Sponsor Message:
Civil Aviation Forum
My Starred Topics | Profile | New Topic | Forum Index | Help | Search 
TK B738 Crash At Amsterdam - Part 8  
User currently offlineHB-IWC From Indonesia, joined Sep 2000, 4516 posts, RR: 72
Posted (5 years 9 months 3 weeks 1 day 3 hours ago) and read 19997 times:

Please continue the discussion on this topic in this new thread.

Previous installments:

Part 1

Part 2

Part 3

Part 4

Part 5

Part 6

Part 7

230 replies: All unread, showing first 25:
 
User currently offlineBoeing747_600 From United States of America, joined Oct 1999, 1295 posts, RR: 0
Reply 1, posted (5 years 9 months 3 weeks 1 day 2 hours ago) and read 19932 times:

727forever wrote (in Reply 213 of Part 7)

As far as the FO not keeping his hands on the throttles, I agree that this was very poor form. His hands should never have left the throttles from the beginning of the approach.


Absolutely. Even after the Capt yelled "My Control!" or something to that effect, he should have kept his left hand on the throttles until the Capt's right hand grabbed them. I wonder if this is procedurally required.

Finally, its a tragedy that the Capt just didnt hit the TO/GA switch the moment he took control which would have cancelled all active modes and made the A/T command max go around power ... unless

(a) The Capt didnt take control until the active pitch mode went to FLARE

AND

(b) TO/GA is inhibited during FLARE (I dont know if this is the case)


Either way, shouldnt the TO/GA switch have been hit once the stall warning went off?!


User currently offline727forever From United States of America, joined Jan 2006, 794 posts, RR: 5
Reply 2, posted (5 years 9 months 3 weeks 1 day 2 hours ago) and read 19867 times:



Quoting Boeing747_600 (Reply 1):
Either way, shouldnt the TO/GA switch have been hit once the stall warning went off?!

I do not fly the 737, but on other Boeings and other manufacturer designs the AP will disconnect upon stick shaker activation. AP will do many things, but flying out of a stall is not one of them.

Additionally, I started thinking about the FOs hands. How do we know that his hand wasn't on the thrust lever? We are assuming that it wasn't, but do we really know. Logically it was probably a deal where he reached over pushed it up, saw that the airplane was entering the stall, and put his left hand back on the yoke to try to fly out of the stall. Only problem was the thrust levers just came on back to idle, probably before the engines even had time to spool much above the idle. My airline teaches that in a stall the first thing you do is disconnect the AT with you thumb as your pushing the thrust levers to the firewall. Then before letting go you press TO/GA.

727forever



727forever
User currently offlineNCB From , joined Dec 1969, posts, RR:
Reply 3, posted (5 years 9 months 3 weeks 1 day 1 hour ago) and read 19782 times:

Osis, I don't think that Boeing will easily admit that there is a problem with their parameters and that they are at fault. They will try to lay all blame on the dead crew.

Considering that there is only about 20 seconds between the airspeed departing Vref and the stick shaker, the crew did get a little chance of noticing it. F/O was probably too busy flying the airplane back on the glideslope, but captain could have seen the decrease in speed.
Nevertheless, when I know that the A/T is on and unless I hear the disengagement tone, I wouldn't be checking on the airspeed as often as if I were flying manually.

The problem is that the logic system is not logic.
The left radio altimeter reads something different from the other RA, the pressure altimeter, the GPS computed altitude, the INS computed altitude, overriding all these systems and making the entire system assume that the airplane is flaring...
Sorry but who designed this? No built-in redundancy.

The warning the crew has gotten, is the landing gear warning.
Since the aircraft can retard its own levers with A/T in flare mode I don't expect a "retard" call. That is probably the only warning the crew may have gotten.

Airbus does not have the same philosophy, the pilot must retard the levers in flare mode, probably partly due to the fact that Airbus considers it important that crew take the levers to retard instead of giving the machine the ability to do that. That is also a very necessary feature considering that Airbus throttles levers do not move when A/T is set as opposed to Boeing levers.

Whoever commented upon it, I'm not against twins, I'm against unsafe aircraft.


User currently offlineJoeCanuck From Canada, joined Dec 2005, 5478 posts, RR: 31
Reply 4, posted (5 years 9 months 3 weeks 1 day 1 hour ago) and read 19729 times:



Quoting NCB (Reply 3):
The warning the crew has gotten, is the landing gear warning.
Since the aircraft can retard its own levers with A/T in flare mode I don't expect a "retard" call. That is probably the only warning the crew may have gotten.

From the summary, below, of the report yesterday, it looks like the crew recognized the radio altimeter failed, when it happened, and "did not see this as a problem".

So, if the report is correct, the crew did have a warning. Whether or not they associated the failure of the radio altimeter with the throttles going into flare mode is another matter. If they didn't, why didn't they?

That seems to be a question that probably needs asking. Is it a training problem with airlines? Is it not spelled out clearly enough in the Boeing literature?

Quoted from TK B738 Crash At Amsterdam - Part 7, PW100, Post 4

Quote:
* Dutch Safety Safety Board: Safety Alert to Boeing:
*** Recommend to review Fligth Manual B737: if radio altimeters are inop, autothrottles and autopilot must not be engaged;


* 9 casualties;
* 80 wounded, 28 still in hospital;
* TK 1951 ETA 10.40
* No problems until last moment;
* 3 qualified pilots in cockpit;
* 1st officer flying the aircraft;
* TK1951 received landing clearance, no delay in approach, direct approach;
* Autopliot engaged during approach;
* DFDR and CVR in good condition;
* at 1950 ft, LH radio altimeter step change read out from 1950 ft to -8ft;
* this faulty altitude was also reported to autothrottle;

* CVR: crew saw altimeter discrepancy;
* Crew did not see this discrepancy as problem, and lowered gear as gear warning was activated due to -8 ft radio alt;
* As a result, autothrottle went into retard mode at 1950 ft rather than 50 ft due to faulty radio altimeter read out;
* Aircraft slowed to stall speed;
* Aircraft flight controls [autothrottles] thought airplane was above runway and went into "flare mode";

* At 450 ft stall warning /stick shaher activated;
* power was increased immediately [!] [TOGA?], but power increase was too late to continue flight;




What the...?
User currently offlineMir From United States of America, joined Jan 2004, 21864 posts, RR: 55
Reply 5, posted (5 years 9 months 3 weeks 1 day ago) and read 19579 times:

From the previous thread:

Quoting Md80fanatic (Reply 208):
The AT logic would be the first suspect in this chain and no scutiny should be spared in this area. The first hole in the cheese exposed the second and the third. Pilots failing to deal with an automation error is only possible when there IS an error. Removing this error is primary, long before blaming the pilots IMO.

Or if you look at it another way, even if the first hole of the AT logic is exploited, there still should be other layers without holes (the pilots) that would prevent the accident from happening. Those layers failed in this case.

I have a problem with the idea that the pilots should get a pass just because their error wasn't the first in the chain. Of course the issue with the RA and AT link needs to be seriously looked at, but the fact remains that there was time for the pilots to figure out that something wasn't right - be it by the change in the FMA, by the fact that they were approaching Vref and the throttles weren't spooling up, or whatever other clues might have existed. Technical errors are going to happen, but they shouldn't necessarily bring down an airplane.

It all goes back to monitoring the automation. The automation should never fly the airplane - that's the pilot's job. If the pilot chooses to delegate that job to the automation in order to free up some brainpower for other tasks, that's fine, but all he is doing is choosing to fly the airplane by using the buttons and knobs on the glareshield rather than by moving the yoke. The need to cross-check the instruments to make sure the airplane is doing what he wants it to still remains. The autopilot is a flight control, not another pilot. It cannot evaluate its own actions to determine whether it is doing the right thing - that is the pilot's responsibility. Unfortunately, it appears that that responsibility was neglected.

-Mir



7 billion, one nation, imagination...it's a beautiful day
User currently offlineStarlionblue From Greenland, joined Feb 2004, 17174 posts, RR: 66
Reply 6, posted (5 years 9 months 3 weeks 1 day ago) and read 19556 times:

Quoting Doktor 71


Turkish Pilots Association Vice president Mr. Izgi was quoted by Turkish Media saying the information presented today was not satisfactory and they expcted more satisfactory results since quite a long time has passed since investigation started. He also said that no pilot would fly with such a big difference in altitude (meaning discrepancy between both altimeters) and that no pilot would neglect such an error.

