A320, London Stansted UK, 2019
A320, London Stansted UK, 2019
On 1 March 2019, an Airbus A320 left engine suffered a contained failure soon after takeoff thrust was set for a night departure from London Stansted but despite the absence of an instruction to cabin crew to begin an evacuation, they did so anyway just before the aircraft was going to be taxied clear of the runway with the Captain only aware when passengers were seen outside the aircraft. The Investigation found that an evacuation had been ordered by the senior member of the cabin crew after she was “overwhelmed” by the situation and believed her team members were “scared”.
Description
On 1 March 2019, a CFM56-powered Airbus A320 (OE-LOA) being operated by Laudamotion, a wholly owned Ryanair subsidiary, and operating a scheduled international passenger flight from London Stansted to Vienna made a very low speed rejected takeoff at night in normal visibility when the left engine failed soon after takeoff thrust had been set. Just as the Captain, having communicated with the cabin crew, was about to taxi the aircraft clear of the runway, passengers were seen outside on the runway and it became apparent that the cabin crew had ordered an emergency evacuation. The serviceable right engine was still running and the 10 minor passenger injuries sustained during the evacuation included people blown over by the jet efflux from the right engine after leaving the aircraft.
The exits used for evacuation in relation to engine intake/efflux hazard areas. [Reproduced from the Official Report]
Investigation
A Field Investigation was carried out by the UK AAIB assisted by relevant recorded data downloaded from the CVR and FDR following their removal from the aircraft. The CVR contained interphone conversations between the flight crew and the cabin crew but not PA use by the cabin crew.
It was noted that the 44 year-old Training Captain in command had a total of 14,128 hours flying experience of which 10,308 hours were on type. Corresponding experience details in respect of the First Officer were not recorded but the flight involved was a line training sector for them. There were five cabin crew on board, one more than usual, but the fifth was not part of the operating crew and was on a familiarisation flight after recently completing her ground training.
What Happened
It was established that, having lined up the aircraft on runway 22, the Captain, who was acting as PF for the flight set rated thrust and commenced the takeoff. About a second after the First Officer had called “thrust set” at a recorded groundspeed of 31 knots, a loud bang occurred and the aircraft began to drift left of the centreline as the Captain announced a rejected takeoff.
The aircraft came to a stop between the centreline and the left runway edge. The Captain then set the parking brake and used the PA system to make the standard announcement “attention crew on station” twice. The First Officer advised ATC that they were stopping on the runway and then worked through the SOP actions for the indicated abnormal conditions which included an ECAM ‘ENG FAIL 1’ annunciation and therefore the shutdown of the engine. There were no signs or indications of a left engine fire.
The RFFS arrived quickly and the Captain contacted them on the UK standard frequency for that purpose, 121.6 MHz to confirm that there were no external signs of engine fire. On receiving confirmation that this was so, the Captain decided to vacate the runway using thrust from the right engine and requested clearance to do so. One minute after the left engine had been shut down, and just as the Captain was about to make a PA to instruct the cabin crew to resume normal operations, a ‘DOOR L FWD CABIN’ caution message appeared on the ECAM. After initially thinking it was a fault, he then reported having seen the Door 1L evacuation slide deployed and saw passengers moving in front of the aircraft.
The Captain called the Senior Cabin Crew Member (SCCM) on the interphone and asked why the evacuation had been initiated and was told that “she believed he had ordered one”, which he (correctly) denied. Following this exchange, the APU was started and the right engine, which was still operating while the evacuation was underway, was shut down. The First Officer then went into the cabin whilst the Captain remained on the flight deck to maintain radio contact with TWR and the RFFS. On entering the cabin, the First Officer found that all the passengers and two of the cabin crew had left the aircraft, the latter in order to assist passengers on the ground, and noted “a lot of baggage near the exits”, all six of which had been opened.
Ten passengers sustained minor injuries during the evacuation, two of whom were taken to hospital by ambulance and the others were treated for minor injuries at the scene. All but the two passengers taken to hospital were eventually transported to the terminal in buses and the aircraft was later towed off the runway to a remote parking position. Several passengers who evacuated via the right side overwing exit (2R) were either nearly blown over, or were blown over several times by the jet efflux from the right engine and some of their belongings blown away. As shown on the illustration above, passengers crossing behind the engine were at risk of exposure to ‘wind’ speeds of 65 mph or more even with the engine running only at idle and it was concluded that in general, there had been “a risk of serious injury due to one of the engines running during the evacuation”.
