BugSmasher1960
Even to use a microwave or mobile phone to its full capability requires the user to study the user manual and understand the system. Most people use these devices for limited applications by seeing how others do it. It is freighting that some pilots do the same in the air while using automation. A pilot started a thread on this forum by stating his bad experienceof of auto thrust not by asking others their opinion but by directly accusing the manufacturer of being irresponsible and producing an inefficient auto thrust system and wanted to create a community protest. When someone pointed out his blunder in using ATHR he quietly disappeared. The SFO and Paphos incidents are examples of crew's ignorance of auto flight/ATHRsystem. What they thought should happen was based on the pilots insufficient knowledge of the ATHR and it simply was not designed to happen. AF447 was a massacre of the innocent because they did not have any idea of what was happening and what needed to be done and out of fear of the unknown did something so bizarre that it defies all logic. Initial training, periodic monitoring and evolving as things happen or don't happen is the way. But it must be incorporated in the SOP with consultation with the manufacturer and not one individual adding something here and another one deleting something there. Profit making is treated as something very mean but don't forget it is the very raison d'etre of airline's existence. What other purpose Airlines are run for?

JS, # 1075,
The earlier post expressed a feeling about the subject, this preceded the main point of concern.
Debating the what, where, why, of a feeling would not contribute any more to this discussion, particularly as individual experiences and viewpoints will differ depending on experience and knowledge.
Few pilots are familiar with determining what a crew may have been doing based on an external video of a landing. The video lacks context, the crew's awareness, intent, the actions, system behaviour, auto / manual landing, etc. Most of the comments were about the human, often being supposition and seeking error and blame, and with little thought of an aircraft system error or malfunction.
Much better to start from the position that the pilots were trying to do their best, and then consider if this was sufficient for the situation, then why, and what could contribute to that.

Many debates in this forum consider how something should be done vs what actually happens in operation; flying by the book (SOP) vs actual flying.
Furthermore with the hindsight of this thread the quoted Boeing text appears incomplete - "When the go-around mode is initiated," - the text assumes a particular situation. What happens in specific, rare conditions where GA mode is not initiated even though selected ... Someone has overlooked the point; a hole in the logic or the text, a latent factor, a hole in the Swiss Cheese.

Dani - when you ease your hand pressure from the throttles, without taking your hand off the throttles, you will feel the throttles attempt to move backwards. If that were to occur at that moment you just disconnect the a/t's and assume manual control.

Been there, done that, got the T shirt, as have many Boeing pilots when there's been a disconnect betwen the expected level of automation and the actually level of automation.

A perfect example of applying the wrong SOP. But why did he do it? Perhaps it was because he wasn't trained to think, didn't think, or even just didn't understand aerodynamics. Perhaps it was fear, perhaps it was lack of training, or perhaps it was a lack of aptitude. Or perhaps it was a combination of all of those things.

The thing is I often see the blind application of SOP's without any thought, along with blind faith in the absolute reliability of automation. Things have been learnt by rote & are applied by rote. I don't know if it is a training issue or an aptitude issue, or perhaps a bit of both. Perhaps laziness & lack of professional discipline, thought & study are part of the issue as well.

Whatever the reasons, if the industry continues to move down the path it is on & pilots forget how to be pilots & lose basic flying skills, there will continue to be an increasing number of accidents. People will continue to ask "why did he/she do that?"

A major incident on a major airline in a major airport and so far there has been no official information released at all. It really is a wonder that this state of affairs has been allowed.
Happened 20 days ago.

