khaosai

The other point is that you never firewall the throttles. The engines are expensive bits of kit.

Weight balance etc is worked out before the pilot sits in the front left hand seat.

Take off and climb speeds are worked our.

I'd be more worried about being hit by a motorbike than dropping out of the sky.

Here is what it looks for me landing a Cessna in Mae Hong Son which has a mountain at the end of the runway. This would scare you

non pilots. And for the pilots yes I'm high on the vasi but it gets bumpy sliding down that approach.

Looks like fun.

Harry: "Does your method use a reduced flap setting for the takeoff and hence a longer take off run?"

That is not a normal procedure with this.

http://www.boeing.com/commercial/aeromagazine/articles/qtr_4_08/article_05_1.html

so this is not the procedure used due to fuel saving.

Using the onboard performance tool computer will allow the crew to generate take off speeds, flap and thrust setting. You input certain information into the computer and it calculates and outputs the performance figures for you. Quite often the flap setting will be lower with higher speeds utilised to get airborne. It's a clever piece of kit, even ensuring you don't exceed the tire speed limits when taking off at high altitude airports.

It's more accurate than the old paper charts and thus the airlines save money. It's like any computer tho, garbage in, garbage out, so you need strict procedures in place to ensure safety remains a priority. It works very well for both take off and landing performance.

Swerved to avoid the flip flop at the bottom of the picture.

Hi,

large aircraft in the holding pattern do look quite close on occasion with the required separation being a minimum of 1000 feet vertically. Do you think the other aircraft was at the same level ? Where you aware of any pitch and thrust change when you witnessed the other aircraft ?

The traffic collision avoidance manoeuvre should be flown smoothly and to be honest most passengers would not be aware of it happening.

The controlling at Heathrow is amongst the best in the world but both pilots and controllers make mistakes from time to time.

There is no harm in you reporting it, but I can assure you air traffic control and both airlines involved will have filled reports if it did occur.

For your main questions I recommend the forthcoming NTSB final report. I have this hypothesis that the pilot's last seconds action to increase the aircraft's pitch to climb may have resulted in the tail-strike on the seawall. Total speculation on my part, but I'd love to know their actual altitude when they realized they were too low and slow. Maybe the NTSB will speculate about what could have been done to avoid the seawall in the last seconds in their report (other than to fly a stabilized approach to avoid the situation).

As to your opinion on aircraft control and in particular as it relates to this incident, perhaps you should watch this [over] reliance-on-automation training lecture by a professional pilot:

http://www.thaivisa.com/forum/topic/651859-boeing-777-plane-crash-lands-at-san-francisco-airport/page-15#entry6622025

I apologize for the details in it (don't know why), but I, and many others, found it to be informative as well as entertaining.

PS (your original question): The crew of the AF 447 incident had just about four minutes to correct their aircraft's stall and did not. Even as high as 10,000 feet with its rapid descent they may not have been able to save the aircraft even if they had accurately deduced what it was doing and acted immediately and correctly.

great video.

Hi,

That is a great video with great delivery highlighting the potential threats with automated aircraft. There are some great pointers that will remain valid for a long time to come. The statement, fly the aircraft needs to be the pilots primary task.

Thing have moved on in the industry with regards to the level of automation and the improvement in the aircraft systems. The B777 is a good example where the auto throttle use is recommended even when in manual flight, however may be disconnected to maintain proficiency. The Airbus 380 traffic collision avoidance manoeuvre is now fully automated and flown by the autopilot and auto throttle.

Auto flight systems can reduce workload and increase safety but it's important for the pilot to decide on what level of automation to use to achieve those goals. If the level of automation is not performing as expected then the pilot should reduce the automation to ensure proper control. Quite often it's lack of understanding of the system that causes the problems. The systems that are designed are excellent but need to be fully understood.

Summary of final two NTSB briefings:

http://www.airsafenews.com/2013/07/summary-of-final-two-ntsb-briefings-on.html

this bit is interesting:

Cockpit automation and its role in the crash

As mentioned in a previous AirSafeNews.com article, use of the autothrottle by the crew to maintain speed was an issue because although the crew was heard on the CVR stating that the target speed was 137 knots, the aircraft was significantly slower than that speed before the crash. In Wednesday's press conference, the NTSB stated that there were five distinct autothrottle modes used in flight, and in the last 2.5 minutes of flight, there were several autothrottle and autopilot modes used.

