Abu Dhabi DH8C at Abu Dhabi on Sep 9th 2012, passenger detects engine overheat

Last Update: September 15, 2015 / 11:18:29 GMT/Zulu time

Bookmark this article

Incident Facts

Date of incident
Sep 9, 2012

Classification
Incident

Airline
Abu Dhabi Aviation

Flight number
AXU-711

Departure
Abu Dhabi, United Arab Emirates

Destination
Das Island, United Arab Emirates

Aircraft Registration
A6-ADB

Aircraft Type
De Havilland Dash 8 (300)

ICAO Type Designator
DH8C

An Abu Dhabi Aviation de Havilland Dash 8-300, registration A6-ADB performing flight AXU-711 from Abu Dhabi to Das Island (United Arab Emirates) with 46 passengers and 3 crew, was preparing for departure when during engine start a passenger seated in the right hand seat row 6 observed fluid coming out of the right hand engine's vent. He commented to his friend in the window seat this was not right but thought this was due to condensation. The fluid however dissipated, the aircraft continued to taxi to the runway. The passenger's attention was now focussed on the engine nacelle, he spotted some hot spot with fumes from the vents at the engine nacelle, but still thought this was due to some residual fluid. The fumes however were continuing and the white colour of the vents changed to brown then black. Shortly prior to the takeoff roll the passenger began waving his hands to attract the attention by the flight attendants, the aircraft however had been cleared for an immediate takeoff from runway 13R and commenced takeoff. After the aircraft had become airborne the flight attendant checked with the passenger and subsequently informed the flight deck about something not right with the right hand engine. The first officer (30, 2150 hours total, 1100 hours on type) left his seat to have a look at the engine while the aircraft climbed through about 2300 feet, at that point the exterior of the panel had already started to blister and bare metal had become visible. The first officer returned to the cockpit informing the captain (47, 9500 hours total, 4500 hours on type) there was a fire inside the right hand engine's nacelle, the crew declared PAN while the aircraft was climbing through 4000 feet towards 5000 feet on a downwind, the crew reported there were signs of fire on the right hand engine. All cockpit indications remained normal, no fire warning occurred, the crew therefore did not discharge the fire bottles. The crew decided to return to runway 13R with both engines operating and shut the right hand engine down after roll out, the cabin was prepared for a possible evacuation. The aircraft landed safely on runway 13R about 10 minutes after departure, the right hand engine was shut down, emergency services in attendance reported no fire, the aircraft taxied to the apron on the #1 engine.

The United Arab Emirates General Civil Aviation Authority (GCAA) released their preliminary report stating, that unknown to crew, emergency services and passengers the #1 engine had suffered a similiar overheat condition, too, because being invisible on the outboard side of the left engine.

Maintenance confirmed that after forcing the access panels enclosing the hot section of the right hand engine open, that hot gases had exited through an open port of the engine. Upon checking the left hand engine maintenance established the same problem had occurred on the left hand engine, too. That port should have been closed with the left hand side ignitor plug. These left hand plugs had been removed on Sep 8th as part of an engine wash, and had been placed inside the engine compartment with the electrical harnesses removed. Following the engine wash the left hand plugs however were not re-installed in both engines.

There were signs of overheating within the compartment including the structure and tubings.

The operator's engineering staff had concluded that introducing a compressor wash would extend the engine life and thus introduced such engine compressor washes (in addition to turbine washes) starting in July 2012, but only on the incident aircraft. Task cards had not been created, the instructions for the compressor wash were distributed verbally. Some of the staff however were away on vacation during the engine wash on Sep 8th, and some staff was not aware of the new requirements.

On Sep 8th 2012 a turbine wash was being conducted, which required to open two specific access panels and remove the left hand igniter plugs. The aircraft also was to undergo some work on the nose wheel steering. After the engine turbine wash had been prepared, the work on the nose wheel steering was completed, the engine access panels were closed in preparation for the engine wash and the aircraft towed outside the hangar, where two other engineers volunteered to do the engine washes. These engineers were aware of the new compressor wash requirements and believed the engine was undergoing a compressor wash. Following the engine wash they therefore continued with a compressor wash, subsequently opened other, smaller access panels following the verbally transmitted steps, then reconfigured the compressor plugs, sense lines and closed the panels, but did not open the larger adjacent access panels and thus did not detect the ignitor plugs laid into there. The turbine water wash did not happen at all. The engineers subsequently verbally informed the supervisor that the wash has been completed, and the supervisor signed the engine cleaning off and released the aircraft to service. The next flight was the incident flight.

On Sep 15th 2015 the GCAA released their final report concluding the probable causes of the incident were:

The Air Accident Investigation Sector determines that the cause, of this incident, which resulted in an 'in-flight turn back', was due to the omission to reinstall the left engine igniters on both of the aircraft’s engines following maintenance work. The maintenance error occurred as result of a number of contributing factors.

The Contributing factors to the event were:

- Unrecorded maintenance work performed on the Aircraft by the Operator’s maintenance personnel.

- Introduction of an engine wash without a maintenance task card.

- Engineer signed off work on the Aircraft without verifying that the work had been performed.

- Mechanics performed unsupervised work.

- Mechanics performed engine motoring without the Operator’s approval.

- Work was performed on the Aircraft without maintenance task card.

