A333, vicinity Orlando FL USA, 2013
From SKYbrary Wiki
|On 19 January 2013, a Rolls Royce Trent 700-powered Virgin Atlantic Airbus A330-300 hit some medium sized birds shortly after take off from Orlando, sustaining airframe impact damage and ingesting one bird into each engine. Damage was subsequently found to both engines although only one indicated sufficient malfunction - a complete loss of oil pressure - for an in-flight shutdown to be required. After declaration of a MAYDAY, the return to land overweight was completed uneventfully. The investigation identified an issue with the response of the oil pressure detection and display system to high engine vibration events and recommended modification.|
|Actual or Potential
|Airworthiness, Bird Strike, Loss of Control|
|Flight Conditions||On Ground - Normal Visibility|
|Type of Flight||Public Transport (Passenger)|
|Intended Destination||Manchester International Airport|
|Actual Destination||Orlando International|
|Take off Commenced||Yes|
|ICL / ENR|
|Location - Airport|
|Airport vicinity||Orlando International|
|Tag(s)||Extra flight crew (no training)|
|Tag(s)||Bird or Animal Strike|
|System(s)||Engine - General|
|Damage or injury||Yes|
|Causal Factor Group(s)|
On 19 January 2013, an Airbus A330-300 being operated by Virgin Atlantic Airways on a passenger flight from Orlando to Manchester UK with an additional heavy crew member present on the flight deck for take offhit birds shortly after making a night take off from runway 35L in Visual Meteorological Conditions (VMC) and high engine vibration was followed by indications of a total loss of oil pressure from the left hand engine, which was shut down. A MAYDAY was declared and an uneventful return to land was completed.
The Investigation was delegated by the National Transportation Safety Board (USA) (NTSB) to the UK AAIB acting for the State of the Operator. Data was obtained from the 25 hour Flight Data Recorder (FDR), the 2 hour Cockpit Voice Recorder (CVR) and the Digital ACMS Recorder (DAR) which provided details for the entire incident flight. It was found that the Operator’s procedures for the protection of the CVR following a Serious Incident did not extend to the FDR and that the preservation of the required data from that source was fortuitous.
It was established that, after flexible thrust take off from runway 35L with the Co-Pilot as PF and a heavy crew member, another Senior First Officer, present in the flight deck for take off, the aircraft had suffered multiple bird strikes as it passed 530 feet aal. A loud noise accompanied the impact of a bird with the nose of the aircraft and shortly afterwards, Electronic Centralized Aircraft Monitor (ECAM) indications of a malfunctioning left engine were annunciated. The crew engaged both autopilots in heading mode and the designated PF continued in that role whilst the aircraft commander analysed the problem with the assistance of the third pilot. Significant vibration was felt through the airframe which indications suggested was likely to be from the left engine which was now showing a low oil pressure annunciation on the ECAM and an indication of zero oil pressure on the gauge. This required idle thrust to be selected on that engine and, when those indications persisted, that the engine should be shut down which was accomplished. A ‘PAN’ call was transmitted to TWR who acknowledged it and instructed a change to radar on a different frequency.
On initial contact with Radar, the flight status was upgraded to ‘MAYDAY’ and a return requested. The aircraft was levelled at 3,000 feet where the crew reviewed the situation and the commander then downgraded their status to PAN and took over as PF. Single engine landing performance overweight was reviewed and radar vectors for an ILS approach to Runway 36R were accepted subject to a single orbit to allow the crew additional time to complete checks. The landing was uneventful and the aircraft taxied clear of the runway for an Rescue and Fire Fighting Services inspection which confirmed hot brakes and so a tow in to the parking gate was provided. The total flight time was 30 minutes.
The comprehensive arrangements implemented by the airport authority for bird hazard control were noted as was the relatively low bird strike rate - 0.36 strikes per 1000 movements in 2011 of which arriving aircraft had accounted for 72% of strikes.
Recorded data on engine function showed that following the ingestion, the left engine N1 shaft vibration had increased from 0.4 to 10 units - the maximum figure, EGT had increased by 10°C to 771°C and all three engine shaft speeds, N1, N2 and N3, increased slightly. Right engine N1 shaft vibration also increased at the same time from 0.2 to 1.8 units. As the left engine speed had reduced, the N1 shaft vibration level had also reduced to 0.1 units. The total time from the initial ECAM warning to shutdown was 1 minute 46 seconds. With MCT set on the right engine, the maximum recorded N1 shaft vibration during the remainder of the flight was 2 units which was below the trigger for an ECAM N1 shaft vibration advisory message.
“Examination of the aircraft in Orlando revealed impact damage to the radome, the left engine nose cowl, three fan blades fitted to the left engine and two fan blades fitted to the right engine. No defects were identified with the engine oil system. Analysis of the bird remains recovered from the engines indicated that the birds were probably Ring-necked Ducks, of between 1.5 lb and 2 lb in weight, and that each engine had ingested one bird.”
The technical part of the Investigation then sought to understand why the left engine oil pressure indication had instantly dropped to and remained at zero. It was established that oil pressure monitoring was provided by three sensors - two oil pressure transducers mounted on the left side of the engine fan case and an oil pressure switch mounted on the gearbox-driven oil pump. The pressure transducers provide oil pressure readings to the EEC which are then fed to various aircraft systems and the oil pressure switch generates a signal in the event of loss of oil pressure. The ECAM generates a low oil pressure message when any two of the three sources of oil pressure sensing indicate low oil pressure. In addition, in order to protect the engine bearings if lubrication fails, an oil pump failure condition will be assumed by the EEC if it detects both transducer outputs to have a specified negative differential pressure within a three second period and the EEC will then overwrite the oil pressure values being transmitted with a zero value which leads to the ECAM reading of zero oil pressure seen after the bird ingestion. Crucially this logic is designed so that once activated, the EEC remains latched in this condition until it resets during engine shutdown.
This means that even if the oil pressure recovers or stabilises, the ECAM message would remain illuminated and require an engine shutdown response from the crew.
Rolls Royce advised their awareness of seven previous high vibration events which had generated a low oil pressure message on Trent 700 engines of which five had led to a precautionary engine shutdown . Five of the seven events had followed IDG failure and one had followed a bird strike. It was noted that the Trent 800 engine, which has a similar lubrication and oil pressure monitoring system, is not set to zero if negative pressure differentials are transmitted from the transducers and there have been no reported low oil pressure events due to high shaft vibration on Trent 800 engines. The Investigation noted that Rolls Royce were in the process of considering an appropriate modification to Trent 700 engines but they were unable to give an indicative timescale for resolution.
One Safety Recommendation was made as a result of the Investigation as follows:
- that Rolls-Royce plc modify the oil pressure indication and failure detection systems of the Trent 700 engine to minimise the possibility of an activation of the Electronic Engine Controller oil pump failure logic as a result of high vibration or an Integrated Drive Generator failure. [2013-15]
The Final Report of the Investigation was published on 12 September 2013.