly what control inputs and thrust lever positions 'feel right'?

The answer to both these questions need to be asked of training departments. My airline actively discourages hand flying. When I mention my concerns to management pilots they typically reply, "if crew hand fly we get more altitude busts, and hand flying above 25,000 makes you non RVSM," plus other lame excuses.

I fear that this issue will be addressed as a result of this accident, not enough paying passengers were injured, and unfortunately the crew will wrongly get the entire blame for this one.

Pressing the TOGA switches after touchdown DOES NOT give TOGA modes on the FD. The TOGA switches are inhibited after touchdown, so pressing them does absolutely nothing. The correct procedure is to advance the thrust levers manually, maintain the flap configuration, ignore the take-off configuration warning and rotate at VREF. When safely airborne, press the TOGA switches and perform a normal go-around from that point. Unfortunately, that procedure is not included in the bog-standard 777 FCOM and I'm betting most 777 pilots aren't trained how to do it.

So pushing the thrust levers up on a GA........ who would have thought? Nearly every student pilot would be able to guess that would be a good idea!

Let's look at how much Boeing trusts TOGA.....

Let's have a look at Windshear memory items:

Windshear TOGA
Thrust lever maximum (aggressively)
A/P disengage
A/T disengage
etc.....


Let's have a look at GPWS warning memory items:

A/P disengage
A/T disengage
Thrust lever maximum (aggressively)
etc....

Go Around procedure

TOGA Flap 15
Go Around attitude
Verify Go Around thrust
Positive rate Gear Up

At no point does Boeing suggest that pushing TOGA by itself is sufficient evidence that thrust has done as TOGA is supposed to do. In single syllables NEVER TRUST TOGA in critical situations, because if it doesn't do as you expect it to do, left uncorrected at a critical time, a disaster will ensue.

Airborne at second 23.
Thrust levers advanced at second 35.

That is 12 seconds. Now slowly count to 12, and realize how unbelievably long 12 seconds are.

gazumped

I have agreed with everything you have said, but for the benefit of the uninitiated, it's probably worth pointing out that attempting a go-around WITHOUT (at some stage!) pushing the TOGA buttons, puts you in a (completely different) world of hurt!!!

Going around after landing is something practiced, (if ever?) once, for every 99+ go-arounds from DA. Though I fully accept it should be briefed moderately frequently.

Depending on how rigorously this crew were trained, it may not be too surprising they got it wrong?

Whichever way you look at it, the airline itself has some very significant questions to answer!

PM should have ignored ATC and equally ATC should not have issued an instruction when it must have been apparent the crew had their hands full. Crew had not asked for any clearance for obvious reasons. The long first landing happened right opposite the tower. Why did ATC assume the crew were going for GA and not a second landing?