Takeoff (climb)

The left engine propeller's uncommanded travel to the feathered position during takeoff for reasons that could not be determined due to impact damage. Contributing to the accident was the flight crew's failure to establish a coordinated climb once the left engine was shut down and the left propeller was in the feathered position.

The accident was the result of many contributing factors which culminated in a stall-induced loss of control. During the initial climb after takeoff, an intermittent discontinuity in engine number 2’s auto feather unit (AFU) may have caused the automatic take off power control system (ATPCS) sequence which resulted in the uncommanded autofeather of engine number 2 propellers. Following the uncommanded autofeather of engine number 2 propellers, the flight crew did not perform the documented abnormal and emergency procedures to identify the failure and implement the required corrective actions. This led the pilot flying (PF) to retard power of the operative engine number 1 and shut down it ultimately. The loss of thrust during the initial climb and inappropriate flight control inputs by the PF generated a series of stall warnings, including activation of the stick shaker and pusher. After the engine number 1 was shut down, the loss of power from both engines was not detected and corrected by the crew in time to restart engine number 1. The crew did not respond to the stall warnings in a timely and effective manner. The aircraft stalled and continued descent during the attempted engine restart. The remaining altitude and time to impact were not enough to successfully restart the engine and recover the aircraft.
The following findings related to probable causes were noted:
An intermittent signal discontinuity between the auto feather unit (AFU) number 2 and the torque sensor may have caused the automatic take off power control system (ATPCS):
- Not being armed steadily during takeoff roll,
- Being activated during initial climb which resulted in a complete ATPCS sequence including the engine number 2 autofeathering.
The available evidence indicated the intermittent discontinuity between torque sensor and auto feather unit (AFU) number 2 was probably caused by the compromised soldering joints inside the AFU number 2.
The flight crew did not reject the take off when the automatic take off power control system ARM pushbutton did not light during the initial stages of the take off roll.
TransAsia did not have a clear documented company policy with associated instructions, procedures, and notices to crew for ATR72-600 operations communicating the requirement to reject the take off if the automatic take off power control system did not arm.
Following the uncommanded autofeather of engine number 2, the flight crew failed to perform the documented failure identification procedure before executing any actions. That resulted in pilot flying’s confusion regarding the identification and nature of the actual propulsion system malfunction and he reduced power on the operative engine number 1.
The flight crew’s non-compliance with TransAsia Airways ATR72-600 standard operating procedures - Abnormal and Emergency Procedures for an engine flame out at take off resulted in the pilot flying reducing power on and then shutting down the wrong engine.
The loss of engine power during the initial climb and inappropriate flight control inputs by the pilot flying generated a series of stall warnings, including activation of the stick pusher. The crew did not respond to the stall warnings in a timely and effective manner.
The loss of power from both engines was not detected and corrected by the crew in time to restart an engine. The aircraft stalled during the attempted restart at an altitude from which the aircraft could not recover from loss of control.
Flight crew coordination, communication, and threat and error management (TEM) were less than effective, and compromised the safety of the flight. Both operating crew members failed to obtain relevant data from each other regarding the status of both engines at different points in the occurrence sequence. The pilot flying did not appropriately respond to or integrate input from the pilot monitoring.
The engine manufacturer attempted to control intermittent continuity failures of the auto feather unit (AFU) by introducing a recommended inspection service bulletin at 12,000 flight hours to address aging issues. The two AFU failures at 1,624 flight hours and 1,206 flight hours show that causes of intermittent continuity failures of the AFU were not only related to aging but also to other previously undiscovered issues and that the inspection service bulletin implemented by the engine manufacturer to address this issue before the occurrence was not sufficiently effective. The engine manufacturer has issued a modification addressing the specific finding of this investigation. This new modification is currently implemented in all new production engines, and another service bulletin is available for retrofit.
Pilot flying’s decision to disconnect the autopilot shortly after the first master warning increased the pilot flying’s subsequent workload and reduced his capacity to assess and cope with the emergency situation.
The omission of the required pre-take off briefing meant that the crew were not as mentally prepared as they could have been for the propulsion system malfunction they encountered after takeoff.

The accident was due to a loss of control that brought the aircraft into unusual attitudes, which the pilot was unable to restore and which led to his fall. The insufficient skills of the pilot when faced with a high performance aircraft, whose systems are complex, contributed to the occurrence of the accident.

The Aircraft Accident Investigation and Inquiry Board determined that the probable cause of this accident was:
- Lack of recurrent training of the flight crew:
Routine flights do not prepare a pilot for unusual situations, whether they are unexpected crosswinds or systems/engine anomalies. Pilots should receive regular recurrent training to include abnormal and emergency procedures.
- The existing runway edge light design:
The PIC tried to recover the aircraft back to the runway but apparently the aircraft left main landing gears already hit or bumped the concrete base of runway edge lights. The design of runway strips or shoulder must be free from fixed objects other than frangible visual aids provided for the guidance of aircraft and must not be constructed
with sharp edges; and where the lights will not normally come into contact with aircraft wheels, such as threshold lights, runway end lights and runway edge lights;
- Human Factors:
Due to deteriorating adverse weather conditions and due to the delay of their initial request for take-off clearance plus the sudden change of flight plan affected the Captain’s ability to perform a take-off procedure as recommended in the aircraft flight manual and instead delegated flight control duties to the F/O resulting in the loss of coordination between the light crew.

