Areas of investigation of TransAsia GE235 accident

A TransAsia ATR 72 crashed yesterday on take off from downtown airport.

Investigators will look at the following:

  • Because the reported Mayday call indicated engine flameout, there will be initial focus on the engines and with it, anything that contributes to flameouts, such as fuel contamination, bird or other foreign object ingestion, and mechanical issues;
  • They will look at why the airplane didn’t climb out under power of one engine, as airplanes are designed to do;
  • Did only one engine flame out or did both;
  • What caused the airplane to drop off dramatically on one wing to a perpendicular position as it crossed the highway–did the plane fall off on one wing due to flight conditions or did the pilot, believing he was going to lose the plane, bank sharply in a desperate attempt to avoid crashing into the city;
  • Pilot actions in handling the flameout and subsequent actions in flying the airplane;
  • Did the pilots deliberately head for the river to avoid crashing into buildings;
    The mechanical history of the airplane; and
  • The history of the pilots.

Recovery of the flight data recorder and cockpit voice recorder should provide information quickly to answer many of the questions above but the full investigation typically takes a year or more.

34 Comments on “Areas of investigation of TransAsia GE235 accident

  1. Tragic accident. We’ll have wait, it seems he tried to ditch but stalled the wing that didn’t have the prop induced lift, while trying to clear the bridge..

    Did the taxi people survive? Talking about bad luck..

    • It appears that the plane was coming in nose up to ditch in the river but the pilot had to bank to miss the lamp posts and signs on the bridge.

      It looks like the wing tip sliced through the middle of the taxi but amazingly the driver and passenger are still alive although suffered some injuries

  2. I believe, flying passengers plane with 2 engines is quite risky and plane manufacturer as well as FAA group have to consider providing 3 engines instead of 2 for all passengers planes, hence in case of any failure for one engine (which represent the loss of 30% of the plane power) the pilot will be able to maneuver and or to stare the plane much more easier than the case with 2 engine plane. Since here the plan will loose 50% of its power if one engine stops for one reason or another. I am sure here that not all pilot are able to stare their plane specially during take off.

    • There has never been a case that I am aware of where having a 3rd or 4th engine would have prevented a plane crash.

    • Ditto, and the cost to put a 3rd engine on an aircraft aka the Tri Islander (or a Ford or Foker Trimotor?) hmmm

  3. After the loss of power in a twin engine, there is sufficient reserve to maintain a gentle climb. The plane is designed to take this loss, and pilots are trained for this.

    However there are significant yaw and roll issues that have to be resolved, among other things. This article on loss of an engine in a light twin gives a nice indication of some of the complications that eventuate:

  4. Keep in mind that the ATR is already underpowered relative to most of its contemporaries, so it’s not something you’d want to have an engine failure in.

    Here are some weight-to-power ratios of modern turboprops at MTOW:

    Embraer EMB 120 Brasilia: 7.3 lbs per hp
    Bombardier Dash 8 Q400: 6.6 lbs per hp
    Saab 2000: 6.1 lbs per hp
    ATR 72: 10.2 lbs per hp

    • Very interesting figures. There is also the matter of wing loading at max takeof weight. But it doesn’t make up for the delta in power-to-weight ratios.

      ATR72-600 : 76.6 lb/sq ft
      Q400 : 95 lb/sq ft

      Overall, there seems to be more margin on take-off with the BBD Audi A4 than the ATR VW Jetta.

      • It is the other way around. The Q400 is “overpowered” because it is build to operate from short runways, longer ranges at speeds close to jets. It makes the Q400 much more expensive to buy and operate too.

        ATR72 engines (any twin) are dimensioned for continuous climb on one engine for set conditions (temp, height etc.). If this aircraft wasn’t capable of doing so, there probably was another problem apart from single engine operation.

        Someone noticed flaps were not out. I’m not sure what procedures would require under these conditions.

        • Pilots practice flying on 1 engine quite a bit. One possibility is that the pilot became confused on which engine was out and pulled back on the throttle on the wrong engine. We should know soon when they look at the flight recorders.

          This is the second major loss of life incident involving Transasia ATR-72s in the past year. This article looks at the ATR’s safety record:

          ATR 72s have been involved in 10 accidents since 1994 that resulted in more than 350 deaths. The smaller ATR 42 has been involved in eight accidents since 1987 that killed 172 people for a total of 522. By comparison the Dash 8/Q400 has been involved in 6 major accidents since 1988 resulting in 162 fatalities. Maybe it is safer to be “over powered”?

          • You should look at the number in service, where and with whom before drawing conclusions about the airplane’s safety. More likely regional safety standards and airline standards are more relevant.

          • From 1981 until the end of 2014, ATR had delivered a total of 1,190 aircraft: 754 ATR 72s and 436 ATR 42s. Since 1984 Bombardier has delivered 1142 Dash 8s/Q400s. There is a list of ATR accidents here: and Dash 8 accidents at the link above. Very similar mix of operators. All in all the number in service, and where and with whom all look very similar. I do not see why ATR’s accident rate and fatalities are three times higher than Bombardier.

          • Nope.

            Dash8 is not only Q400, but the older, smaller Q100/-200/-300 too.

            471 Q400s were delivered.

          • If you look at only the ATR-72/Q400 accident rates, the difference is even more shocking. ATR 72s have been involved in 10 accidents since 1994 that resulted in more than 350 deaths. The Q400 has only had one accident since 1999 with 53 deaths. Yes there are 754 ATR-72 vrs 471 Q400s but you have to admit that 1000% difference in accident rates has to make you question the safety of ATR-72s.