After that TALPA president Mr. Yilmaz just said that "they haven´t received the original investigation report yet" and that "at the press conference earlier this day just initial findings were presented and after detailed investigation the final report will be published".


Seriously, have these guys no shame? Time to gracefully back up instead of just making even bigger a**es of themselves...



"There are no stupid questions, but there are a lot of inquisitive idiots."
User currently offlineSudden From Sweden, joined Jul 2001, 4130 posts, RR: 6
Reply 7, posted (5 years 9 months 3 weeks 1 day ago) and read 19423 times:

Very sad and unfortunate all this! By all means!

But I do wonder how the aircraft was able to be on idle for 100 seconds!!?? without the crew noticing that they must have been way below glideslope due to abnormal sink rate as a result of lost forward motion.
I am in no position of assuming the pilots did what ever wrong as we have experts out there that are paid to do that (laywers?), but it seems a bit strange to me.
Or am I missing something?

Aim for the sky!
Sudden



When in doubt, flat out!
User currently offlineCaribillo From Spain, joined Jul 2006, 218 posts, RR: 0
Reply 8, posted (5 years 9 months 3 weeks 1 day ago) and read 19364 times:

Quoting Mir (Reply 5):

It all goes back to monitoring the automation. The automation should never fly the airplane - that's the pilot's job. If the pilot chooses to delegate that job to the automation in order to free up some brainpower for other tasks, that's fine, but all he is doing is choosing to fly the airplane by using the buttons and knobs on the glareshield rather than by moving the yoke. The need to cross-check the instruments to make sure the airplane is doing what he wants it to still remains. The autopilot is a flight control, not another pilot. It cannot evaluate its own actions to determine whether it is doing the right thing - that is the pilot's responsibility. Unfortunately, it appears that that responsibility was neglected.

I thought that was pilot's job. Being there if something goes wrong and solve the situation.

If that isn't pilot's job, what is the difference between a 100% automatic airplane without crew and the current comercial airliners?
If the two guys on the cockpit can't solve that, isn't cheaper save the crew and fly the plane in automatic (it is possible actually) or flying it from the ground (possible aswell)?
Why did they need a whole life training and they couldn't solve that (they were on three inside the cockpit!)?

Let's the fire start...

[Edited 2009-03-05 00:49:56]

[Edited 2009-03-05 00:50:56]


Red, orange and yellow...with a big crown!
User currently offlineSpitfire From France, joined Feb 2001, 801 posts, RR: 7
Reply 9, posted (5 years 9 months 3 weeks 23 hours ago) and read 19332 times:

Quoting Sudden (Reply 7):
But I do wonder how the aircraft was able to be on idle for 100 seconds!!??

Just read the posts concerning the "Low Drag, Low Noise" approaches in part 7.... You can be on idle till around 1000ft (even lower...) without problem. But everything has to be closely monitored during that phase of flight.....

Even with AT/AP approaches there must be at least one pilot monitoring their actions and ready to take over at any time.

[Edited 2009-03-05 01:01:38]


Sabena ... Never to be forgotten (12 years already , what a shame !! )
User currently offlineMir From United States of America, joined Jan 2004, 21864 posts, RR: 55
Reply 10, posted (5 years 9 months 3 weeks 23 hours ago) and read 19274 times:



Quoting Starlionblue (Reply 6):
Seriously, have these guys no shame?

No, I don't think they do. Well, if they want to look like arrogant idiots....

Quoting Caribillo (Reply 8):
If the two guys on the cockpit can't solve that, isn't cheaper save the crew and fly the plane in automatic (it is possible actually) or flying it from the ground (possible aswell)?

It would be cheaper. But I can confidently say that this is not the only time that automation has screwed up - I've been on such a flight, and it ended with a perfectly normal landing. So despite these little hiccups, we don't have that many crashes. Even the earlier flights involving this very airplane where the radar altimeter malfunctioned didn't end in accidents. So it's fair to say that while this crew may not have been able to solve the problems, many other crews are. And that's plenty of argument for having pilots up front. In fact, this accident is a prime reason why you MUST have pilots up front - this crew at least had a chance to make things right. A fully automatic airplane would have been doomed the moment the radar altimeter failure happened.

-Mir



7 billion, one nation, imagination...it's a beautiful day
User currently offlineSbkom From United States of America, joined Feb 2009, 85 posts, RR: 0
Reply 11, posted (5 years 9 months 3 weeks 23 hours ago) and read 19272 times:



Quoting Mir (Reply 5):
It all goes back to monitoring the automation. The automation should never fly the airplane - that's the pilot's job.

Well, it is an interesting point. Yes, technically you are right. Pilot is responsible for everything.

Somewhere in the previous archived thread I think MD80fanatic said something like:
-pilots fail, it is the pilot's fault
-automation fails, it is still the pilot's fault

If there is automation, we should be able to rely on it to some degree. Yes, they are not human, they are there to help humans fly the plane, but still, they should share some sort of responsibility.
Imagine the conditions are worse, it is pitch black night, rain, or snow. The same CAT 1 approach will most likely end in disaster. Maybe we will blame the pilots less that time.

Again imagine a CAT3c approach with autoland. We rely solely on the automation.

I am not discounting the fact that the RA failure should have been understood and managed better. I also have questions about the crew's understanding that the LH AR was feeding the AT despite the fact that Channel B was selected.

I still think that there are logical design problems in the a/c systems. Like redundancy, failover, alerting. Also some poor user interface problems like the autopilot can disengage when the controls are moved, but you need to press a button to disengage the AT. The fact that AP and AT are two different systems, in fact, they should be one system. Just because they are invented in different times shouldn't make them different systems.

The bottomline is, yes, the pilots failed to intervene, or were late identifying the problems, but if there is automation, it should be more reliable and responsible too. After all, technically it is not impossible to make a better system. If the technology is there, it should have been used to a better degree to prevent these failures.

It is not humanly possible to control everything that the computers do, we are not here to QA those systems.

Sad, very sad. This could have been prevented.

Quoting Starlionblue (Reply 6):
Seriously, have these guys no shame?

Forget TALPA. They don't know what they are talking about. They are in such a denial state, blinded by saving the face of the pilots that they are being ridicule. Instead of fingerpointing, they should contribute to the findings and improve the safety, training and operating procedures of their pilots.


User currently offlineSudden From Sweden, joined Jul 2001, 4130 posts, RR: 6
Reply 12, posted (5 years 9 months 3 weeks 23 hours ago) and read 19267 times:



Quoting Spitfire (Reply 9):
Just read the posts concerning the "Low Drag, Low Noise" approaches in part 7.... You can be on idle till around 1000ft (even lower...) without problem. But everything has to be closely monitor during that phase of flight.....

Even with AT/AP approaches there must be a pilot monitoring their actions and ready to take over at any time.

Thanks for clarification.

Aim for the sky!
Sudden



When in doubt, flat out!
User currently offlineLifelinerOne From Netherlands, joined Nov 2003, 1938 posts, RR: 8
Reply 13, posted (5 years 9 months 3 weeks 23 hours ago) and read 19234 times:



Quoting Starlionblue (Reply 6):
Seriously, have these guys no shame? Time to gracefully back up instead of just making even bigger a**es of themselves...

TALPA and Turkish newspapers this morning are still suggesting that the crash was caused by a wind shear or vortex by the B757 of Norhwest which landed before TK1951. They claim their pilots are still heroes who prevented further loss of lives by crash landing the plane under difficult circumstances. They claim that the Dutch ATC is to blame and that this is a cover up!

Cheers!  wave 



Only Those Who Sleep Don't Make Mistakes
User currently offlineSpitfire From France, joined Feb 2001, 801 posts, RR: 7
Reply 14, posted (5 years 9 months 3 weeks 23 hours ago) and read 19211 times:

Here you can find about those 100 seconds in idle :




This message is sent to all 737-100,-200,-300,-400,-500,-600,-700,-800,-900,-BBJ customers and to respective Boeing Field Service bases, Regional Directors, the Air Transport Association, International Air Transport Association, and Airline Resident Representatives.