As explained by the NTSB, the autopilot helps pilots manage pitch, roll, attitude, and heading; while the autothrottle helps to control speed or thrust. The two systems can work together, and the NTSB has to determine, with the help of Boeing, the following:

• Whether autopilot and autothrottle modes were commanded by the pilots or activated inadvertently,
• How the various autopilot and autothrottle modes are designed to work, and
• What are the ways the systems are expected to respond in the various modes.

=> maybe some of the 777 pilots here can comment on reasons why the autothrottle didn't maintain speed.

There is this from page 12 of this thread: http://www.thaivisa.com/forum/topic/651859-boeing-777-plane-crash-lands-at-san-francisco-airport/page-12#entry6600692

Also, buried somewhere in this thread, I believe there is an in-depth, overkill technical description of the auto-throttle subsystem that leaves one (me, anyway) wondering how it could ever work due to its complexity.

The Professional Pilot's (and others) Rumor Network has 122 pages of speculation: [link removed by author - oops]

Knock yourself out (reading all the speculation, I mean).

ah yes, very informative post by Khaosai - I overlooked it.

One of my questions - if these altimeters fail in the last approach phase and if it further causes a throttle retard, do the pilots reasonably have enough time to react?

Hi,

I personally think you would have enough time to react.

This is an interesting Thai website in English Lingo.

Here is a link to a really bizarre post, with many pics, I have no idea if this is in Thailand or not, but, non the less a great intro to this site.

http://thailandaviation.blogspot.com/2010/10/abandoned-soviet-aircraft-with-bba-14.html

PS, the link is above, don't even bother trying to guess it !!!

Saw something similar in Ann Summers !

This is 100% in Thailand.

What is its significance?

This is a guess - spotting one ! Fire Away Gentlemen

Maybe I am a whack job, but if every politician learnt to jump (not a tandem) the world would be a much better place, but most jumpers I know are extremely safety conscious individuals. (Of course there are some real nutters as well that are down right frig-gen nuts). If I was filthy rich, I would glide and jump every week.

Hi,

Cool pic, but even if someone offered me a million dollars to do so, I could never jump out of an aircraft. Granted it must be some buzz, but not for me.

I am pretty sure that the Slide designers did not have in their design brief :-

" able to still functional after a cartwheel crash and subsequent complete mangling of fuselage"

Any designer who can do that and still hit fuel and load economy requirements is a magician

I think the brief would have been more about the slides not activating except for the very specific actions of the crew on the controls. Certainly something was a little too easily triggered and there will probably be an AD on it soon.

Not prob brief, I mean input requirements for slide design.

As u know " cabin crew arm and cross check doors " is basically hooking the slide to the door threshold, so when u open the door, the slide activates.

With a twisting and deformed frame, it should still not deploy, but with all the buckling and twisting - not easy to w sure deployment does not happen.

Hi,

Good point. The doors on the 777 are of the translating plug type. They first move in, then up, then out. If structural damage has occurred this may not be possible. It's probably something that the crew never even considered would occur.

The plane can carry 349 pax maximum.

Apparently, the man was throwing water at people. He was also reported to have been acting in a bizarre manner.

The diversion was made after the man allegedly made physical threats to the crew when the FAs tried to calm him.

The passenger was taken to the Queen Elizabeth II Health Sciences Centre in Halifax for a psychiatric assessment.

550 passengers is an exaggeration, but I have seen in the region of 450 passengers on board. 428 seats and then infants in addition to that.

The disruptive passenger definition is quite specific and really is a last resort if other methods of defusing the situation have failed.

Hi,

Nice video and a nice flight deck.

On the B777 you first get a auto call out at 2500 ft based on the radio altimeter. The next auto call out will be at 1000ft based on the barometric altimeter so indicates height above the airfield. The next call will be at 500ft, again a barometric altitude. The next call will be at 200ft based on radio altimeter. You will also get a call when 100ft above the minima you have set for landing and then an actual call when at the minima which will say "minimums". From there as the nose disappears below the threshold you will get more radio altimeter calls of 50, 40, 30, 20 and 10.

The only auto call out that refers to speed is the V1 decision speed call out on take off.

These calls may vary between operators I guess but they are all geared towards awareness. In the older generation aircraft the height calls would typically be made by the pilot monitoring and acknowledged by the pilot flying so the concept is the same here.

In addition to the above some airlines have enhanced ground proximity indications to let you know if you configuration is not correct, your high, your landing too fast or landing too long.

Hi Skippy,

Where do you fly the simulator, Bangkok ?

Flight Experience - Convent Road Silom rd - BKK.

