The Boeing bulletin on the Turkish Airlines crash

Thanks to the commenter Dicks Airbus who posted the bulletin text in the comments on my earlier post.

Just to clarify, here it is again:

FROM: THE BOEING COMPANYTO: MOM [MESSAGE NUMBER:MOM-MOM-09-0063-01B] 04-Mar-2009 05:29:01 AM USPACIFIC TIMEMulti Operator MessageThis message is sent to all737-100,-200,-300,-400,-500,-600,-700,-800,-900,-BBJ customers and torespective Boeing Field Service bases, Regional Directors, the AirTransport Association, International Air Transport Association, andAirline Resident Representatives.SERVICE REQUEST ID: 1-1228079803ACCOUNT: Boeing Correspondence (MOM)DUE DATE: 10-Mar-2009PRODUCT TYPE: AirplanePRODUCT LINE: 737PRODUCT: 737-100,-200,-300,-400,-500,-600,-700,-800,-900,-BBJATA: 3400-00SUBJECT: 737-800 TC-JGE Accident at Schiphol Airport, Amsterdam – 25February 2009REFERENCES:/A/ 1-1222489391 Dated 25 February 2009Reference /A/ provides Boeing’s previous fleet communication on thesubject event. The US NTSB, FAA, Boeing, the Turkish DGCA, theoperator, the UK AAIB, and the French BEA continue to actively supportthe Dutch Safety Board’s (DSB) investigation of this accident.The DSB has released a statement on the progress of the investigationand has approved the release of the following information.While the complex investigation is just beginning, certain facts haveemerged from work completed thus far:- To date, no evidence has been found of bird strike, engine orairframe icing, wake turbulence or windshear.- There was adequate fuel on board the airplane during the entireflight.- Both engines responded normally to throttle inputs during the entireflight.- The airplane responded normally to flight control inputs throughoutthe flight.The Digital Flight Data Recorder (DFDR) data indicates that the crewwas using autopilot B and the autothrottle for an ILS (InstrumentLanding System) approach to runway 18R at Amsterdam Schiphol airport.During the approach, the right Low Range Radio Altimeter (LRRA) wasproviding accurate data and the left LRRA was providing an erroneousreading of -7 to -8 feet. When descending through approximately 2000feet the autothrottle, which uses the left radio altimeter data,transitioned to landing flare mode and retarded the throttles to theidle stop. The throttles remained at the idle stop for approximately100 seconds during which time the airspeed decreased to approximately40 knots below the selected approach speed.The two LRRA systems provide height above ground readings to severalaircraft systems including the instrument displays, autothrottle,autopilots and configuration/ground proximity warning. If one LRRAprovides erroneous altitude readings, typical flight deck effects,which require flight crew intervention whether or not accompanied by anLRRA fault flag, include:- Large differences between displayed radio altitudes, including radioaltitude 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 duringapproach- Unexpected Configuration Warnings during approach, go-around andinitial climb after takeoff- Premature FMA (Flight Mode Annunciation) indicating autothrottleRETARD mode during approach phase with the airplane above 27 feet AGL.There will also be corresponding throttle movement towards the idlestop. Additionally, the FMA will continue to indicate RETARD after thethrottles have reached the idle stopBoeing Recommended Action- Boeing recommends operators inform flight crews of the aboveinvestigation details and the DSB interim report when it is released.In addition, crews should be reminded to carefully monitor primaryflight instruments (airspeed, attitude etc.) and the FMA for autoflightmodes. More information can be found in the Boeing 737 Flight CrewTraining Manual and Flight Crew Operations Manual.Operators who experience any of the flight deck effects described aboveshould consult the troubleshooting instructions contained in the 737Airplane Maintenance Manual. Further, 737-NG operators may wish toreview 737NG-FTD-34-09001 which provides information specific for the737-NG installation. Initial investigations suggest that a similarsequence of events and flight deck indications are theoreticallypossible on the 737-100/-200/-300/-400/-500. Consequently the aboverecommendations also apply to earlier 737 models.

, , ,

21 Responses to The Boeing bulletin on the Turkish Airlines crash

  1. John S March 4, 2009 at 3:44 pm #

    “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. ”

    No crosscheck between the left and right LRRA data to the autothrottle? No autothrottle inhibit for an LRRA fail?