SERVICE REQUEST ID: 1-1228079803
ACCOUNT: Boeing Correspondence (MOM)
DUE DATE: 10-Mar-2009
PRODUCT TYPE: Airplane
PRODUCT LINE: 737
PRODUCT: 737-100,-200,-300,-400,-500,-600,-700,-800,-900,-BBJ
ATA: 3400-00

SUBJECT: 737-800 TC-JGE Accident at Schiphol Airport, Amsterdam - 25 February 2009

REFERENCES:
/A/ 1-1222489391 Dated 25 February 2009

Reference /A/ provides Boeing's previous fleet communication on the subject event. The US NTSB, FAA, Boeing, the Turkish DGCA, the operator, the UK AAIB, and the French BEA continue to actively support the Dutch Safety Board's (DSB) investigation of this accident.

The DSB has released a statement on the progress of the investigation and has approved the release of the following information.

While the complex investigation is just beginning, certain facts have emerged from work completed thus far:

- To date, no evidence has been found of bird strike, engine or airframe icing, wake turbulence or windshear.
- There was adequate fuel on board the airplane during the entire flight.
- Both engines responded normally to throttle inputs during the entire flight.
- The airplane responded normally to flight control inputs throughout the flight.

The Digital Flight Data Recorder (DFDR) data indicates that the crew was using autopilot B and the autothrottle for an ILS (Instrument Landing System) approach to runway 18R at Amsterdam Schiphol airport. During the approach, the right Low Range Radio Altimeter (LRRA) was providing accurate data and the left LRRA was providing an erroneous reading of -7 to -8 feet. When descending through approximately 2000 feet the autothrottle, which uses the left radio altimeter data, transitioned to landing flare mode and retarded the throttles to the idle stop.
The throttles remained at the idle stop for approximately 100 seconds during which time the airspeed decreased to approximately 40 knots below the selected approach speed.

The two LRRA systems provide height above ground readings to several aircraft systems including the instrument displays, autothrottle, autopilots and configuration/ground proximity warning. If one LRRA provides erroneous altitude readings, typical flight deck effects, which require flight crew intervention whether or not accompanied by an LRRA fault flag, include:

- Large differences between displayed radio altitudes, including radio altitude readings of -8 feet in flight.
- Inability to engage both autopilots in dual channel APP (Approach) mode
- Unexpected removal of the Flight Director Command Bars during approach
- Unexpected Configuration Warnings during approach, go-around and initial climb after takeoff
- Premature FMA (Flight Mode Annunciation) indicating autothrottle RETARD mode during approach phase with the airplane above 27 feet AGL. There will also be corresponding throttle movement towards the idle stop. Additionally, the FMA will continue to indicate RETARD after the throttles have reached the idle stop

Boeing Recommended Action
- Boeing recommends operators inform flight crews of the above investigation details and the DSB interim report when it is released. In addition, crews should be reminded to carefully monitor primary flight instruments (airspeed, attitude etc.) and the FMA for autoflight modes. More information can be found in the Boeing 737 Flight Crew Training Manual and Flight Crew Operations Manual.

Operators who experience any of the flight deck effects described above should consult the troubleshooting instructions contained in the 737 Airplane Maintenance Manual. Further, 737-NG operators may wish to review 737NG-FTD-34-09001 which provides information specific for the 737-NG installation. Initial investigations suggest that a similar sequence of events and flight deck indications are theoretically possible on the 737-100/-200/-300/-400/-500. Consequently the above recommendations also apply to earlier 737 models.





I hope this is clear for everybody now ! It's official from Boeing ! (not from a newspaper or armchair pilot....)

[Edited 2009-03-05 01:30:22]


Sabena ... Never to be forgotten (12 years already , what a shame !! )
User currently offlineSbkom From United States of America, joined Feb 2009, 85 posts, RR: 0
Reply 15, posted (5 years 9 months 3 weeks 23 hours ago) and read 19175 times:



Quoting LifelinerOne (Reply 13):
TALPA and Turkish newspapers this morning are still suggesting that the crash was caused by a wind shear or vortex by the B757 of Norhwest which landed before TK1951. They claim their pilots are still heroes who prevented further loss of lives by crash landing the plane under difficult circumstances. They claim that the Dutch ATC is to blame and that this is a cover up!

Could you give me some links? I couldn't find such news this morning.
Thanks.


User currently offlineSbkom From United States of America, joined Feb 2009, 85 posts, RR: 0
Reply 16, posted (5 years 9 months 3 weeks 23 hours ago) and read 19152 times:



Quoting Spitfire (Reply 14):
Unexpected removal of the Flight Director Command Bars during approach

What is this? Can someone explain please?


User currently offlineJpiddink From Netherlands, joined Feb 2009, 105 posts, RR: 0
Reply 17, posted (5 years 9 months 3 weeks 23 hours ago) and read 19118 times:

As I am not a pilot, I have some difficulty understanding exactly what happened in the automated systems. Perhaps someone else can shed some light on this question that keeps popping into my mind:

If the (one) altimeter suddenly dropped from 1950 ft to -8 feet, and as a result the throttles went to idle and the airplane began to loose altitude, somehow either
- the failing altimeter should have sunk below ground level (I assume the Polderbaan is at about -20 ft?), resulting in the automated systems recognizing they weren't functioning properly anymore (flying below the earth surface), or
- the failing altimeter still showed -8 ft and didn't drop with throttles in idle. Shouldn't that also be an indication that the automated systems recognize as unexplainable behavior, hence shut down A/T and/or A/P?

Just my 2 pence trying to gasp what happened.


User currently offlineExSR From Germany, joined Sep 2008, 83 posts, RR: 0
Reply 18, posted (5 years 9 months 3 weeks 23 hours ago) and read 19098 times:



Quoting LifelinerOne (Reply 13):
Quoting Starlionblue (Reply 6):
Seriously, have these guys no shame? Time to gracefully back up instead of just making even bigger a**es of themselves...

TALPA and Turkish newspapers this morning are still suggesting that the crash was caused by a wind shear or vortex by the B757 of Norhwest which landed before TK1951. They claim their pilots are still heroes who prevented further loss of lives by crash landing the plane under difficult circumstances. They claim that the Dutch ATC is to blame and that this is a cover up!

Cheers!   

For some cultures, simply everything is a question of honour. It´s more than ridiculous.


User currently offlineDogBreath From United Kingdom, joined Mar 2008, 265 posts, RR: 0
Reply 19, posted (5 years 9 months 3 weeks 23 hours ago) and read 19089 times:

As a B737-800 pilot, I find this accident appalling. Who was flying the airplane?? What was the 'safety pilot' doing? Another case study for failures in CRM (Crew Resource Management).

Regardless of the incorrect data from the Captains (LH) Radio Altimeter, this should not have brought down this aircraft. Proper scan technique and adherence to the basics of aviation should have detected the reducing airspeed.
I can't for the life of me understand why the safety pilot didn't monitor the FO? If he had been doing his job, this would've most likley ended differently.

No aviation accident happens due to a single event (in this case the incorrect RA), it happens due to a number of problems. The James Reason model uses the swiss cheese theory that if all the holes of defence are lined up then an accident will happen.

Notable to me is that this aircraft had been written up on previous flights due to the same fault. Any good Captain will read back through the Tech Log and highlight to himself and the FO any recurring faults and to note this as a potential 'threat' for the flight. Whether this was done or not we may never know, but if it was, surely they it have been known during the approach. Reading the report is says that they discussed the incorrect RA and dismissed it as not a problem (can't remember the exact words from the report).

What was the Safety Pilot doing? His job is purely and simply to monitor the actions of the FO. From his position on the jump seat he has the best seat in the house to monitor what going on. Like the Flight Engineer of old, he can easily scan the entire flight deck to provide valuable added support to both pilots. Why didn't he see the dangerous situation developing?

There seems to be so much debate on this topic about the incorrect data from the RA, about 100 seconds of Idle thrust, blah, blah, blah. A shame that the aircraft was allowed to develop into a deep stall in the first place. It's not like they had a split second to make a judgement call and act on it they had plenty of time to save this aircraft. It should not have happened if the pilots had been doing their job.