- Engine washes were not considered a critical task by the Operator.

- Performing similar tasks on both engines during the same maintenance visit.

- Not carrying out an engine run after the engine washes were performed

- Not performing a system check of the engine ignition system after engine wash normalization.

- Not attaching a telltale streamer to indicate that parts have been removed and are in a concealed area.

- Operator’s quality oversight, as unrecorded work was being performed regularly prior to the Incident.

- Mechanics not signing for work performed, following engine washes.

- The removal, in 2009, of the engine wash card which was requiring a signature by the mechanic, before the engineer signoff.

- The effect of fatigue on the decision making process of the Engineer due to his shift pattern of working an average of 8.5 hours a day for 32 days with 2 staggered days off.

- The Engineer, in addition to supervising the shift work, was required to enter data into the Operator’s electronic system.

- Application of the Operator’s human factors training, as unrecorded work was a practice associated with engine washes.

- The Operator’s SMS implementation, since there were GCAA audit findings between 2009 and 2012.

- Lack of guidance provided by the GCAA, and the Operator, of the effect of shift duty times, and management of the risk associated with fatigue.

The GCAA analysed: "The passenger who witnessed and reported this Incident to the crew, stated that it was difficult to hear the pre-flight safety announcement, due to the noise of the engines and propellers. As there were no cockpit warning indications during engine start, taxi and the initial flight phase to indicate any abnormal situation, the flight crew were unaware that both engines were operating with one igniter not installed on each engine. The gradual change in the paint color, the paint blistering of the right engine nacelle panel, and the potential effect to the continued safe operation of the aircraft was not a situation that the cabin crew member (CCM) had experienced before, nor was it part of the cabin crew CRM training. His interaction with the passenger was not conclusive as the CCM was not technically prepared for such situations. Had the flight crew workload increased due to engine warnings, or any Aircraft system degradation, the critical phase when the CCM entered the cockpit, less than a minute after takeoff, could have impacted on the performance of the flight crew. From the information given to the flight crew by the CCM, verification was done by the co-pilot to enable them to have a clear understanding and aid their decision making process. The flight crew discussed during the interviews that the sight of exhaust gases emitting from the engine vents is not unusual and they have had similar passenger concerns previously. Their initial assumption was based on this knowledge. The decision to return to the departure airport was made after the co-pilot had heeded the persistent request of the concerned passenger which was made known to the flight crew by the CCM."

With respect to the engine washes the GCAA analysed:

On the night prior to the Incident flight, the duty Engineer in charge of the shift had physically prepared both engines for turbine washes, as per the planned work referenced in document EGEN F/W 1150. He removed engine access panels 415AL and 425AL, the igniter leads and the left igniter from each engine. The right igniters were not removed as they were not required to be removed for the turbine wash.

The removed igniters were left inside the recess of the engine compartment without placing them in protective bags.

After rectification of the NLG defect, the Engineer instructed that the Aircraft be towed out of the hangar to the designated wash bay area as the aircraft required an external wash. Before towing, the engine panels 415AL and 425AL that had been opened to allow the removal of the igniters, were both closed.

However, no external tell-tale indicator, such as a streamer, was used to indicate that items had been removed and left inside the compartment.

The Aircraft was towed outside the hangar and the Engineer returned to the control room where he proceeded to update the electronic data system.

Two mechanics, who were assigned to another aircraft, returned to the control room, as they had completed their assigned tasks. The Engineer then requested both of them to perform the engine washes on the Aircraft. This instruction was verbal, without specifying what engine wash was required, and the Engineer did not mention that the engines had already been prepared for the turbine washes.

The mechanics were experienced in performing turbine washes, but they were not authorized by the Operator’s Quality Director to perform unsupervised engine motoring. They were aware that this Aircraft had a special requirement related to compressor washes. However, as mentioned during the interviews, they were not sure whether or not the turbine washes also had to be performed on the Aircraft, and they concluded that only the compressor washes were required.

The mechanics removed panels 413AL for the LH engine and 423AL for the RH engine, in order to access the compressor plugs. The single access plug (figure 5) on the external engine compressor casing was then removed.

The compressor washes do not require panels 415AL and 425AL to be opened.

After the engine washes were completed, both engine compressor plugs and access panels were normalized. At no time during this wash sequence was any aircraft manual referred to, nor was any attempt made to access the turbine section in order to perform the turbine water washes.

Both mechanics returned to the control office around 1 hour before the end of shift and verbally informed the Engineer that the engine washes were completed. The assumption made by the Engineer was that the mechanics were referring to the completion of the turbine washes. He was unaware that the mechanics had actually performed compressor washes on both engines, and had not performed the planned turbine washes.

The Engineer, as he was dealing with other documentation in the control office, was not involved with the engine washes nor did he physically verify that the engine igniters were normalized after the washes before signing off the work in the Aircraft maintenance log book.

The reason for the Engineer’s decision not to inspect the Aircraft is unknown. However, his decision making process may have been influenced by the closeness of the time to the end of the shift, his confidence in the mechanics, the practice of allowing mechanics to perform the engine washes, cumulative fatigue due to the number of days on continuous duty, as well as the fact that on the day before he had worked overtime of about 3.5 hours. His duties as a supervisor in electronically entering data could also have influenced his decision.