737-800 FBT

First flight full fee, after that 50% discount. About 3400 baht per hour after discount

I like it, just wished I could afford it. Flight Experience Wonder if I can put the time in my log book?

Hi,

It looks like a well set up operation. Might take the wife along next time in Bangkok but not sure she will like my back seat input !

Just had a look on the website. It's a Cat B approved sim and says if you hold an Australian licence you can log certain things.

Regarding the airport authorities having the instrument approach off the air due to work in progress, I personally don't think any blame can be apportioned to them. There is a fantastic RNAV approach available to that runway which gives you accurate lateral and vertical guidance to the decision height. A crew could have made the decision to fly that approach prior to departure.

Some airlines publish a list of different approaches in order of preference. The RNAV approach will be most likely be given the same level of preference as and instrument landing system (ILS).

Just did SFO on real 737 FBT flight simulator 1st visual approach , fully manual, clear day ( easy peasy - confused how 2 pros could crash last week ), The rest was with fog down to 500 ft AGL. Not easy - quite hard for novice flying fully manual but managed. Last landing on localizer and on glide slope ( first fog landing had to go around )

Hi Skippy,

Where do you fly the simulator, Bangkok ?

Air America ish,

Sky van ish, herc ish, nomad ish.....anhedral wing....russian......not a clue to be honest.

Hi,

Airspeed is definitely something that is required in the pilots scan. The NTSB will have the answers on what human factors on that particular day contributed to this accident.

Regarding the auto throttle logic, it will not provide stall protection when below 100ft. Certain engaged vertical modes will not provide stall protection when the auto throttle is in HOLD mode, i.e no auto throttle wake up. I am not sure when or if they got the stick shaker, but based on the speeds quoted its pretty likely they did.

If the auto throttle is in SPD mode it will correct for any change in speed, fast or slow. So lets say below 100ft on the approach you get a speed reduction the thrust will increase to counter act that change until between 25ft and 50ft where it goes to IDLE as part of the landing.

It could be that lack of technical knowledge was a contributory factor in all of this where the auto throttle modes were not fully understood.

Exiting a plane somewhere in Thailand..

Airline and model please.

IMG_1735.JPG

Airbus....320 ?

Guess on Cebu Pacific

Hi,

I think it was some sort of demo or air show. Sends a chill through the spine watching that. I don't have any experience operating the airbus product but it's certainly very automated. The A380 can even do a traffic collision avoidance manoeuvre via the automation which is impressive.

Airline management promote a fairly healthy safety culture so any subsequent delay will be of no major concern

And once again, I refer you to GF072...

Terrible accident which should have been avoided. I should stress my personal experience with airline management has been a good one when it comes to flight safety.

Interesting Article on why the autothrottle did or didnt do what the pilots expected.

Asiana 214 - Why the NTSB’s Focus On Autothrottles

"Each time the automation mode is changed there is the possibility that what the autothrottles will be trying to do will also be changed. It turns out that one of the things the autothrottles will do in certain modes is an automatic application of power to rescue the jet from a critically slow airspeed condition. It also turns out that here are also a number autothrottle functions that are influenced by the height of the aircraft above the ground - such as power reduction at 100 feet for touchdown during an automatic landing. But in a case where this is the desired function the logic of the automation cannot also be to add power rescue the aircraft from a slow condition when this is a natural and desired precursor to a landing. So in this case the programming of the autoland function establishes a logic in the automation that is different from other ways it would behave autoland is not programmed."

Link article on asiana214 autothrottle

Hi,

Just to clear up some inconsistencies from above the auto throttle on this aircraft will go to idle between 25ft and 50 ft during an auto land. The aircraft flare manoeuvre begins at 50ft during an auto land which differs slightly from the recommended technique when landing manually where is states to flare between 20ft and 30ft. These figures are in the Boeing flight crew operating manual and is what is taught to pilots new to the aircraft.

Hi IMA_FARANG,

there are some very good explanations to the questions you have asked. Here is my take on things too.

Any delay due to a go around will vary depending on many factors. These factors may include how busy the airport is, whether the controllers have radar available, what sort of weather is existing at the airport etc. it will also depend on the type of approach you have flow as some approaches have different missed approach procedures. I would plan on an additional 15 minutes and 2000 kg of fuel as an approximation. As stated by another poster if fuel management becomes an issue then you can expedite your approach by declaring a mayday call.