Truth, Honour, Loyalty
User currently offlineMir From United States of America, joined Jan 2004, 21864 posts, RR: 55
Reply 20, posted (5 years 9 months 3 weeks 22 hours ago) and read 19007 times:

Quoting Sbkom (Reply 11):
Somewhere in the previous archived thread I think MD80fanatic said something like:
-pilots fail, it is the pilot's fault
-automation fails, it is still the pilot's fault

The second line should be "if the automation fails, and the pilots don't catch it, it's the pilot's fault". Which is entirely correct. If this were a FBW airplane, and the throttles just refused to respond to pilot input, then I wouldn't fault the pilots - they tried to solve the issue, but the automation wouldn't let them. But that isn't the case here - they didn't try to fix the issue until too late.

Quoting Sbkom (Reply 11):
If there is automation, we should be able to rely on it to some degree.

The automation was very reliable - it did exactly what it was programmed to do. Yes, the input was bad, and that's not the fault of the pilots. But that's why the pilots are there - as I said before, the autopilot cannot think in the big picture, it can only figure out what it should be doing at that precise moment in time. The pilots are the ones with the big picture, and they need to ensure that whatever the autopilot is doing is in line with that big picture.

Quoting Sbkom (Reply 11):
Again imagine a CAT3c approach with autoland. We rely solely on the automation.

Not at all. We rely solely on the automation to make control inputs, because the automation can do that more precisely than a human. But the pilots are still VERY involved, making sure that the autopilot is doing what it needs to be doing, and ready to intervene at the first sign of trouble. Remember: the pilots are ALWAYS flying. They can choose to fly through the autopilot, but they should never let the autopilot make decisions for them.

Consider this: if one pilot had been flying by hand, without autopilot or autothrottles, and noticed that the RA was reading -8, and, deciding that the airplane was on the ground, flared and brought the throttles to idle accordingly, should the other pilot have intervened, seeing an incorrect course of action? Most everyone would say yes. Why should it be any different when the autopilot is making the control inputs?

Quoting Sbkom (Reply 16):
Quoting Spitfire (Reply 14):
Unexpected removal of the Flight Director Command Bars during approach

What is this? Can someone explain please?

The Flight Director is something that shows up on the PFD that provides instructions to the pilot on how to fly the airplane in both pitch and roll. The information it uses is based off of the Mode Control Panel in the 737 (as with most airliners), which is on the glareshield and is commonly referred to as the autopilot controls. But what the MCP actually does is control the flight director - the crew can either choose to follow the flight director themselves (hand-flying) or to have the autopilot do it by engaging the autopilot. The only button that actually controls the autopilot direcly is the autopilot on/off button.

For example, if the airplane was in level flight on a heading of 180 with the autopilot off, and the crew turned the heading knob to read 090, the flight director would show that a turn to the left was necessary. The pilots could choose to turn that way or not - the autopilot, were it engaged, would start the turn automatically.

During the flare (which is what the errant RA indication caused), something happens to the flight director. I can't remember what, but will do some investigation later today.

Hope that helps.

Quoting Jpiddink (Reply 17):
If the (one) altimeter suddenly dropped from 1950 ft to -8 feet, and as a result the throttles went to idle and the airplane began to loose altitude, somehow either
- the failing altimeter should have sunk below ground level (I assume the Polderbaan is at about -20 ft?), resulting in the automated systems recognizing they weren't functioning properly anymore (flying below the earth surface), or

This is a radar altimeter we're talking about, which measures height above terrain. It doesn't matter what elevation that terrain is at on the radar altimeter means you're on the ground, whether you're below sea level (like at AMS) or a mile above it (like at DEN). It is competely independent from the barometric altimeter, which is set to read altitude above sea level. At most airports, the barometric altimeter will read higher than the radar altimeter, since most airports are at least a few feet above sea level. AMS is one of the few airports where the barometric altimeter will read lower than the radar altimeter.

The radar altimeter is in the fuselage (generally the nose), but is calibrated so that a reading of 0 means that the main wheels are on the ground. But the nose is much higher than the main wheels at touchdown, and when the nose comes down to the runway, it sees the ground coming closer than it was at the 0 mark. That's how you get readings such as -8. It's perfectly normal when the airplane is on the ground.

Quoting Jpiddink (Reply 17):
- the failing altimeter still showed -8 ft and didn't drop with throttles in idle. Shouldn't that also be an indication that the automated systems recognize as unexplainable behavior, hence shut down A/T and/or A/P?

That's a good question. There is also the issue of why it couldn't detect a change from 1950 to -8 in a matter of moments as being odd. Both of those are things that Boeing is going to have to look into.

-Mir

[Edited 2009-03-05 01:59:20]


7 billion, one nation, imagination...it's a beautiful day
User currently offlineExSR From Germany, joined Sep 2008, 83 posts, RR: 0
Reply 21, posted (5 years 9 months 3 weeks 22 hours ago) and read 19000 times:



Quoting DogBreath (Reply 19):
As a B737-800 pilot, I find this accident appalling. Who was flying the airplane?? What was the 'safety pilot' doing? Another case study for failures in CRM (Crew Resource Management).

Regardless of the incorrect data from the Captains (LH) Radio Altimeter, this should not have brought down this aircraft. Proper scan technique and adherence to the basics of aviation should have detected the reducing airspeed.
I can't for the life of me understand why the safety pilot didn't monitor the FO? If he had been doing his job, this would've most likley ended differently.

No aviation accident happens due to a single event (in this case the incorrect RA), it happens due to a number of problems. The James Reason model uses the swiss cheese theory that if all the holes of defence are lined up then an accident will happen.

Notable to me is that this aircraft had been written up on previous flights due to the same fault. Any good Captain will read back through the Tech Log and highlight to himself and the FO any recurring faults and to note this as a potential 'threat' for the flight. Whether this was done or not we may never know, but if it was, surely they it have been known during the approach. Reading the report is says that they discussed the incorrect RA and dismissed it as not a problem (can't remember the exact words from the report).

What was the Safety Pilot doing? His job is purely and simply to monitor the actions of the FO. From his position on the jump seat he has the best seat in the house to monitor what going on. Like the Flight Engineer of old, he can easily scan the entire flight deck to provide valuable added support to both pilots. Why didn't he see the dangerous situation developing?

There seems to be so much debate on this topic about the incorrect data from the RA, about 100 seconds of Idle thrust, blah, blah, blah. A shame that the aircraft was allowed to develop into a deep stall in the first place. It's not like they had a split second to make a judgement call and act on it they had plenty of time to save this aircraft. It should not have happened if the pilots had been doing their job.

Very very well said!


User currently offlineLifelinerOne From Netherlands, joined Nov 2003, 1938 posts, RR: 8
Reply 22, posted (5 years 9 months 3 weeks 22 hours ago) and read 18974 times:



Quoting Sbkom (Reply 15):
Could you give me some links? I couldn't find such news this morning.

Well, only in Dutch I'm afraid:

http://www.nu.nl/algemeen/1927177/ie...eeft-schuld-behalve-nederland.html

For example headline in Turkish paper Hürriyet was "Everyone is to blame, except The Dutch".

Good example of the sentiment.

Cheers!  wave 



Only Those Who Sleep Don't Make Mistakes
User currently offlineMir From United States of America, joined Jan 2004, 21864 posts, RR: 55
Reply 23, posted (5 years 9 months 3 weeks 22 hours ago) and read 18967 times:



Quoting DogBreath (Reply 19):

Since you're an NG pilot, maybe you could answer this question:

We know that a -8 RA will command idle on the engines. Will it also command flare and rollout modes for pitch and roll, respectively?

I'm trying to figure out what cues the crew would have had to the problem other than the reducing airspeed.

-Mir



7 billion, one nation, imagination...it's a beautiful day
User currently offlineRheinwaldner From Switzerland, joined Jan 2008, 2289 posts, RR: 5
Reply 24, posted (5 years 9 months 3 weeks 22 hours ago) and read 18888 times:



Quoting JoeCanuck (Reply 4):
So, if the report is correct, the crew did have a warning. Whether or not they associated the failure of the radio altimeter with the throttles going into flare mode is another matter. If they didn't, why didn't they?

