Turkish airlines accident at Schiphol

The Turkish Airlines Boeing 737-800 that crashed on final approach to Amsterdam Schiphol on 25 February is the second aircraft in a little more than a year to land short of a runway on approach to a major airport.

The other event involved a British Airways Boeing 777. It landed about 350m short of runway 27L at London Heathrow in January 2008.

The two events may not turn out to have much in common when we finally know the causes, but it’s worth looking at the similarities.

Both hulls were badly damaged by the impact with the ground, but there were no fatalities in the case of the British Airways flight and in the Turkish Airlines 737 only a few fatalities among the one hundred and thirty-four people on board.

And, of course, neither aircraft caught fire.

The Turkish 737′s hull looks as if the vertical speed at impact was greater than for the 777, which may account for the fatalities that occurred, but it was clearly survivable by most of those on board.

In neither case was the weather on approach – in itself – a challenge for the aircrew, given that they were approaching runways that have first class navigation aids and high intensity approach and runway lighting.

The interim report on the British Airways event says that the reason the crew had to put the aircraft down short of the runway is that, when they called for power, having put the aircraft into landing configuration, the engines didn’t respond, so the crew had to steepen the approach to maintain flying speed.

The cause of the engines’ failure to respond is believed to have been a temporary restriction to the fuel supply caused by ice crystals in the fuel. If this is confirmed in the final report, it will have been a unique accident cause for modern jet aviation.

As for the Turkish Airlines 737-800, it had been in service for six years, and the -800 is one of the so-called Next-Generation 737s – the latest variant of the long running 737 Series.

So here we have a latest-generation aircraft landing at one of the world’s great hub airports in misty weather – but with more than adequate visibility – and somehow it didn’t make it to the runway. What happened?

Frankly, nobody knows at this stage. Did this crew suffer a power failure of some kind? We don’t know, but certainly the unique circumstances surrounding the long-haul British Airways flight were unlikely to have been repeated in this short-haul case.

Could it have been a birdstrike, like the one that the US Airways Airbus A320 suffered before it ditched in the Hudson recently? Maybe, but we have no evidence right now.

As modern airlines go, Turkish Airlines does not have a good accident record, having experienced two fatal crashes in the last ten years.

They lost a Boeing 737-400 in 1999, and an Avro RJ100 in 2003. The Avro RJ event also occurred on final approach, but in poor visibility.

7 Responses to Turkish airlines accident at Schiphol

  1. Esteban 25 February, 2009 at 7:44 pm #

    “Only a few fatalities among the one hundred and thirty-four people on board.”

    So, 9 dead people are not that bad??

  2. David Learmount 25 February, 2009 at 7:46 pm #

    Those are your words. I am glad that 125 people survived, and I’m sure you are too.

  3. Oliie361 26 February, 2009 at 6:03 pm #

    BA 038 crash and Turkish Airline crash at Schiphol
    I have always thought through my whole flying career that if I needed thrust on short final and the engines would not respond (spool up) for any reason, I would have two instinctive options:
    1) Crash land the “bird” wherever possible
    2) Reach the runway somehow.
    Crash landing suggests to keep the landing gear down on uneven terrain or up on smooth terrain.
    Not having the privilege of choosing where to put down the “bird”, because of the location of most airports, it is obvious that in that situation reaching the airfield is the best choice.
    I always kept in mind this possible event and trained myself mentally to retract the gear if the above circumstances would materialize.
    Fortunately in my 30 years flying career this sort of event never materialized.
    My hunch is that if both flights met the above circumstances on short final, maybe they should have thought and execute the retraction of the landing gear to gain those few hundred yards to reach the runway performing a belly landing.
    I remember that one the important memory items of an emergency descent on DC9 B737, B747, was to put the gear down if airspeed was under M 0.80/0.82 or 320 kts IAS.
    The landing gear effect was as effective as, if not more than, the spoilers to augment the drag allowing a much steeper descent.