Airline management promote a fairly healthy safety culture so any subsequent delay will be of no major concern but may pose challenges if the airline operate the hub and spoke type arrangement where passengers have to make connecting flights. As a pilot that is way down the list of priorities and would not influence continuing an unstable approach.

Its important to note that go arounds are a normal manoeuvre which occur on a daily basis, occurring perhaps more often at certain airports than others for a multitude of reasons. The airline would not be punished for such a manoeuvre.

As mentioned in previous posts most airlines will have strict stabilisation criteria that must be met by a specific point on the approach. This is non negotiable and must result in a go around. Management would take a very dim view on any crew continuing such an approach. It's a huge threat to the industry and is totally unacceptable. You may find that the aircraft will destabilise after that point, again that must be mitigated by going around.

Lets start one hour before top of descent. A good briefing highlighting specific threats is a good starting place. Both crew members in the loop. Have some specific gates as you descend. These aircraft don't go down and slow down that well so you need to allow a lot of distance to manage the descent effectively. Lets say your cruising along at 35000 ft. You would need to start descent at approximately 120 miles from touchdown. its pretty simple arithmetic that most pilots will continue to do in their minds during descent to ensure the automation is providing them with good sound information. Speed management is also an important consideration so also requires planning. The B777 will take approximately 10 miles and 2 minutes to slow from 310 knots to 220 knots in level flight, more so whilst descending.

If its not working out then do something about it. Tell air traffic control your not happy and request extended track miles prior to landing. If they are not willing to accomodate your request then be assertive and tell them what you will do. Always rememeber who is in charge.

I don't know the Asiana stabilisation requirements but I would say this approach should have been discontinued at the latest 1000ft or if became unstable below that height then immediately.

It's ok for me to sit here typing in a relaxed environment, but on the day with lots of potential distractions who really knows what went on in that flight deck. Sad that people lost lives and I hope that lessons can be learned to further enhance what already is a pretty safe industry.

Hi,

There is an bulletin in circulation highlighting some failures that could cause some potential issues whilst landing. Not sure if its a factor in this accident but it's something the investigators will be looking at as a potential cause.

The aircraft has 3 radio altimeters fitted. When one or more produce erroneous altitude indications there can be some additional effects and conflicting information on the aircraft instrumentation.

No auto throttle wake up protection available which could have been an issue here.

False or missing ground proximity warnings.

Erroneous radio altitude on either primary flight display.

Airspeed errors.

Auto throttle retard.

Auto throttle disconnect and inability to re engage in speed mode.

Engines in ground idle rather than approach idle thus increased acceleration times in the event of a go around.

The recommendation, whether flying manually or with automation is to monitor the primary flight instruments, namely speed and altitude and the flight mode annunciations and to keep one hand on the thrust levers even with the auto throttle engaged.

Certainly food for thought but not one I have heard mentioned so far thus probably been discounted.

Hi skippy,

Yeah interesting dynamics for sure. Lets see how they were communicating in the flight deck. Regarding the idle descent, may well be the case. All goes back to approach stabilisation, as power setting appropriate to the configuration is a parameter that needs to be considered.

Resilience needs to emphasized during crew training.

I always thought that hearing was the first sense to go until one of my peers recently highlighted that it is in fact touch that goes first then hearing.

Hi,

I need to dig out some text books relating to performance A aircraft, but I think for the stall speed calculation it is as follows:

Reference speed for landing (VREF) is 1.3 or 1.23 times the stall speed.

When the landing weight is worked out based on the actual zero fuel weight added to the expected fuel on touchdown we get the VREF speed which is then entered into the flight management computer (FMC) via the control display unit. That then appears on the primary flight display speed tape located on the left hand side.

With auto throttle engaged we actually fly that speed plus an additional 5 knots, with the objective being to touch down at VREF. The actual landing distance is worked out based on that additional 5 knots of speed.

The B777 auto throttle logic during approach is designed to cope with gusty conditions and as a result is recommended during all phases of flight.

Engine spool up times will vary depending on the phase of flight. During approach with the flaps selected then the engines are at approach idle. The reaction time for the engine in the event of a go around will therefore be reduced. Based on experience my estimate is that the engine will reach full thrust after approx 4 to 5 seconds.

On approach the engines N1 parameter would be in the region of 55 to 60% when fully configured, i.e. gear down and flaps set at 30. The equivalent engine pressure ratio (EPR) would be 1.060 to 1.070. After landing you should cancel reverse thrust completely when the engines have reached idle which is approx 23% N1. From that you can see that full thrust would be achieved sooner on approach than lets say during take off.