The problem is the unnoticed impact on other systems. How close is the relationship between speed (by A/T) and RA? There is NO link which immediately lets ring a bell.

It is quite evident that because the plane was fast and high it descended already in idle mode when the RA went off. That means there was NO visible or aural indication that the automation just has switched the state. Not one instrument to show any difference, no movement of the levers. How good is that for awareness?

The A/T completely under the hood has switched into that disastrous retard mode. The plane continued for the largest part of those 100 seconds to do what it was supposed to do: Bleed off speed. All the checking and monitoring of the world would have signaled:  checkmark , checkmark , checkmark .

In accordance to the dropping speed the PF probably commanded more an more flaps. Pilots are evry aware about the speed in that phase because the actual speed dicates the required flap setting.

Only in the end for a very short time the anomaly started to show up.

I stay with my assessment: That completely unanticipated automatic retard-mode-selection mainly killed those people.

Is it right to assume that the plane could have been saved if the manual powersetting would not have been idled again by that doomed A/T?

If yes the poor airmanship factor looses even more weight.

Quoting Sudden (Reply 7):
without the crew noticing that they must have been way below glideslope due to abnormal sink rate as a result of lost forward motion.

The autopilot holds the glide path until the very last moment. It does not scarifice altitude but speed. In fact the A/P does not care about speed. The A/T does. The A/P just holds the flight path.

IMO this is an example of automation contributing the major part to a crash. I don't say that automation did not prevent a much larger number of potential crashes. But here the systems are to blame.
In additon I must mention that this accident probably says not a lot about automation in general because this A/T seems to be designed so poorly.

Each of the following simple policies could have been implemented to prevent that outcome:
- Use the DME to the runway as additional criteria when to select the retard mode the earliest. Expect that this DME signal could be "wrong" too.
- Use all available RA's to select the retard mode.
- Use three RA's to get a validated-by-two signal
- If only two RA's are used validate them against each other AND DO SOMETHING in cases of discrepancy. Something at least would be the refusal to go ever in retard mode. A warning and a A/T disconnect should be expected. Though even that could cause confussion because the source of that event still is not obvious.
- Any manual thrust lever movement in retard-mode immediately ends the retard mode and disconnects the A/T. Unbelievable that automation reclaims the thrust authority after a pilot evidently overruled a retard-mode.

A retard mode in general must be designed extremely cautious because very close to the ground it aims to reduce power. That is the opposite of all safety systems that are in action for all the other flight phases.

This may be the reason why Airbus requires this step fully to be done manually. The systems which prevent a stall or a controlled flight into terrain would otherwise have to be designed to manage a bullet-proof automated transition into an off-mode. By nature building such a transition is dangerous. With the manual retard mode, (which in other words tells to the machine "landing is imminent") the burden to detect that condition always correctly is taken away.

As a summary:
The system design in my eyes is inexcusable. When did the manufacturer ever release so soon after the accident a risk note?