    What is the point in having two LRRAs if you have no safeguards in place to inhibit a faulty LRRA?

  2. Harry P March 4, 2009 at 4:24 pm #

    Agree with previous post completely.

    Given the extremely high importance of these devices (Radio Altimeters) and the high confidence self-checks that are built into these, I am very interested to learn what the root cause of the inaccuracy in the left LRRA turns out to be.

    Does anyone know which manufacturer designed this Rad Alt?

  3. Phil March 4, 2009 at 4:31 pm #

    From the operation description in the report, there is no automatic comparator that looks at both rad/alts…requires crew monitoring and intervention.

  4. Phil March 4, 2009 at 6:26 pm #

    To add to my last comment….3 Radar altimeters would be required to have a true comparator and voting system….the defective unit would be voted out because it did not agree with the other 2. With only 2 units, the comparator only knows there is a miscompare, but not which unit is defective, so it requires crew monitoring and intervention….the crew has to delete the defective unit and select the correct one, or manually fly the approach.

  5. david foster March 4, 2009 at 9:32 pm #

    “Inability to engage both autopilots in dual channel APP (Approach) mode”…this sounds like there *is* a mode in which both A/Ps operate simultaneously…with some form of comparator between their actions?…with different RA inputs?

  6. Michael March 4, 2009 at 11:17 pm #

    In the report is written that autopilot B was in command. Even if the leftside LRRA senses a -8 ft altitude, how could it be possible that the thrustlevers were positioned in the retard mode. As far as I know the retard mode is an autoland A/T mode only available when both autopilots are in command.

  7. Dean K March 5, 2009 at 4:15 am #

    Jesus Christ! Throttle remained at idle for 100 seconds! This is far beyond an eternity, what were the flight crew thinking during this time! I say this as a fellow pilot, seriously, I’m not sure what was going on in there.


    I stopped engaging autopilot on approach, hand flying from now on starting at 10k feet. I’m not going to let a machine do my job anymore. Two aircraft have gone down this year due to autopilot over use.

  8. Michael March 5, 2009 at 9:06 am #

    Landing FLARE and RETARD are dual autopilot modes at a B737. How is it possible to have those modes engaged if only autopilot B was engaded…?

  9. Ed March 5, 2009 at 10:19 am #

    Please listen
    Start (sub)
    if leftLRRA not equ to rightLRRA;
    disengage AT

    How hard can it be

  10. Alemaobaiano March 5, 2009 at 12:55 pm #

    The 100 second comment came from Boeing, not the investigation, and is disingenuous at best and a slur on the deceased crew at worst. It seems to be a point picked up by many as a stick to beat the crew with, wondering what they were doing for 100 seconds.

    At what point did the 100 seconds start? At some point earlier in the approach is the answer, before the radalt jumped to -8′, which would be entirely consistent with a slightly high approach, indeed a long period of idle was considered as a possibility in the BA038 incident. Several members on other websites have estimated that the time from radalt change to impact as 18-20 seconds, too long certainly, but not an eternity of 100 seconds.

    Did the crew fail in their duties? Probably, but until the CVR data is released we can’t be certain.

    Were they let down by their equipment? Certainly. How can the A/T system rely on a single unverified data source?

    Just to clarify one point here, the A/P and A/T systems are independent, with the A/T system taking it’s altitude reference from the LH radalt ONLY.

  11. Gurer March 5, 2009 at 9:27 pm #

    (…right (LRRA) was providing accurate data and the left LRRA WAS!!! providing an erroneous reading of -7 to -8 feet.. )
    I don’t understand; when left LRRA begin malfunction? and after that an eternity of 100 secs… at idle stop! plenty of time to disengage the autopilot.
    why the autothrottle system actuator responded so quickly? Is it a programming glitch? from 2000ft to -8ft in nano secs.. and the autothrottle goes to idle stop immediately, no matter what other feeds -if any- were?
    Anyway, the pilots sees the throttle movement, and how come they stay put for 1.5 minute? with alarming ground proximty warnings at all.
    If there is a training session going on, why the airplane was on ILS? I can understand the human decision error in excess technology (Inability to engage both autopilots in dual channel APP (Approach) mode) but it seems that the combination of many errors was in place at that unfortunate moment.
    I simply can not understand 100 secs of inactivity.