          • Trooper misses the other side of the equation. The ATR has much deeper market penetration than the Q400. One must also look at where the ATRs operate (ie, what countries and regions of the world) and who the operators are. Safety varies from region-to-region and airline-to-airline. One must also look at the probable causes. How many were pilot error vs a mechanical reason?

          • 1994, American Eagle Flight 4184, an ATR 72–212 crashed due to icing in Roselawn, Indiana killing all 68 people on board.
            2002, TransAsia Airways (TNA) cargo flight 791, an ATR 72–200, crashed due to icing. Both crew members were killed.
            2012, UTair Flight 120, a ATR 72–201 crashed soon after takeoff from Roshchino International Airport in western Siberia. 33 of the 43 passengers and crew on board were killed due to icing.
            The Q400 does not have any problems with icing.
            Even crashes due to pilot errors can be partly due to human factors in the design of the cockpit. All in all it appears you are safer in a Bombardier Q400 vrs an ATR-400.

          • It is absolutely myopic to talk about aircraft safety based on just the number of accidents. You’re assuming that the aircraft was at fault for all of these which you know very well that this isn’t true as there are so many other factors too.

            Same goes for the number of deaths. Two different aircraft can have fatal accidents but if one flight carriers 10 pax and the other 50, it doesn’t make one safer over the other.

            I concur with leeham that regional safety standards is an issue, kind of like driving safety IMO – a safe car can still have different accident rates worldwide.

  5. So, perhaps it was my Mk I, peanut-gallery eyes, but looking at the video I was left wondering how much, if any, flap was set. It didn’t look like there was much if there was any. Should there have been in an engine-out situation?

  6. The other relevant question is why was the pilot talking on the radio when he should have been flying?

    It isn’t like he is telling anyone something they need to know, plane going down in the middle of a city is kind of obvious

    • The ATC recording hasnt been released, so this may be a social media generated ‘news’

    • If the article is correct it could be another BMI flight 92..

      Aircraft systems preventing the pilots crashing aircraft might be a bridge too far, maybe discussions about addition warnings/ advises?

      • My understanding is that the right engine came back to idle on its own whereas the left one was shut down by the pilot; which turned a manageable problem into a catastrophe. So it could be just a pilot error. At this point I can hardly make a connection with BFI92 other than the fact that the good engine was shut down in error after the other one had failed. An engine out take-off is one of the most practised emergency procedure by pilots. It’s almost a routine manoeuvre and most of them could do it with there eyes closed. Now, was the technology a contributing factor in this particular case? To find out we may have to wait until next month when the preliminary report is published.

  7. Pretty amazing, though the cross wire alarm situation would be a consideration (Boeings 737s that showed the wrong engine failure)

    And the spin starts with how nobble the pilot was avoiding building etc to his supreme sacrifice.

    Of course you avoid those things, they kill you first. That assumes he was actually in any control and it simply did not stall out where it did.

    Not to mention the spin of tarnishing of the tourist industry image when in fact its the victims and their families that are the ones really impacted.

    • I like to think that the pilot was aiming for the river, and had the bridge not been in the way he might have been successful. But an ATR in no A320 and the Keelung is not as wide as the Hudson.

      • I guess I put that in useless speculation (granted I indulge in it myself so point a finger at me as well)

        what does it matter what the pilot was intending to do, not do or was not doing anything?

        There seems to be an ideal we put wonderful intentions on pilots parts when they are just like everyone else, reacting to and most of what occurs is pure happenstance to what is going on.

        • As opposed to movies there are so few heroes in real life that we have to create them. A pilot who acts professionally quickly becomes a hero; often despite their own protest: “I only did my job”, or “I did what I was trained to do”.

      • I guess I put that in useless speculation (granted I indulge in it myself so point a finger at me as well)

        what does it matter what the pilot was intending to do, not do or was not doing anything?

        There seems to be an ideal we put wonderful intentions on pilots part when they are just like everyone else, reacting to and most of what occurs is pure happenstance to what is going on.

  8. This is in from the Aviation Herald

    The right engine had an issue or it thought it did. It eventually auto feathered. That means it lost torque so it looks like a engine issue on the right of some kind.

    As soon as the Master Warning went off for the right engine, the left one was put back to idle and shutdown. (fuel off)

    • Here is a key excerpt from The Aviation Herald: <>

      Why was the left engine throttled back after the right one had failed? Something must have confused the pilot(s).

      • Here is the excerpt:

        37 seconds after becoming airborne, a master warning activated related to the failure of the right hand engine, at 10:52:43L the left hand engine was throttled back and at 10:53:00L the crew began to discuss engine #1 had stalled. At 10:53:06L the right hand engine (engine #2) auto-feathered. At 10:53:12L a first stall warning occured and ceased at 10:53:18L. At 10:53:19L the crew discussed that engine #1 had already feathered, the fuel supply had already been cut to the engine and decided to attempt a restart of engine #1.

  9. They shut down the wrong engine and must have realized it.

    No latitude at 1500 feet.

    No indications the wrong engine indicated fault, so one of those weird inexplicable screw ups (unless indeed the right engine was cross wired to the wrong indicator, i.e. correct indication lighting up wrong light, Boeing had that as I recall)

    Same as Spaceship 2, why did he pull the feather lever?

Leave a Reply

Your email address will not be published. Required fields are marked *