Separately from the way the emergency evacuation was initiated and conducted, it was observed that yet again, such an evacuation had involved passengers taking their carry-on baggage with them when evacuating. In this case, it had particularly occurred at the overwing exits where the evacuation was not supervised by crew members. It was noted that the CS25.803 requirement that it must be demonstrated that an emergency evacuation of all occupants from an aircraft can be achieved within 90 seconds “does not require any of the test ‘passengers’ to retrieve their own baggage and attempt to leave the aircraft with it”.
Why it Happened - the Evacuation
The cabin crew reported having heard a loud noise soon after the takeoff roll started and had detected the drift to the left before the aircraft came to a stop a few seconds later. All cabin crew except the SCCM reported hearing Captain’s “attention crew on station” PA following which they all stood up at their assigned exits. None of them observed any danger and all awaited further instructions from the Captain. After a slight delay, the SCCM spoke with one of the rear-stationed cabin crew which was followed by the SCCM ordering an emergency evacuation over the PA system.
An evacuation was immediately commenced. All exits except Door 3L, where the slide did not appear to deploy correctly and the exit was blocked, were used for the evacuation. At both the forward and rear exits several passengers brought their hand baggage with them, but it was removed from them before they were permitted onto the slide.
All members of the cabin crew were interviewed the following day and further interviews were subsequently conducted with the SCCM and the two cabin crew who had been assigned to the rear of the cabin.
During her interviews, the SCCM stated that she had not heard the Captain’s “attention crew on station” PA and that the noise of the engine failing had been “very loud" and that this and the movement of the aircraft to the left had “scared her”. She stated that when the other cabin crew all stood up (on hearing the Captain’s PA) she did so too and had been “aware of all the passengers looking at her”. She added that she had “felt under pressure because of this” and had generally been “feeling shocked and overwhelmed”. Her initial attempts to contact the rear-stationed cabin crew by interphone had been unsuccessful and “there was confusion while they attempted to communicate using a combination of the interphone, hand signals and the PA”, although the darkness in the cabin had made the use of hand signals difficult. She explained that she had wanted to obtain information from the rear cabin crew “to help her decide whether an evacuation was needed” and noted that she had gotten the impression that the (male) crew member she had communicated with was himself “scared and shocked” and that his nearby (female) colleague was “completely shocked”. She stated that at some point in this exchange with the rear cabin crew, she had at some point said “evacuate, evacuate, evacuate” over the interphone. The rear crew member who had heard this explained that when she said it, he had not understood why she would order an evacuation over the interphone so he told her to announce it over the PA which she had then done.
The SCCM stated at interview that she was aware of the guidance in the “flight safety manual” about the circumstances in which cabin crew might need to initiate an evacuation on their own initiative but “was not thinking about this at the time”. It was found that she had initially qualified as cabin crew in May 2017 and had then operated as a junior crew until November 2017 when the operator under its previous ownership had ceased trading. After the change of ownership and restart of operations, she had returned to cabin crew duties at the operator in April 2018 and after completing SCCM training was promoted accordingly in May 2018. The practical training for this promotion included scenarios using a Cabin Emergency Evacuation Trainer (CEET) but all of these training scenarios ended with a simulated evacuation. It was also noted that none of the cabin crew training conducted by the operator included any aspect of pilots’ activity when responding to an emergency or the potential effects of startle and surprise on cabin crew performance. However, all cabin crew training met the relevant EASA requirements and had been approved by the Austrian Aviation Regulator.
The OM Part A was found to include a statement that “the designated senior (cabin crew) must have at least one year’s experience as an operating cabin crew member”. Other than this, the operator was found not to have any requirements for the composition of the cabin crew team assigned to each flight duty in terms of experience.
Why it Happened - the Engine Failure
The engine involved was a CFM 56-5B which had been manufactured in 2007 and installed on the subject aircraft since new. The operator had taken delivery of the aircraft in December 2018 and had not performed any maintenance on the engine since then other than oil replenishment. However after the aircraft’s return to the lessor from a previous operator, the return-from-lease inspection had disclosed several left engine defects which required off-wing repair at a specialist engine overhaul facility. During this repair work, Inlet Guide Vanes (IGV) and the Variable Stator Vane (VSV) actuation rings were split to remove the top half of the High Pressure Compressor (HPC) case to enable the replacement of two damaged HPC blades. This work required the disassembly of the connecting links on the IGV and VSV actuation rings. These rings and their connecting links were reassembled after reinstallation of the HPC case and “no defects were noted during the post-repair inspections or the subsequent engine test cell runs” and the engine was subsequently refitted to the aircraft.