The accident occurred as result of the aircraft departing the side of the runway after the commander rejected takeoff after having been unable to use the elevator because of the yoke's locked position. The roll beyond the edge of the runway was likely caused by the flight engineer while attempting to operate the handle to release the flight controls lock while the aircraft was already accelerating for takeoff. The accident was thus caused by this combination of factors:
- violation of requirements by FCOM to ascertain the flight controls were free and usable before engine start,
- failure by the crew to execute the checklists to check elevator, rudder and ailerons were free to move before takeoff,
- flight crew receives insufficient practice in real flight to maintain skills acquired during simulator training in the management of the aircraft and its systems resulting in negative impact during emergency situations.

During the attempted takeoff, the rudder was central from 40 kt and remained so until approximately 65 kt. Between approximately 52 and 65 kt, the aircraft turned right slightly before it turned left sharply at approximately 65 kt. Given that the rudder was central, this change of direction might have been caused by one, or a combination of the following factors:
a. Differential braking
b. Asymmetric thrust
c. A change in wind speed and direction
d. A nose wheel steering input
Data from the FDR showed that thrust was applied symmetrically throughout the takeoff run, and the manufacturer did not consider that the data for longitudinal acceleration and indicated airspeed supported the use of differential braking.

The loss of power to the right engine for reasons that could not be determined during postaccident examination and teardown and the pilot’s failure to properly configure the
airplane for single-engine flight.

This was a private flight which could not depart in conditions of less than 400 m RVR. RVR cannot be measured at the threshold end of Runway 03 but the prevailing visibility was reported as being more than 10 km. The crew reported that there was moisture on the windscreen from the mist and they could see a “glow” around lights which were visible to them. They were also aware while taxiing that there was some patchy ground fog on the airfield. The ATC controller transmitted that visibility had not been measured in the fog patches but there seemed to be ‘very low, very thin fog from the zero three threshold to approximately half way down the runway’. With hindsight, this piece of information is significant but, at the time, the crew did not consider the fog to be widespread or thick; operating under FAR Part 91 in the United States, they were used to making their own judgments as to whether the visibility was suitable for a takeoff. However, after the aircraft came to a halt following its abortive takeoff attempt, the controller could only see the top of the fuselage and tail above the layer of fog. It is likely, therefore, that the visibility was worse than the crew appreciated at the time N103CD taxied from Holding Point J1. The route from J1 to the runway The information on the aerodrome chart used by the crew, and the source of information in the UK AIP, suggested that the aircraft would be required to taxi in a straight line from J1 to the runway and then make a right turn onto the runway heading. In fact, in order to taxi from J1 onto the runway, an aircraft must: taxi in a straight line; follow a curve to the right onto runway heading but still displaced to the right of the runway itself; turn left towards the runway; and then turn right again onto runway heading. The UK AIP states that there is no centreline lighting on Runway 03, and that the pavement width at the beginning of the runway is twice the normal runway width. It recognizes the potential for confusion and urges crews to ensure that they have lined up correctly. This information was not available to the crew on their aerodrome charts and both crew members believed that the runway had centreline lighting. Further, the light from those left-side runway edge lights covered in fog would have been scattered, making it harder for the crew to perceive them as a distinct line of lights. The situation is likely to have been made worse by the bright lights reflecting off the top of the fog layer, making the underlying runway lights even harder to see, or swamping them completely as shown in Figure 5. The CCTV images in Figure 5 show that peripheral lighting can interact with low fog layers to reduce the visibility of underlying aerodrome lighting. Current standards associated with apron lighting only address the minimum light levels required to make the areas safe and there are no standards relating to light spilling into other areas.
Human and environmental factors Five of the factors identified by the ATSB as being present in misaligned takeoffs were present in this accident:
1. It was dark.
2. It was potentially a confusing taxiway environment given that the aerodrome chart did not reflect the actual layout of the taxiways. Pilots had previously reported having difficulty when vacating the runway near the Runway 03 threshold because of a lack of taxiway lighting.
3. There was an additional paved area (the ORP) near the runway.
4. There was no runway centreline lighting and the runway edge lights before the displaced threshold were recessed.
5. There was reduced visibility.
It appeared that the information available to the crew caused them to develop an incorrect expectation of their route to the runway. Both crew members believed that the runway had centreline lighting and, when the first right turn almost lined the aircraft up with some lights, their incorrect expectation was reinforced and they believed that the aircraft was lined up correctly. Cues to the contrary, such as runway edge lights on the other side of the runway, or the fact that the first three lights ahead of the aircraft were red (indicating that they were edge lights before the displaced threshold), did not appear to have been strong enough to make the crew realize that they had lost situational awareness. Figure 8 indicates that the apparent intensity of the white left-side runway edge lights was significantly less than that of the right-side lights, when viewed from the position where the aircraft lined up. This, along with other visual issues relating to contrast and the fog, is a plausible explanation as to why they were not noticed by the crew. The aircraft began its takeoff roll from a location beyond the first red runway edge light and approximately 46 m short of the next light, as shown in Figure 1. Aircraft structure only obscures approximately the first 13 m of pavement ahead of pilots within a Gulfstream III aircraft and therefore these lights would not have been obscured by the aircraft. However, it is likely that the recessed nature of the red edge lights before the displaced threshold made them less compelling than the elevated white edge lights beyond, which would explain why their significance – that they could only have been runway edge lights – was not appreciated by the flight crew.

A failure of the right engine magneto distributor drive gears, which resulted in a total loss of engine power during takeoff. Contributing to the accident was the operator's failure to inspect and maintain the magnetos in accordance with the engine manufacturer's specifications.

The pilot's failure to maintain lateral control of the airplane after a reduction in left engine power and his application of inappropriate rudder input. Contributing to the accident was the pilot's failure to follow the emergency procedures for an engine failure during takeoff. Also contributing to the accident was the left engine power reduction for reasons that could not be determined because a postaccident examination did not reveal any anomalies that would have precluded normal operation and thermal damage precluded a complete examination.