I agree partially with that. If the SFO crew had been using manual thrust for the last two months the accident would never have occurred. The auto throttle should be nothing more than one tool that can be used to move the thrust levers, the other method, and it should be the primary go to method, is the pilots hand. We have set up a system whereby that is not the case, the go to method is the automated mode. It may not say that in an FCOM but that is the reality. The brain quickly develops preferred methods of achieving tasks and we have created the situation where the preferred method is the automation and manual manipulation is a bit foreign to us. So ingnorence of the system is a problem, but not as big a problem as re wiring our brains to favour automation over manual handling.
My two cents.

misd-agin,

Yes it is the pilots responsibility to look through the FD and not blindly follow it 100% of the time (try telling that to some sim checkers!). Usually it is fine but as we all know that isn't always the case..... that is where basic knowledge of pitch and thrust settings comes in. What happened to AF447 was tragic but easily avoidable and could easily be the same all over again with EK.

framer
I am afraid the primary method has to be automation backed up by manual skill. A pilot cannot keep switching off automation at the drop of a hat because he has insufficient knowledge of automation. Without AP You won't be able to stay in RVSM or operate CAT3. SFO happened basically because they didn't monitor speed and didn't know how the auto throttle works.

Not to mention the inappropriate use of FLCH SPD mode during an approach!

I hear what you are saying Villas and I don't necessarily disagree with you. The topic is complex and I think we may be speaking at cross-purposes a little bit.
I agree 100%. So if we break it down in search of a resolution we get,
1/ Why didn't they know how the Auto-throttle system works?
2/ Why were they not monitoring the airspeed?
I'm talking about number two. The automation is the reason that they were not monitoring the airspeed. There are many different ways that we could legislate to make manual manipulation less foreign to our current generation of airline pilots. Thats the specific part of the problem I was talking about.
Cheers

Under Annex 13 to the Chicago Convention, States in charge of an investigation must submit a Preliminary Report to ICAO within thirty days of the date of the accident, unless the Accident/Incident Data Report has been sent by that time. Preliminary Reports may be marked as confidential or remain public at the investigating State’s discretion. So, ten days to go to see if the preliminary report is made public.
It's interesting that the Chairman and CEO of Emirates, Sheik Ahmed bin Saeed al-Maktoum, is also:
- Chairman of Dubai Airports,
- President of the Dubai Civil Aviation Authority, and
- a Board Member of the General Civil Aviation Authority of UAE.
When you have someone sitting on the Board of the investigating authority who also represents the airline, the airport and the regulatory authority involved in the accident you might expect the whole process might move along a little faster; you certainly wouldn't expect any conflicting or argumentative testimony from any of the three parties the Sheik represents.

The probability of getting any objective and truthful report about a plane crash from any arab state is zero..even if the Sheik-your-money-or-your-life is not involved

At SFO the time below Vref was measured in tens of seconds(45??). It's professionally stunning that three pilots never looked at their airspeed for that amount of time. Or they did see the airspeed and never called it out or took corrective action.

FullWings
Unless the Dunlops are already being tucked up in bed ... and there is one of the key questions. As an onlooker, I may presume that even ONE Trent 982 could overcome the extra drag of leaving the gear down for another 15 seconds. Does the PM make the call? If the PM was the FO, then we might expect life will be uncomfortable.

Of course, the report may well confirm that the FD gave a cabin PA very shortly before landing to the effect that there was a problem with the gear. Or it may not ...

Alright already! We all know that... what's the solution? I'll tell you: get pilots engaged in "flying" a bit more! You lot can waffle on about SOPs and not knowing the ins and outs of the grand airbii and intricacies of other types all you like but if pilots don't have basic flying skill, mainly because they are not allowed to keep current doing it, then don't blame them when the likes of AF447 happens.

Isn't that so true? Simple little things like turning off the flight director in the climb or descent gives a little relief from the boredom of the automatics and sharpens the scan - providing that is not against the Koran or the local Deity of course.:E

In the simulator, unless during type rating or recurrent training he is not hamstrung by regulatory box ticking or a frowning management, then so much depends on the personal attitude of the simulator instructor towards practice of raw data hand flying. Simulator practice can be made enjoyable if the right personality is in the instructor seat.

To those who remark sarcastically that simulator training was never meant to be enjoyable, I say that is quite wrong. Even Maths can be made enjoyable in the class-room given the right teacher. But have the grim wrong teacher and maths becomes a loathed subject as many students have found out. Me for one. Same with simulator instructors where the Horror Box is often rightly named. Wise management will allow for more than just a couple of minutes manual non-automatics practice during simulator training. That policy, if regularly implemented, would have saved hundreds of lives in the past and also in the future.