25 Post contains links Doktor71 : Here is one link: http://www.cnnturk.com/2009/turkiye/...oylemek.yanlis/516426.0/index.html "Blaming pilots is wrong" If needed I might translate tha
26 Sbkom : This is a big statement, and open for debates. Don't get me wrong, I agree that the pilots failed to recognize the problem, but it is hard to draw a
27 Mir : There should be indications on the FMA of the mode change. Well, except for getting 40 knots below Vref. If I'm getting close to Vref and my throttle
28 Mir : I agree. But as a programmer, you should understand that while we would like automation to be 100% reliable, things are going to happen from time to
29 Sbkom : Yes, thank you. I missed that one. I usually check other papers (Hurriyet etc) for statements like that, I wouldn't expect that from CNN. In fact, Mr
30 DogBreath : All I can say is that on my Company's aircraft if on single channel Autopilot - no. Dual channel Autoland - yes. I'm not aware if Boeing would have m
31 Post contains images Sbkom : In this case it is less than 50% reliable It failed (or forgot) to pick the correctly functioning sensor. If there are 2 RA sensors, it should use bo
32 Sbkom : Very interesting! Please keep investigating.
33 Doktor71 : I agree, that the whole investigation must come to an end. But this statement from TALPA will not get much attention from others since their first re
34 LTBEWR : With the possibility brought up in the preliminiary investigation of a faulty most likely USA made altitude measuring device and a USA made aircraft,
35 Mir : What I'm thinking is that if the autothrottle went to idle (which we know it did), the FLARE mode would have engaged, which would just have held a pi
36 Max777geek : This is just plain wrong at design level. Everything crucial in the airplane systems is crosschecked with its redundant counterpart, why the hell the
37 IDISA : So far there are some features that are unclear to me... 1) How is possible for such a "hi-tech" aircraft such as the 737-800 that there is so poor re
38 Jpiddink : And this also is exactly why the research by the Safety Board can by law not be used in a lawsuit. Everyone consulted or involved must be able to spe
39 Rheinwaldner : Of course. Therefore I wrote "for the largest part". Contrary to other claims the time to do that was not 100 seconds but maybe only as less as 20. T
40 TristarSteve : Correct, both AP must be engaged for an autoland. No. The AT went to retard because it was at -8 ft from the L RA. The AP continued the approach beca
41 Qualitydr : Md80, I agree. And two other crews who saw a radalt failure in the same machine had no problems. The data I'm missing here: Were the other 2 failures
42 EMA747 : What the monitoring FO was doing for me is a BIG question and one which we know virtually nothing about at this stage. Maybe he was filling in some t
43 Rheinwaldner : You write a lot of reasonable things about the design. However the step alone must not always mean that the RA failed. But a discrepancy between two
44 DRAIGONAIR : Fully agree with you...its the first thing you learn when you are training to be pilot. I disagree. Yes automation did contribute, but that wasnt the
45 DRAIGONAIR : One more question about the emergency service..Im getting a lot of talk here that they were slow? Does anyone have any hard facts available how long i
46 DogBreath : Just to clarify, when flying in single channel on an ILS, there is no 'flare' mode. That is no A/T reduction to Idle and no pitch change to flare the
47 Md80fanatic : I'm not so sure we can simply blame the pilots after less than a week of investigating. Too many interlinked interdependent systems, plus the possibil
48 NCB : Well worded and I think that it is the accurate analysis. All instrument scans would have shown that nothing was wrong until in the last 10-20 second
49 Gonzalo : Hi Everybody. I think we can made some "conclusions" at this point. First, there is a potential problem with the RA/AT automation in all the 737's, ac
50 Max777geek : Aiming only at one of the RA systems to a 143knots speeding object at 1950 feet ? Ok, ok, let's say your plan works. How did his beam direct only to
51 Woody71 : When the rad alt changed to the incorrect altitude the plane was still in approach config, flaps at 15 and gear retracted. When the aircraft changes t
52 EMA747 : I thought from some post above that they came in fast and high and the thrust was at idle as they were slowing down? (I'm not trying to say you are i
53 Bahadir : i hope some of the turkish media doesn't pick up on this it will be a nice conspiracy theory to feed the people.. my countrymen love conspiracy theor
54 Mir : I could buy that, but DogBreath's posts indicate that the AT shouldn't go to retard due to a -8 RA on a single channel approach. Please let's not sta
55 DRAIGONAIR : That doesnt matter, they still have their normal altimeters..Assuming correct QNH was set, with both RA's inop they would still know their altitude.
56 Rheinwaldner : Not when there is still speed in excess. In this phase usualy the speed reduction from KIAS 220 .. 200 .. 180 down to IASref is commanded. After that
57 Gonzalo : I hope the same. The newspapers here in Chile are publishing the Turkish point of view, with the minister talking about the Dutch investigation as "w
58 Mir : It's fair to guess that at least one of those happened in VMC conditions. But the thing is that the weather conditions aren't really relevant - the A
59 FlyinTLow : FIrst of all I have to say that I do not fly a Boeing, I only fly Airbus, and so I can only come from that side and share my thoughts. What most peopl
60 PHBUF : Hello all, After having browsed the Airliners.net forum for many years now, I have finally decided to join. I have read most of the previous 8 pages r
61 Mcdu : No, the automation did not kill these people. Failure to manage and operate the automation killed these people. The pilots still have to pilot the ai
62 FlyinTLow : I don't think it stands in direct violation to it. For one thing, and I know you will hate me for saying this, since this is just legal stuff and has
63 PHBUF : Only on 2 out of 8, yes, but that still would make me believe the crew would have been notified about this, either by the THY technicians or by lookin
64 TristarSteve : I have not seen any statement that there was a REPORTED fault with the RA. From the FDR they have seen that it has failed a couple of times in previo
65 Rheinwaldner : How do you know? The accident has nothing to do with trusting the wrong altitude. On a glideslope the basic altitude information comes from the glide
66 PHBUF : Okay, that certainly would explain a lot. But still, I would like to think that the crews of those two flights would have reported the RA malfunction
67 FlyLKU : Agreed. It appears from available information that the aircraft was in flyable condition when it crashed (unlike the Hudson Glider). The radar altime
68 Mir : True. However, the RA had malfunctioned before, so if something had been written up, it would have been in the MEL before the flight, and they would
69 PHBUF : Exactly. It would be very interesting to see what the THY 737 MEL says about inoperative radio altimeter(s). Much of the blame the pilots will probab
70 Mir : Agreed. It has to do with the pilots letting the autopilot take them to someplace they weren't expecting. That's the million dollar question. And, ev
71 Woody71 : Does anyone know who the manufacturer of the auto throttle system is?
72 Mcdu : Planes don't just fall out of the sky on approach as your post suggest. There are clues the airspeed is low and getting lower. There are target pitch
73 Hiflyer : While the Dutch preliminary report stated that this radalt was the third for the aircraft in the past 20-30hrs per the FDR nowhere did it say (that I
74 Gonzalo : How do i know they weren't capable of maintain altitude and speed ? Because they crashed. Sad but true, they crashed their airplane. I don't want to
75 DRAIGONAIR : On the speed tape the airspeed would have also shown a trend vector pointing down, showing a decrease in speed!
76 RedFlyer : Great post, Thilo. I have a question regarding the above. When a pilot finds himself/herself in a situation where they don't understand what the airp
77 PHBUF : It has not been stated yet, but if you look at reply 60 you'll see what the KLM 738/739 MEL says about it. Basically it says one RA may be inoperativ
78 Md80fanatic : It's called playing "devil's advocate". That's what you have to do before casting blame on the dead guys. All other avenues need be exhausted complet
79 PHBUF : With all due respect, I wouldn't use the word 'saved.' What saved those 92% of the passengers was the absence of a fire breaking out, which I can't i
80 Md80fanatic : Flying a fully loaded 738 at less than 90 knots without dropping a wing is what then, luck? A massive disaster was averted by someone up front, that i
81 PHBUF : Well I surely hope your right and that the FDR/CVR will prove you right. But I think at 90 knots, in a stall situation, there is not much the crew can
82 SSTsomeday : I am following these posts with interest, especially those of airline pilots who shed very informative light on how things run in the cockpit - and t
83 Hiflyer : While the TALPA style defense is admirable the Dutch press conference and statements referred to the CVR when it stated "provisional data indicates t
84 FlyinTLow : Absolutly agree with you there! But you are getting this all wrong! The RAs failed, and from what the preliminary findings of the investigation say,
85 PHBUF : I agree with you that apparently they did not fully understand the consequences; they did not realize that an inoperative RA would influence the autot
86 Mcdu : It appears from the information at hand the recovery might have occurred had they had more altitude. The 737 in flaps full config will be controllabl
87 Md80fanatic : It depends on whether the hull landed wings level or not. I don't think this happens by luck. Wings level is a result of control, a controlled landin
88 Theredbaron : We really need to hear the CVR to assess: 1) What in Gods name was the 3d pilot Doing there? Munching pretzels, reading the NYT? Somebody had to be mo
89 Gonzalo : Huh ? How is it possible that the pilots ignore the MEL recommendations ???? And since when the KLM and CO MEL's do not connect the two systems ?? It
90 Sbkom : Maybe that's what the Safety Pilot was doing all that time! One clue about the faulty RA being left unreported is the words of the infamous Mr. Sen o
91 Boeing747_600 : No, but IDLE would be annunciated as the MCP Thrust Mode SInce they came in fast and high, this may not necessarily have been the case I take it that
92 Sbkom : I wouldn't be so sure about this with absence of crucial CVR transcriptions. Let's not be so conclusive. We still don't know what the pilots thought,
93 NCB : Exactly. I find it insulting that people blame pilot error for this accident this easily. Do you think that the crew fell asleep for 100 seconds? I m
94 Boeing747_600 : While I personally believe that pilot error was the dominant cause of this crash, I think the logic behind using only the LH RA as input to the A/P an
95 Sbkom : Welcome to the thread. This has been covered in the previous threads. Please read the threads. It is being discussed for a week now. This has been ta
96 Boeing747_600 : No responsible pilot would ignore a speed bleed of over 1 KIAS/sec on approach for a 15 second period. How about the Thrust Mode going to IDLE on the
97 Post contains links TK787 : http://www.airkule.com/default.asp?page=yazar&yazar=unalb A very detailed written article from airkule.com, a Turkish aviation site. ( In Turkish only
98 CityhopperNL : Well yes, this makes sense to me, I didn't look at it this way before but you are certainly right. The first people however who should take this into
99 JoeCanuck : Flying a fully loaded 737 at 90 kts might be termed negligence. I think it's probable that the pilots didn't understand the correlation between the f
100 TristarSteve : Define failure. When avionics boxes fail they send a discrete to the monitoring equipment. In this case a large yellow RA on the PFD. But this box di
101 Michi : Concerning the MEL confusion: You check the MEL before the flight, in case there is a MEL relevant item in the tech log book. In case something fails
102 Osiris30 : Your statement, and indeed your entire post is false. There was multiple indications, including a gear warning at 1950ft (far too high), as well as i
103 Spitfire : During all the approach (and certainly when established on the ILS) BOTH pilots have to monitor their instruments. When out of the clouds the PF will
104 DRAIGONAIR : What i have learnt is that the PF is looking at the instruments and PNF is checking. Approaching minimums the PNF flying looks up to look for the run
105 TristarSteve : IDLE is an AT mode triggered by the faulty L RA. FLARE is an AP function. The R AP was engaged getting signals from the normal R RA.
106 Woody71 : The r/a didn't fail, it read an incorrect altitude but was working properly as far as the system knew. Now I agree that a severe jump in altitude sho
107 PHBUF : I'll give you that, the wings-level-impact might have helped save lifes and might be due to control inputs by the crew I am not being conclusive at a
108 Sbkom : Indeed a very unbiased article. It summarizes our findings here. I wish TALPA learns from that.
109 JoeCanuck : So regardless of automation, typically someone is monitoring the flight instruments at all times, correct? I am loathe to blame pilots for accidents
110 DogBreath : The FMA will annunciate FLARE but only dual channel autoland approaches. On a normal single channel approach the AP must be disengaged prior to at le
111 PHBUF : while not exactly what you're looking for, maybe interesting nonetheless..from the 737-600/-700/-800/-900 Flight Crew Operations Manual (from the Boe
112 CityhopperNL : Would AT increase throttle as a reaction on gear extraction even if airspeed was still too high? I don't think so personally, knowing that AT is not
113 Derik737 : *********************************************************************** 737 AUTOPILOT AND AUTOTHROTTLE OPERATION *************************************
114 Sbkom : Inability to perform the expected function. Please do not think only hardware failure, where the electronics fail and stop sending information. There
115 Md80fanatic : 8 feet below ground level? --------------------------------------------------------- To group: The bad RA reading did not lead to pilot confusion of
116 PHBUF : maybe not of the actual altitude, but it's very well possible it created confusion onboard the flight deck regarding the autothrottle. And it's also
117 JoeCanuck : 0 feet is when the plane is on the landing gear. -8 might indicate collapsed gear. The sensor is in the fuselage.
118 CityhopperNL : Could it be that retard mode overrides any other speed selection, so also the alpha floor protection? Apart from this, this crash WILL have consequen
119 Md80fanatic : True Joe, but in the air (higher than the gear struts) it should never read a negative.
120 B2707SST : This was addressed above: TOGA is not inhibited at any point until reverse thrust is selected on the roll-out. This makes sense, becuase you may need
121 JoeCanuck : ...which, to me, it should have been further investigated instead of ignored.
122 PHBUF : it still would have been an incident, but perhaps not an accident..
123 Post contains images Md80fanatic : Missed that, thanks.   That is how it purportedly supposed to work, but then again, retard shouldn't occur at 2000 feet. There really is no telling
124 NCB : Change that to: "No pilot" instead of "No responsible pilot". Also, you can only decide to ignore information once you have received and analysed the
125 Derik737 : A/P Go-Around Arm (Only in DUAL CHANNEL) A/P go-around arms when these conditions are true: The airplane is below 2000 feet radio altitude Both autop
126 JoeCanuck : I assumed you use power to adjust to the glideslope. If that is the case, as soon as they saw they were below the glideslope, they should have increa