  4. David Nicholas 27 February, 2009 at 11:09 am #

    I recall from my dispatch days that a 737 is very easy to trim. You just put the bags 50/50 in the two belly holds and seat the passengers evenly forward and aft of the Centre of Gravity. Then the only trim change that the crew will experience will be as a result of fuel usage (this is why Ryanair blocks off the same number of seat rows at the front and read of the cabin on less-than-full flights).
    However, complications arise with mixed-class passenger loads, and cargo. You can end up with the aircraft close to the forward or aft trim limit (and allowance for the movement of the CG must be made when trimming and loading the aircraft, so that it remains in trim throughout the flight). It is possible for an aircraft that has been improperly trimmed and/or loaded to go out of trim during flight.
    Pure speculation, but the loss of control while low and slow, the increasing ange of attack observed by witnesses and passengers, and the extremely tail-low impact suggest that the aircraft was out of trim for its speed and configuration.
    This is explained comprehensively by Boeing in this extract from a publication on the avoidance of jet upsets:
    In a situation where the airplane pitch attitude is unintentionally more than 25 degrees nose high and increasing, the kinetic energy (airspeed) is decreasing rapidly. According to the energy management principles, the energy is actually being stored as potential energy. As airspeed decreases, the pilot’s ability to maneuver the airplane also decreases. If the stabilizer trim setting is nose up, as for slow-speed flight, it partially reduces the nose-down authority of the elevator. Further complicating this situation, as the airspeed decreases, the pilot could intuitively make a large thrust increase. This will cause an additional pitch up for underwing-mounted engines. At full thrust settings and very low airspeeds, the elevator — working in opposition to the stabilizer — will have limited control to reduce the pitch attitude.

    In this situation the pilot should trade the potential energy of altitude for airspeed, and would have to maneuver the airplane’s flight path back toward the horizon. This is accomplished by the input of up to full nose-down elevator and the use of some nose-down stabilizer trim. These actions should provide sufficient elevator control power to produce a nose-down pitch rate. It may be difficult to know how much stabilizer trim to use, and care must be taken to avoid using too much trim. Pilots should not fly the airplane using stabilizer trim, and should stop trimming nose down when they feel the g force on the airplane lessen or the required elevator force lessen. This use of stabilizer trim may correct an out-of-trim airplane and solve a less-critical problem before the pilot must apply further recovery measures.
    …all well and good unless you have run out of altitude. The normal method of avoiding a stall involves pushing the stick forward until the speed recovers, but you can only do this if you have the height to spare.

    Rather than the B777 event at Heathrow, I see a commonality with both the A320 near Perpignan and the Colgan Dash 8 – in both cases control was lost at an altitude too low for recovery.


  5. Oliie361 27 February, 2009 at 6:24 pm #

    The book says, if you crashland keep the gear down as it absorbs kinetic energy and helps reduce speed faster on ground. I agree with that.
    Many colleagues of mine are mesmerized by the old adage that landing gear doors operation, when trying to pull the gear up in dire short final conditions, compounds the drag problem.
    I am not sure that is a problem today with RAT (Ram Air Turbines that supply electrics and hydraulics in case of all engines flameout), engine driven hydraulic pumps that have fantastic performances with engines at idle, very different to the old piston driven ones, make the landing gear retraction and the gear doors operation time quite short.
    Has anyone tried it in a class D simulator?

  6. bill randolph 27 February, 2009 at 9:57 pm #

    One thing I would like to say ICE CRYSTALS o come on it might fool the general public but not me these aircraft are state of the art FADEC controlled engines in both cases plus you would step decend into London airspace not straight from cruise altitude.Fuel starvation is likley min fuels etc etc. Heathrow and Schiphol have long holding patterns.To carry fuel costs money, times are hard.Gone are the days where you land with 10 ton still on board we all know that.
    A double engine flame out on these engines will not happen unless there is on common thing missing,fuel.

  7. David Nicholas 2 March, 2009 at 1:54 pm #

    Ollie, who said the gear wasn’t down?They were only about half a mile from the threshold so how significant is that?
    The 737 has unfaired main undercarriage doors so no extra drag there while the wheels retract/extend. At the speed at which the aircraft was observed to be flying when it stalled, the RAT would probably not have even extended, let alone run.
    At high angles of attack, such as this, accelerating an engine from approach speed to TO/GA would be slow, and tisk compressor stall.
    For whatever reason: Too low – too slow, nowhere to go……