I agree however sadly simulator time hasn't been about training or sharpening skills in a very long time, it is all box ticking exercises now. I can count on one hand the number of times I have actually received some constructive feedback in the debrief or given some time "off the record" to actually do some training, it just never happens anymore. The simulator has been abused and is so far from what it was originally designed to do.

The pilots aren't banned from looking at their airspeed indicators during the approach OR banned from looking at their pitch attitude while in cruise. Two simple things that would have made a world of difference.

My whole airline didn't know how the auto throttles worked. Our manuals, approved by the FAA, and our training, told us that with the Autothrottles ON, we would be speed protected. We didn't know about this autothrottle (non-speed protection) mode until the SFO accident.

The above is a MAJOR problem. Boeing knew how the system worked, but didn't communicate that effectively to my airline.

The NTSB report on Asiana OK214 flagged the issue of the crew's poor understanding of the airplane’s autoflight system as a contributing factor (Finding 10. As a result of complexities in the 777 AFCS and inadequacies in related training and documentation, the pilot flying had an inaccurate understanding of how the autopilot flight director system and autothrottle interacted to control airspeed, which led to his inadvertent deactivation of automatic airspeed control.)
The NTSB recommended that the Federal Aviation Administration (FAA):
- Require Boeing to develop enhanced 777 training that will improve flight crew understanding of autothrottle modes and automatic activation system logic through improved documentation, courseware, and instructor training.
- Once the enhanced Boeing 777 training has been developed, as requested in Safety Recommendation, require operators and training providers to provide this training to 777 pilots.
- Require Boeing to revise its 777 Flight Crew Training Manual stall protection demonstration to include an explanation and demonstration of the circumstances in which the autothrottle does not provide low speed protection.
The NTSB also recommended that Boeing:
- Revise your 777 Flight Crew Operating Manual to include a specific statement that when the autopilot is off and both flight director switches are turned off, the autothrottle mode goes to speed (SPD) mode and maintains the mode control panel-selected speed.
When that report was released two years ago, the FAA and Boeing danced around those recommendations such that they are currently classified as "Open - Unacceptable Response" on the NTSB register.
Interestingly, in noting the FAA's inaction as unacceptable, the NTSB stated that they were "... concerned that training improvements will be unnecessarily delayed and that the “lessons learned” will be limited to a single airline’s analysis, that neither the FAA nor other Boeing 777 operators will be aware of the problems found, and that no resultant training revisions will be made."
That was December 2014; 18 months later on a hot August day in Dubai and the NTSB's concerns seem to be prescient given what probably transpired.

I don’t have the “gear down landing rate of climb” table for the RR892-powered 777-300 but at TOGA thrust and ALAW you’re probably right.

PM will call positive rate, based on altimeter, VSI, radalt, etc. It is quite possible that those instruments did confirm the aircraft climbing but the airspeed would likely have had a nice negative trend arrow on it. In moments of stress we fall back on what we normally do and there may have been just enough confirmation to trigger the call from the PM.

atpcliff
I would like a discussion between you and M. Mouse because as TRE on 777 and 787 he feels it is categorically stated by Boeing. After SFO accident even FAA asked Boeing to have a look at the design off course it was one of the 31 factors contributing to the crash. It will be interesting for non 777 guys.

Full Wings:

If you have access to a 777 FCTM you will find that postive climb is NOT called with reference to the VSI or radalt but solely with reference to the altimeter.

About a third of the people I see in the sim are doing it wrong.

The VSI can and will show positive climb when the gear is on the ground because it is only a measure of acceleration.

The radalt reading is the delayed product of an algorithm and not a direct reading of height. Thus it can and will be tricked by aggressive handling in pitch as one might do in a windshear encounter near the ground.

1104 comments later and we no nearer the truth of a crew surviving, photo'd fully documented FDR/CVR crash. Not even a sniff of a preliminary guess. Must make the rest of the B777 guys a bit nervous, not to mention Boeing.