127 MD80fanatic : Not if you consider the spool up time from idle, there will be a multi-second delay until useful thrust and several more until full thrust is availabl
128 Michi : Some technical aspects: How does a computer (a flight warning computer, that generates all the cautions) know, that it gets bad information? Someone h
129 SEPilot : The problem with any automatic control is that the user is not the one who thought out the logic of the control. The operator seldom has a full under
130 Michi : I should add, that with 2 RAs and different outputs, there must have been some kind of warning, to indicate the pilots the difference. If I remember i
131 Mir : If I made the same errors they apparently made, I'd expect to be criticized harshly. I'm not interested in going easy on people who can't defend them
132 SSTsomeday : Thank you for the detailed reprimand. I guess the points I mentioned are now closed to discussion. Good thing you were here to put me in my place.
133 CityhopperNL : Actually you've made your points about the 15 seconds response time more than once, and I agree with you that this is a pretty short period of time.
134 Derik737 : I don't believe we know if the DFCS went into Alpha Floor or not. Also, I'm not sure where the DFCS will stop pitching up to maintain G/S. Mir, your
135 MD80fanatic : But they weren't handflying, since a dual channel approach was needed to have automatic retard activate (if I have that right from memory on this thr
136 JoeCanuck : In the wee Cessna I fly, on approach I use power to maintain glide angle and pitch to maintain speed. How are descent rate and angel controlled in air
137 TristarSteve : Sorry you have lost me now. Please explain how Alpha floor works on a B738 Autopilot.
138 MD80fanatic : Interesting the data points we have show airspeed finally settling down below 90 knots....the exact value is not as important as the consistent readi
139 Osiris30 : My understanding is that the AP was disengaged and only the AT was engaged. If I have missed this I apologize. The automatic controls on these aircra
140 Mir : The report says that the throttles went to retard, which, as has been mentioned, implies an autoland. It also says that the B autopilot was engaged,
141 XT6Wagon : Its also not like its a huge secret where the autothrottle is putting the power setting... IT MOVES THE THROTTLES to match its setting. So if you let
142 Derik737 : See my post above (#113)
143 Spitfire : To al those who are not professional pilots here, just a simple remark concerning instruments monitoring and time: During an instrument approach (ILS,
144 Mcdu : Often when hand flying with the A/T engaged you will have inputs to the throttles that you do not want and it requires overriding the servos. So to a
145 Ciaran : * aircraft airpseed slowed 40kt below reference speed; * Aicraft descended through glideslope; * Captain was "coaching" First Officer in conducting th
146 Post contains images Vikkyvik : No, that's not logical. Commercial airplanes are generally designed to be able to perform a go-around up to thrust reverser activation. If the engine
147 ULMFlyer : Mir, I'm just the messenger here, so please bear that in mind. But the consensus over at PPRuNe now is that the A/T will go into "retard" mode, even
148 Ciaran : Sorry about the spelling guys (thats what you get when you rely on automation) but the question was in earnest
149 Md80fanatic : Yes, I remember that now from the Congonhas discussion. Thanks for the head's up.
150 JoeCanuck : Sorry...not my quote.
151 Md80fanatic : That was my comment. This is what happens when a poster quotes from someone else's quote rather than going back to when the person originally said it
152 THEENGINEER : Is there an official accident report thats available online? Thanks.
153 Derik737 : Sure they do, just maybe not in your FCOM. See bottom of post 125 above. That verbiage is from the Boeing 737 AMM.
154 THEENGINEER : Is there an official report from the Dutch yet and I'm not talking about the press conference I am trying to find out If something came out today? Th
155 Sbkom : Michi, Even with only one RA sensor the computer should be able to tell. It is called the logical approach. Imagine this: how in the real world the a
156 RedFlyer : Why didn't they FLY the airplane when the stick-shaker went off? The stick shaker is there to warn the pilots of an impending stall. Which means it's
157 THEENGINEER : No one should jump into conclusions here. I know the Turkish side is being very defensive about their pilots here and I don't see the necessity of thi
158 ULMFlyer : Sorry I missed the bottom of that post. FCOM? What's that? The stuff I fly doesn't even have an ADF Seriously, though, thanks for your technical post
159 TrojanAE : I really don't understand on why you guys are so hellbent on defending the pilots. You make them out to be heroes or martyrs that gave their lives to
160 TrojanAE : Let me put it this way. A bus driver has cruise control on and the speed sensor fails, sending the bus way over the speed limit. For 15 seconds the bu
161 Luniew : The crew should have ben paying attention to the airspeed, just because the auto throttle is managing the speed is it does not mean they didnt have to
162 Mir : I'm not a "shoot the messenger" sort of guy. I finally got my hands on a 737 manual (albeit a -300 instead of an -800, but IIRC the autoflight system
163 XT6Wagon : Only to make it comparible, its a minute and a 1/2 that things are going wrong, the driver knows it, and does nothing much at all
164 OV735 : Based on the evidence at hand, I'm having a really hard time thinking of the crew as heroes of some kind. On the most critical phase of flight, they d
165 Md80fanatic : It's really quite simple..... A few years back the checkers at my local grocer were directing people to the new automated self-checkouts. They though
166 Vikkyvik : The vast majority of what you stated, whether true or not, should have absolutely no bearing on an aircraft accident investigation. If you are defend
167 Mir : This accident should do exactly the opposite. This accident was preventable. Were the airplane fully automated, it wouldn't have been. -Mir
168 TrojanAE : While I completely understand what you are pointing at, probably more than 50% of all aircraft accidents have been blamed on pilot error, and there i
169 Theredbaron : Completely agree ! Specially if previous crews with 4 eyes instead of 6 monitoring eyes did not crash. best regards TRB
170 Osiris30 : So to summarize, because you don't want pilots replaced you'll go out of your way to blame anything but the pilots at the risk of more lives down the
171 Md80fanatic : I have numerous -other- reasons (none emotional), Vikkyvik.....and if you had been keeping up you'd know that. I do not mean to say that I would defen
172 Whoopwhoop : it takes dual channel (A&B) on a 737 for an autoland sequence. 2 out of 3 on a 75-76. At least the throttles move on a boeing when the A/T's command a
173 Vikkyvik : That's fine. However, in response to the question, you specifically responded with reply 165, so I was responding to that. None of your other reasons
174 Osiris30 : MD80Fanatic: How many times do you need someone to post the facts again to you? It is *not* conjecture that the flight crew allowed the throttle to r
175 Sbkom : I am sorry but I don't agree. The money they will save from the pilot's salaries will be paid to the programmers in multiples. I know that because I
176 Post contains links Khobar : TAM 3054 demonstrates that no system is perfect. In the TAM case, the throttles were set incorrectly and, as a result, subsequent throttle lever move
177 MD80fanatic : I think most are missing my point.....I have not jumped off the deep end into some wallow of emotion. With 1 week since accident, A.net appears to be
178 Osiris30 : Uhhh.. OOP is prehistoric relatively speaking. No one has ever accused MS (including themselves) of inventing OOP. While I agree with the overall poi
179 Sbkom : Dear OV735 and TrojanAE, I don't know what you do for living, but flying an a/c, especially professionally is far more complex than flying WWII simul
180 Osiris30 : 1) Enough data is availble for me to know that there was defintely one fatal decision made by the pilots and potentially another. 2) You will note th
181 MD80fanatic : Ummm, let's don't go there. More programmers than ever, more programs, and you know what.....they all look and operate exactly the same. They were al
182 Sbkom : It is not that black and white. All we are trying is to find the reason for that distraction. Otherwise, a plane crashes, we all know pilots are ther
183 MD80fanatic : Thanks Osirus30, I understand you have enough data to make a conclusion. That's cool. I wish I did. There will be a time, when perhaps enough data is
184 Post contains images Osiris30 : Sbkom: I'm talking about the 6 seconds where the throttles re-retarded. This is not excusable by distraction. Any of the actual pilots on this forum
185 Sbkom : I feel obligated to defend myself since more than one person pointed out this. I said "invented" in quotes, you should sense the irony here, that Mic
186 Osiris30 : Sbkom: I got the irony, but had to point it out as others may not have any idea what you're talking about and as such may not get it. I got a chuckle
187 B2707SST : Although I haven't posted much on this thread series, I've followed it with great interest, probably more than any other A.net thread in years. Mostl
188 MD80fanatic : Osiris, you are assuming no fault in any other system but RA, and I think that is premature at this juncture. If the plane proved to be fully worthy u
189 Osiris30 : Admitedly I am. Based on the information I have available we've had the investigators tell us the plane responded to all pilot inputs correctly. Shou
190 Sbkom : Nice summary, thank you. I guess, everybody agrees that some degree of pilot error is the "resulting cause" of this sad accident. The discussion is t
191 Commander_Rabb : Correct Sbkom! In classic Safety Board conclusion speak: “The pilot in command’s failure to maintain sufficient altitude from terrain and violati
192 B2707SST : Bear in mind that the RAs are only active below about 2,500 feet, so autothrottle deactivation would only apply below this altitude, and visual and a
193 Sbkom : I didn't hear that! It seems like we have a few friends here writing from Turkey, maybe they can shed a light to this?
194 PlanesNTrains : True, but I think a logical reply might be that they are not trying to "defend" anything or anyone. Instead, I think for many of us, posting or lurki
195 NCB : Once speed starts dropping at 2kts per seconds to below VSref on the worst of the power curve, good luck with recovery. TOGA should still be availabl
196 Spitfire : I have checked my B737 300-400-500 Operations Manuals, QRH and Non-normal check listand procedure, NOTHING ABOUT THIS !!!! The only place you find so
197 JoeCanuck : What I have the most trouble with is that with 3 sets of trained eyes in the cockpit, nobody caught when they descended below the glideslope and the
198 PlanesNTrains : Well, there definitely seems to be a divergence. Some (majority?) seem inclined to hold the crew somewhat more accountable than the aircraft for the
199 PHBUF : I think I'll have to disagree on that one..no human, no pilot, no air traffic controller, nobody is 'perfect' and I think it's unfair to expect anyon
200 Spitfire : Sure...and the very LAST one is the pilot in command.
201 JoeCanuck : It may not be fair, but that's what everyone expects from the folks behind the iron door of the cockpit. People expect to get to their destination al
202 PHBUF : Agreed..and I'm sure the investigation will explain as to why even the very last line of defense failed in this case
203 PHBUF : But I think there's a difference between the expectation of the pilots (or anybody else for that matter) being perfect and the expectation to get to
204 DingDong : It's far much more likely that certain TK employees here were mercilessly hounded by certain forum participants under the guise of supposed benevolen
205 JoeCanuck : I'm actually just expecting them to fulfill the most basic requirements of their job; fly the plane. They had to shoot a pretty basic approach with t
206 PHBUF : Ah..now that I can live with I don't think that's so strange, I'm sure Boeing, the Dutch investigation team and many other parties involved in this i
207 Rheinwaldner : The MEL does not help. Such things must be addressable from memory. At the first flight when the RA fails the crew had no possibility to learn from t
208 PHBUF : I disagree..the MEL was invented because it's simply impossible for the crew to know any consequence of any technical deficiency by memory. I also do
209 JoeCanuck : If they were monitoring all of the flight instruments, which were working fine, they would have realized they were below the glide slope and slowing
210 Iakobos : One of these outside influences might be the "spoofing" of the RA by someone on the ground. If someone had a device that emitted the precise frequency
211 Spitfire : OF COURSE !!!! That's the way it should work.
212 XT6Wagon : Yup, its blowing my mind that people can't see that 3 people failed to fly the plane... No excuse possible. This situation didn't need the best pilot
213 Rheinwaldner : Sure in that case it must work like that and the MEL is a great help to handle the systems. But when looking at the safety system as a whole there is
214 EZEIZA : But I recall reading (if I'm mistaken I apologize ) that they were perfectly aware of the RA problem and considered it to be a non issue, so that mea
215 Spitfire : -On the ILS, around 500 feet /ground you have the stick shaker : ONE THING TO DO ======> GO AROUND !!! and then investigate when you are in a holding
216 Starlionblue : Some cultures place "honor" in front of "truth". The cultures that are too far down this path tend to have a hard time dealing with the "modern" worl
217 PHBUF : especially when death is involved..I'm guessing the heroic status of the pilots would have been a lot less if they had survived
218 NCB : I think that this can indeed be considered as a clear mistake on the part of the unexperienced F/O. They did not know that the aircraft was in flare
219 CityhopperNL : Yes, you are right, the moment that the FO put his hands on the throttle at such a low airspeed with the throttles in idle, he should have known what
220 PHBUF : aren't those two statements contradicting eachother?
221 CityhopperNL : No it does not. So as the pilots should have been aware of this, they can be blamed for allowing throttle levers to roll back. I actually think that
222 CityhopperNL : Sorry, I should have formulated it more carefully, should have said: In the best case for the pilots, they are only responsible for the drop in airsp
223 FlyinTLow : absolutly agree with you there. Whenever you do not get along with the automation, take control. It's the easiest thing to do, back to the basics and
224 PHBUF : Indeed, hopefully the investigation will also determine if those six seconds would have mattered or not.
225 OV735 : Thanks for your reply. My post was primarily meant as a reply to the the posts like below, where the crew is praised for saving lives. Perhaps not ex
226 PHBUF : hey, that was my line! j/k, agree with you completely. Hopefully the investigation will show 'forgiving circumstances' for the crew. THY and Turkey i
227 Oldtimer : Why do you keep on about dual channel and they must have been in Autoland, read the previous threads and the reports from Boeing and the Dutch Author
228 WhoopWhoop : ^^^^ becasue only IN DUAL CHANNEL...which is an autoland requirement on the 737, the autothrottles recieve input from the RA. Single channel you can f
229 SEPilot : After reading most of the posts on all of these threads (I have not offered many posts because my knowledge of the particular issues here is limited)
230 Post contains links HB-IWC : Please continue the discussion of this topic in the following thread: TK B738 Crash At Amsterdam - Part 9 Part 8 is now archived.
Top Of Page
Forum Index

