On 12 December 2009, an Airbus A340-600 being operated by Virgin Atlantic Airways on a scheduled passenger flight departing from London Heathrow in night Visual Meteorological Conditions (VMC) was slow to rotate and the aircraft settled at an initial climb speed below VLS - defined as the lowest selectable speed which provides an appropriate margin above the stall speed. This prompted the PF to reduce the aircraft pitch attitude in order to accelerate which resulted in a poor rate of climb of between 500 and 600 fpm. The flaps were retracted on schedule and the aircraft continued its climb. At no time was full takeoff thrust selected. Later in the climb, the crew looked again at the take off data calculation and realised that they had made the departure with insufficient thrust set and using Vr and V2 speeds which were too low for the actual aircraft weight. The flight to the planned destination was completed.
An Investigation was carried out by the UK AAIB. It was found that the aircraft had taken off using a Vr of 143 KIAS and a V2 of 151 KIAS instead of the correct values of 157 KIAS and 167 KIAS respectively. In addition, the reduced thrust used during takeoff had been reduced too much from full thrust because an assumed temperature of 74 °C had been used to determine it instead of the correct value of 63 °C. It was noted that although the flight crew had discussed the unusually high assumed (FLEX) temperature, it had not prompted a check of the take off data calculation. The origin of these errors was found to have been the entry of incorrect figures for ZFW and Fuel On Board into the FMGS so the calculated gross weight had been incorrect.
Operator SOPs were found not to have been properly followed so that there was no opportunity for the usual gross error check. In addition, the investigation noted that the cross checks that were conducted “were ineffective”.
It was noted by the Investigation that some Airbus A340 operators, although not the incident operator, use an Airbus-designed Electronic Flight Bag method to compute takeoff performance which includes in its outputs a value for a "Green Dot Speed", which is the speed which gives the best lift-to-drag ratio in clean configuration. The FMGS also calculates the Green Dot Speed independently using the ZFW, the position of the centre of gravity at the ZFW and the sector block fuel which are entered separately into the FMGS by the flight crew. This means that one calculation of Green Dot Speed has used the takeoff weight entered into the EFB and the other has used the ZFW entered separately into the FMGS. Any discrepancy between the two values for Green Dot speed then indicates a data entry error and provides a trigger for the flight crew to check all the data.
As a result of the event, the aircraft operator reiterated to its crews the correct procedure for entering take off data and the importance of the independent crosscheck, and initiated a review of their existing take off data calculation and loadsheet procedures which resulted in some minor changes. It was also concluded that incorporating the Airbus ‘Green Dot’ gross error check into their SOPs would significantly enhance them and this option was being reviewed as the Investigation was being written up.
Two Safety Recommendations were made as a result of the Investigation:
- That the European Aviation Safety Agency develop a specification for an aircraft takeoff monitoring system which provides a timely alert to flight crews when achieved takeoff performance is inadequate for given aircraft configurations and airfield conditions. 2009-080
- That the European Aviation Safety Agency establish a requirement for transport category aircraft to be equipped with a takeoff performance monitoring system which provides a timely alert to flight crews when achieved takeoff performance is inadequate for given aircraft configurations and airfield conditions. 2009-081
The Investigation also concluded that, pending a resolution from European Aviation Safety Agency (EASA), the best way to avoid a repetition was to use a comparison of the two independent calculations of ‘Green Dot Speed’ to provide an effective gross error check.
The Final Report was published on 8 July 2010 and may be seen in full at SKYbrary bookshelf: AAIB Bulletin: 7/2010 EW/G2009/12 /04