India A20N near Kolkata on Jan 5th 2019, fuel leak

Last Update: February 27, 2020 / 18:31:42 GMT/Zulu time

Bookmark this article

Incident Facts

Date of incident
Jan 5, 2019

Classification
Incident

Cause
Fuel leak

Airline
Air India

Flight number
AI-335

Departure
Bangkok, Thailand

Destination
Delhi, India

Aircraft Registration
VT-EXV

Aircraft Type
Airbus A320-200N

ICAO Type Designator
A20N

An Air India Airbus A320-200 Neo, registration VT-EXV performing flight AI-335 from Bangkok (Thailand) to Delhi (India) with 145 people on board, was enroute at FL380 about 210nm southeast of Kolkata (India) when the crew detected a fuel leak on the right hand engine, declared emergency, shut the engine down and drifted down to FL220. The aircraft diverted to Kolkata for a safe landing about 35 minutes later.

The airline reported the aircraft diverted to Kolkata due to a technical issue. A replacement aircraft having arrived into Kolkata as flight AI-22 from Delhi took the passengers to Delhi. The occurrence aircraft ferried to Delhi.

A replacement Airbus A321-200 registration VT-PPF departed Kolkata as flight AI-335 about 4:45 hours after landing and reached Delhi with a delay of 5 hours.

The occurrence aircraft remained on the ground in Kolkata for about 31.5 hours, then returned to service.

On Feb 27th 2020 India's DGCA released their final report concluding the probable cause of the incident was:

The cause of the incident was fatigue failure of fuel hose connecting the Main FUEL-Oil Heat Exchanger to the Engine driven FUEL PUMP.

The DGCA reported the captain (31, ATPL, 5,880 hours total, 5,680 hours on type) was pilot flying, the first officer (35, CPL, 1,460 hours total, 710 hours on type) was pilot monitoring.

The DGCA described the sequence of events:

Crew monitored both Fuel on-board and BURN FOB at waypoints en-route as required and made record of the same in the flight navigation log, the flight was uneventful till waypoint ITN'. At 16:03UTC an ECAM warning was generated indicating a disagreement between the Fuel onboard and the fuel used. The aircraft was in contact with Yangoon ATC at this point of time.

The cockpit crew called for the SNY (supernumerary crew-PIC rated) from the cabin and he assisted the flight crew in carrying out the fuel leak procedure. Four 04) minutes after ECAM warning generation the flight crew was advised by Yangoon ATC to contact Kolkata ATC. The crew was unable to locate any visual evidence of leaks from the aircraft's RH wing or engine. The operating crew coordinated with Kolkata ATC with the help of two other aircrafts, who relayed the messages for AIC-335, as aircraft was unable to contact Kolkata ATC on radio. However, the crew was able to receive the radio broadcast from Kolkata ATC. Upon reaching waypoint `IBITA' the fuel on-board had reduced to 8.5tons against an expected 9.5 tons.

The flight was diverted to Kolkata, and 'MAYDAY' was declared by the crew with SQWAK code set to '7700'. Following the procedure, crew suspected fuel leak from Engine #2, as Engine#2 fuel consumption was observed to be approximately 180kg more than Engine#1 in 15 minutes. Subsequently Engine#2 was shut-down at 1625UTC by the operating crew and descended the aircraft to lower flight levels, and continued to descent in co-ordination with Kolkata ATC. After shutdown of engine#2 the fuel leak stopped.

After landing at Kolkata fuel leak was observed from Engine #2 and during wet motoring, fuel was found leaking from the fuel hose connecting Main Heat Exchanger to Main Fuel pump.

The DGCA analysed:

The crew did not detect the fuel leak form the fuel readings taken at waypoints en-route, as the fuel readings recorded against time check (1545UTC) at waypoint PTN was found to be recorded as 3010kg(Fuel Burn) and 10600kg(Fuel-on board); whereas as per FDR data the fuel remaining on-board at this point of time was only 9899kg. The discussion between the cockpit crew members did not indicate that, either of them noticed the discrepancy of the fuel figures prior to the system generated ECAM warning.

The crew became aware of the fuel discrepancy only on activation of the ECAM generated warning at 1603UTC. After activation of the ECAM warning the SNY(SUPERNUMERARY CREW-PIC RATED) was called to the cockpit and he assisted the crew in the further actions.

The First officer performed the visual check of RH engine and wing and did not find any trace of fuel leak, following which the crew calculated the drop in fuel levels in RH and LH wing inner tanks for 06, 13 minutes and they ascertained that RH wing inner tank fuel depletion would be more than 300kg in 30 minutes in comparison to the LH inner wing tank reduction.

The flight was diverted to Kolkata and a "MAYDAY" was declared. The RH side wing tank fuel depletion was calculated to be more than the fuel consumption of the RH engine. Following QRH procedures, the RH engine was shutdown at 16:27UTC and crew further monitored the fuel remaining w.r.t time and confirmed the fuel leak was from Engine#2. The fuel levels stabilised after shutdown of Engine#2. Hence, Engine#2 was maintained in the shutdown condition until landing at Kolkata.

The DGCA found about 1045kg of fuel had been lost due to the fuel leak.

The occurrence engine had been sent to the CFM shop for inspection in August 2018. The fuel hose between fuel pump and FOHE had been requested to be replaced during the shop visit.

In Nov 2018 CFM identified another leak issue on the new hose type and issued a Service Bulletin instructing to replace the hose every 1200 hours. The time of compliance with the SB had to be extended however as insufficient numbers of fuel hoses were available.

The left hand engine of the occurrence aircraft went past the 1200 hours limit as result of the shortage of fuel hoses. The right hand occurrence engine's fuel hose however had accumulated 1079 hours only and thus was still within the requirements of the SB.

The DGCA stated:

A crack was confirmed and located in extremity of the weld seam between the rigid part and the convoluted part of the fuel tube, measuring about 0.47 in (12.1mm) long. Fuel leak was confirmed during the test in the flexible part of the tube.

A transgranular fracture surface in HCF vibratory fatigue under an unilateral bending mode, initiated on the inner wall side at the junction of the weld seam and the convolute. Crack propagation in the flex hose due to interaction between hose frequency response and fuel system.

A re-designed fuel hose is being made available for retrofit.