  12. Dirk van der Laan March 7, 2009 at 11:10 am #

    100 seconds non-reaction? NO WAY. 20 seconds at most. The first 80 seconds the aircraft was decreasing from intitial approach speed (180 or more) to 140 on idle power which is perfectly normal when descending on the ILS (lowering the gear and extending the flaps). It was only when reaching 750 ft (20 seconds before stall) that the A/T should have reacted.
    Why didn’t the crew notice? There are many many possible reasons why the F/O in those 20 seconds might have been distracted from the normal monitoring cycle (which every pilot learns and learns and learns at hours and hours of training). Fe the captain might have been ‘talking him through the before landing checklist’ as was suggested at this site.
    We don’t know what happened in the cockpit so stop these stupid comments!

  13. Jan March 7, 2009 at 11:57 pm #

    If the LRRA began to malfunction at approximately 2000 feet then the crew would not have noticed the thrustlevers moving to idle, since they were already reducing to idle thrust in order to decelerate to final approach speed. Decelerated ILS approaches are the norm these days! Rumour has it the crew were also slightly high on final approach, so they needed to bleed off both kinetic and potential energy, thus extending the time that idle thrust would be needed.

    The timeframe between the moment that the airspeed dropped below the FAS and the moment of impact was much less than 100 seconds. Judging this crew for not noticing idle thrust in a flight phase where idle thrust would be _expected_ for a prolonged time does not strike me as very knowledgeable. Every pilot posesses great airmanship – in hindsight.

  14. Kenneth M. Murphy March 25, 2009 at 4:40 pm #

    Concerning the “RETARD” mode in the 737-800, neither autopilot needs to be engaged. With autothrottle only, the throttles will automatically retard during landing. They can be manually adjusted if you don’t want that to happen.

  15. Mike March 25, 2009 at 6:27 pm #

    We could probably agree that distractions should be avoided during the approach phase. That being said, this airplane has a wonderful airspeed indicator … it even changes color. When you decelerate to the minimum maneuver speed it changes from white to yellow. At that point it is sitting on top of a red and black tape that indicates the area below stick shaker speed. The airspeed tape is below the FMA (Flight Mode Annunciator) autothrottle indicator which would have being annunciating RETARD in green letters. It will be interesting to learn if the investigation turns up other failures that could have distracted the crew from the task at hand.

  16. August 31, 2013 at 1:05 am #

    Truly helpful as well as superb structure of articles or blog posts, now that’s fantastic.

  17. Oversteel Fashion September 28, 2013 at 2:16 am #

    There are men’s bracelets with the Greek Key design, other etched and complexly cut designs such as
    skulls, scorpions, even the Christian cross to name a few.
    Without a doubt, the jewelries from those huge brands are fantastic.
    Also, it does require that the piercer to have special skills.

  18. Ouida October 1, 2013 at 4:20 pm #

    You need to make sure that yoou adhere to all
    of the restrictions that should be followed after youu get your post lip piercing such
    as avoiding any facial exercises andd drinking.
    Sellf expression, grace, poise, composure, self confidence and
    spiritual poweres find a sudden surge. Due to this, onee of the best Valentine’s
    Day gifts that you can give to your love one is stainless steel jewelry.

  19. Www.Ziddu.Com October 7, 2013 at 6:21 pm #

    Body piercing has become a school of commerce today.
    Without a doubt, the jewelries from those hhuge brands are fantastic.
    People haave risked their lives to journey to our country.

  20. subway Gift Card October 18, 2013 at 3:43 am #

    Find a movie and pop some popcorn and snuggle down together watching
    a comedy that will get everyone laughing. Cover and let rise
    in a warm place for 45 minutes or until doubled in size.
    Potato soup, bean soup, ham and pea soup, carrot soup and others would
    be great soups to include that are fairly simple to make.

  21. nordstrom November 5, 2013 at 4:58 am #

    The brans themselves have their owwn sales, which normally run laterr than the Independent
    Shops sales. In the USA: New York, Los Angeles and San Francisco all have
    fashion weeks. If you are going foor tthe Roman or Grecian look,
    you can go with a toga style or Grecian gown type dress.

Leave a Reply