Rat, Boeing, Rolls, and the NTSB are on-site and involved in the investigation, granted in an 'advisory' role - the local authority will be prime. So at some level Boeing knows what's going on. As I've mentioned previously, everyone involved is under something of a gag order - all information release should come through the local authority. Even around Boeing, information sharing is on a 'need to know' basis - in part to avoid leaks although it also comes in handy when the inevitable lawsuits start popping up. :rolleyes: That's why I can publicly post about it - I literally know nothing about the investigation (insert poor Sargent Schultz imitation here :E). If I was in the know I'd need to STFU.
When the final report comes out, if Boeing or Rolls feel they've been unfairly thrown under the bus, then they can publicly disagree with the report - not before (see the Egypt Air 767 for a prime example).

I am curious to know what the crew did after what appears to have been the second landing. Would they have shut engines down or selected reverse thrust? Seems quite remarkable the aircraft slid along in a straight line for around 3000m with wings level. Helped no doubt by the large flat fairing over the gear bays. As the speed wound down did a point come when the aircraft had to tip to the left or right or were the crew able to influence this outcome in any way? It went right and then executed a turn to the right of around 155 degrees presumably because the right engine dug in. Aircraft comes to rest on approx heading of 275 and pictures show wind appearing to be blowing across aircraft at right angles ie from 185. How fortuitous this all happened with runway to spare ( though not much). I find it hard to believe reports that pax were unaware all this was going on.

Yes, that’s what we do but some airlines have different SOPs. FCOM > FCTM.
You can get it on up-sloping runways, too. It’s small effect but if you lost an engine at MTOW you might not be expecting much of a climb rate to begin with...

I alway thought the minimum concept of raising the gear was when the a/c fuselage was at a greater height than an extended gear AND increasing. Doesn't that fall under the Positive Climb definition which is confirmed primarily by an altimeter backed-up with a (I)VSI?

What happens is that pilots with reasonable survival instinct would become proactive and intervene MANUALLY by pulling back the stick/yoke/wheel and shoving and holding the throttles/thrust levers forward. At most Airlines there is a statement on page 1 of any operating manual which says that: Trained procedures, manuals and SOPs cannot cover every conceivable operating snafu, and that pilots are not constraint from exercising common sense. No?

Been there-done that and it worked well, thank goodness. Night landing in a 737 to Guam in the Western Pacific Region. ILS to 6L and when visual, cleared to side step to land 6R. Panam 747 holding on taxiway joining the approach end of both runways and awaiting clearance to line up and take off 6L as soon as we confirmed sidestepping to land 6R. In other words he was holding at 90 degrees to 6L with his tail facing the threshold of 6R about 60 metres away.


We called sidestepping and ATC cleared Panam to take off on 6L. No ILS or VASIS on 6R. We came over the threshold of 6R slightly high due no glide slope. It was dark.
About 50 feet the wings rocked and the 737 seemed to fall out of the sky with PM calling urgently Vref minus 20. It happened so quickly. I instinctively firewalled the engines and pulled back to 15 degrees body angle on instruments and waited for the impact. As it turned out we caught the sink in time and the wheels did not hit before we climbed away.

The cause of the incident was the 747 had opened up to break-away thrust while on the taxiway at 90 degrees to 6R and we caught the jet blast from his engines as we passed behind him at 50 feet. If the crew of the 747 had acted immediately to taxi on to 6L when cleared by ATC the incident would have never happened. Instead the 747 crew failed to commence taxiing when cleared, maybe briefing or reading checklists while staying put on the cross taxiway before starting to roll on to 6L. It was an uncomfortable moment for us.

A thought re the lack of GA near touchdown. Is the aircraft still fail operational?
CS AWO 316
(a) The aircraft must be capable of safely executing a go-around from any point on the approach to touchdown in all configurations to be certificated. The manoeuvre may not require exceptional piloting skill, alertness or strength and must ensure that the aeroplane remains within the obstacle limitation surface for a Category II or III precision approach runway as specified in Annex 14 Chicago Convention.
(b) For decision heights below 15 m (50 ft) automatic go-around must be provided.
(c) When automatic go-around is provided, it must be available down to touchdown.
(d) When automatic go-around is engaged, subsequent ground contact should not cause its disengagement.