It became evident at an early stage in the Investigation that the contained failure of the engine had been a result of the release of several stage 1 HPC blades. These blade failures were attributed to crack progression due to High Cycle Fatigue failure cause by tensile overload and several other HPC blades were also found to be exhibiting fatigue initiation in the blade dovetail. It was then found that the blade fatigue surfaces “were consistent with failure due to a known aerodynamic excitation phenomenon caused by an ‘off-schedule’ Inlet Guide Vane (IGV) / Variable Stator Vane (VSV) condition" which creates stresses in the blade which exceed the design limit. It was considered that the only origin of this “off-schedule condition” which arises from “improper engagement of IGV/ VSV lever arms with their connecting link” was the December 2018 pre-delivery repair work on the engine. It was noted that in comparison with other blade liberation events with the same origin, the engine which had failed in this case had done so with a comparatively low time since the suspect engine repair work which indicated that the extent of the “off schedule” vane condition at release from repair must have been more significant than in most other cases.
It was noted that although the relevant Engine Shop Manual task covering the December 2018 repair contained instructions for a visual inspection to verify the proper installation of the IGV / VSV lever arms, for reasons which could not be determined, mis-assembly was not identified in this way and as reassembly of the whole engine progressed, it would have become increasingly unlikely that the single disconnected IGV lever arm which had been the origin of the excitation problem would be detected since it would have been quickly obscured by external hoses, pipes and brackets. It was also considered very unlikely that the required post-maintenance engine runs would have disclosed the error.
The engine manufacturer observed that it considered incomplete installation of components such as those involved in this case to be ‘‘a common skill-based error’’ and noted that in general, such occurrences can only be reliably prevented by an error-tolerant design from the outset. In this case, it had concluded that “mitigating or eliminating the possibility of an improperly assembled IGV/VSV lever arm would require an engine redesign which it does not consider feasible based on the low rate of occurrence” and noted that “there is currently no means, other than visual inspection, to detect improper lever arm assembly”.
The formally documented Conclusion of the Investigation was as follows:
- The Engine Failure
The left engine experienced a contained failure. All the damage found in the engine was consistent with the release of one or more first stage high-pressure compressor blades as a result of high-cycle fatigue arising from aerodynamic excitation of the blades. A single inlet guide vane lever arm, which had been improperly assembled in the connecting link on the inlet guide vane actuation ring, was identified as the source of the stimulus that resulted in the blade release.
- The Emergency Evacuation
As a result of the engine failure and subsequent rejected takeoff, the Senior Flight Attendant ordered an emergency evacuation that was not necessary in the circumstances. This was probably the result of a combination of factors that heightened her emotional response to the event and affected her decision making. The factors included inexperience as a flight attendant, weaknesses in her training and communication difficulties during the event.
As a result of the flight crew not being consulted before the evacuation was commenced, the right engine remained running for the first few minutes of the evacuation. This led to an increased risk of serious injury to those passengers that evacuated on the right side of the aircraft. Indeed, several passengers sustained minor injuries having been blown over by the exhaust.
During the evacuation several passengers hindered the evacuation by taking their cabin baggage with them. While some were removed by the flight attendants at the supervised exits, this was not possible at the overwing exits.
Safety Action taken in response to this and other recent similar A320 family events was noted as having included the following:
Laudamotion:
- Augmented the team responsible for training with the addition of a deputy manager of flight attendant training.
- Introduced a maximum limit of 30 trainees on initial flight attendant training courses.
- Added practical training in the Cabin Emergency Evacuation Trainer (CEET) to the senior flight attendant course as standard and also incorporated practical training into their annual recurrent training.
- Improved the number of training scenarios used in the CEET to include scenarios that result in a return to normal operations rather than an evacuation.
- Expanded the flight attendant training syllabus to include the performance effects of startle, an improved 30-second review technique (a process where cabin crew mentally rehearse the steps they would have to take during an evacuation before each takeoff and landing) and enhanced communication training.
- Produced a video training aid that will introduce flight attendants to the actions of the flight crew after a rejected takeoff.
- Extended the aeroplane familiarisation phase during initial training with additional familiarisation flights.
- Amended the OM to require that a minimum of two “experienced” cabin crew shall be part of the operating cabin crew complement, which represents 50% of the operating crew members on their A320 aircraft which are operated with four cabin crew.
CFM International evaluated the use of a High Pressure Performance analysis tool which makes it possible to determine if the effects of a mis-assembled lever arm could be identified from a detectable shift in engine performance.
Two Safety Recommendations were made as a result of the Investigation as follows:
- that the European Union Aviation Safety Agency commission research to determine how to prevent passengers from obstructing aircraft evacuations by retrieving carry-on baggage.
- that the European Union Aviation Safety Agency consider including a more realistic simulation of passenger behaviour in regard to carry-on baggage in the test criteria and procedures for the emergency demonstration in CS-25.
The Final Report of the Investigation was published on 6 August 2020.