This topic is archived and can not be replied to any more.

Printer friendly format

Similar topics:More similar topics...
TK B738 Crash At Amsterdam - Part 5 posted Wed Feb 25 2009 18:35:07 by WILCO737
TK B738 Crash At Amsterdam - Part 4 posted Wed Feb 25 2009 09:52:20 by HB-IWC
TK B738 Crash At Amsterdam - Part 3 posted Wed Feb 25 2009 06:10:53 by HB-IWC
TK B738 Crash At Amsterdam - Part 2 posted Wed Feb 25 2009 03:44:01 by HB-IWC
TK B738 Crash At Amsterdam posted Wed Feb 25 2009 01:49:48 by Luvaulter
BA B772 Crash Lands At LHR - Part 6 posted Sat Jan 19 2008 12:26:32 by B747forever
BA B772 Crash Lands At LHR - Part 5 posted Fri Jan 18 2008 11:54:38 by ANCFlyer
BA B772 Crash Lands At LHR - Part 4 posted Fri Jan 18 2008 01:13:30 by 777ER
Small Plane Crash At Santa Monica Airport posted Wed Jan 28 2009 17:57:45 by N83SF
CO Flight 1404 Off Runway At DEN - Part 2 posted Sun Dec 21 2008 20:58:50 by HB-IWC