Also AMC AC 316 paras 1.1 b, and 2 c.

1 Safety Considerations
1.1 Effects of Contact with the Runway: For aircraft in which a go-around from a very low altitude may result in inadvertent runway contact, the safety of the procedure should be established giving consideration to at least the following:
The guidance information and control provided by the go-around mode should be retained and be shown to have safe and acceptable characteristics throughout the manoeuvre,
Other systems (e.g. automatic throttle, brakes, spoilers, reverse thrust and alerting systems) should not operate in a way that would adversely affect the safety of the go-around manoeuvre.

1.2 Inadvertent Go-around Selection. Inadvertent selection of go-around mode after touchdown should have no adverse effect on the ability of the aircraft to safely roll out and stop.

2 Performance
Height losses from a range of altitudes during the approach and flare should be determined when under automatic control and when using the landing guidance system as appropriate.
Height losses may be determined by flight testing (with typically 10 flight demonstrated go-around) supported by simulation.
The simulation should evaluate the effects of variation in parameters, such as weight, centre of gravity, configuration and wind, and show correlation with the flight test results.
Normal procedures for a go-around with all engines operating should be followed.

PEI 3721: Are you implying the B777 doesn't meet those requirements? If so, I'd suggest you are wrong!

There are suggestions throughout these posts that on this occasion, and possibly another, that there could be capability inadequacies in the aircraft and the crew. It is up to an investigation to determine the contributions of man, machine and environment.

autoflight: That may be true, but the B777 DOES meet all the certification requirements mentioned in PEI 3721's post above. Perhaps those requirements need a rethink (ie automatic go-around capability after touchdown??).

Buzz, the thought #1112 was not intended to imply anything; certainly not right or wrong. The objective was to reflect on the general uncertainty and judgement required in aviation, including certification.
Assuming that the nuances of the GA mode were fully appreciated during certification (was this the flt test or the operational evaluation team), then the certification is based on the judgement of skill, alertness, safety, and acceptable characteristics. This opinion was most likely based on experiences and standards at that time, and where projecting this over the lifetime of the aircraft type would be extremely difficult.

Certification judgement remains a 'dark art'; any one view is unlikely to represent a world wide standard, nor accommodate changing standards overtime. Thus even though an aircraft has been certificated, certain aspects are open ended, based on assumptions which may longer hold true.
What operators lack is a clear explanation of what assumptions have been made during certification (skill, knowledge, workload, etc), i.e. operators lack guidance for determining procedures and training requirements. Also, any assumption has to be balanced by the likely human performance in a range of situations, not all of which can be foreseen.

It's all very well for NTSB to ask FAA/Boeing to review system designs, but unless there is a clearly defined standard for judgement, then the industry should not be surprised by a conclusion "that the system was alright then, so it's all right now".
This issue represents the growing gap between 'flying as imagined' at the time of certification and how flying actually takes place today; the regulators have to appreciate changes in the industry and avoid resorting to 'more training' for systems knowledge and awareness, where human limitations in memory and recall associated with these factors probably contributed to the problem.
You cannot expect to solve a problem with the same features which contribute to it.

Low level windshear encounter or wake encounter at low altitude/speed?

That was my first thought underfire and then a botched go-around.

Regards

Orion Man

There has been a lot of talk about windshear, pressure altitudes and temperatures, you have to realise these conditions are pretty well standard operating conditions for us in the summer, nothing special, it's what we are paid to deal with and do so on a routine basis. Are they a causal factor of course, but pitch without power is never going to make an airplane go upwards for very long

It was the COMBINATION of a significantly elevated Density Altitude (Pressure, Humidity and Temperature) AND the existence of at least TWO different types of windshear, that are both likely to have been contributing factors to this accident.

Datum, until we know more you could surmise that 521 was just unlucky. That combination was experienced by a stream of aircraft, some went around some landed safely one did not...