BE20, Nadi Fiji, 2010

BE20, Nadi Fiji, 2010


On 25 April 2010, a Beech King Air touched down at Nadi with its landing gear in the transit position after flying a night approach during which a significant electrical system failure had occurred. The landing gear retracted and the aircraft left the runway to the side and came to a stop resting on its fuselage. The Investigation attributed the electrical failure, which directly affected the landing gear operating system and required two diodes to have both failed was likely to have meant that one would have failed on an earlier occasion with no apparent consequence.

Event Details
Event Type
Flight Conditions
Flight Details
Type of Flight
Public Transport (Passenger)
Flight Origin
Intended Destination
Take-off Commenced
Flight Airborne
Flight Completed
Phase of Flight
Location - Airport
Non Precision Approach
Inappropriate crew response (technical fault), Ineffective Monitoring, Plan Continuation Bias, Procedural non compliance, Stress
Directional Control, Off side of Runway
Electrical Power
Component Fault in service
Damage or injury
Aircraft damage
Non-aircraft damage
Non-occupant Casualties
Off Airport Landing
Causal Factor Group(s)
Aircraft Operation
Aircraft Technical
Safety Recommendation(s)
None Made
Investigation Type


On 25 July 2010, a Beech B200 Super King Air (F-OIAN) being operated by Air Alizé on a public transport flight transporting medical personnel from Nouméa-Magenta, New Caledonia to Nadi, Fiji touched down at destination with its landing gear in the unlocked transit position after flying a night Visual Meteorological Conditions (VMC) approach during which a significant electrical system failure had occurred. The landing gear retracted and the aircraft left the runway to the side and came to a stop resting on the fuselage with the propellers bent and flaps damage. The pilots and the two passengers on board were able to exit unaided. One runway edge light was destroyed.


An Investigation was delegated to the French BEA by the Fiji Islands civil aviation authorities and a field investigator was sent to Nadi from the French overseas territory of New Caledonia to begin the Investigation.

Neither an Flight Data Recorder (FDR) or Cockpit Voice Recorder (CVR) were fitted to the aircraft and were not required to be. It was noted that Air Alizé was authorised to perform medical evacuation flights between the Fiji Islands Nouméa and routinely used the Beech King Air to do this. Both the 60 year old aircraft commander and the 54 year old First Officer were experienced on the aircraft type. The latter worked only part time for Air Alizé and otherwise was a qualified helicopter pilot and TRE(H) working for another company. Although the Beech King Air is certified for operation by a single pilot, when used as a public transport aeroplane by Air Alizé, it was flown by two type-rated pilots.

It was established that the landing at Nadi was to facilitate refuelling only and that the flight was destined for Wallis and Fortuna aerodrome to pick up and transfer a patient to Nouméa. About two hours after taking off from there, the night descent towards Nadi began in Instrument Meteorological Conditions (IMC) and a clearance was subsequently given to make an ILS approach to runway 02. Shortly after overflying a locator beacon positioned at 9.6nm from touchdown on the localiser, a landing clearance was given with a wind check of 110°/13 knots and advice that it was raining and the runway was wet.

The crew stated that having selected the landing gear down, they had seen 'three greens' indicating that it was locked down. As the aircraft emerged into VMC at about 1800 feet, the crew noticed, as a result of the windshield wipers stopping and the landing/ taxi and centre instrument panel lighting going off, that there had been a partial electrical failure. The crew located pocket torches and decided to continue the approach and did not advise ATC of any technical problem. They also subsequently said that the prevailing weather conditions around the aerodrome (see below) did not encourage them to abandon the approach.

At about 1,000 ft, the extension of the flaps to the landing position triggered the aural warning for non-extension of the landing gear but the crew presumed that this (and the now unlit 'three greens' landing gear down and locked indication) were a consequence of the observed partial electrical failure and after checking the position of the flap and landing gear levers, continued the approach. They did not consider the possibility that the gear might not be locked down.

The aircraft in its final stopping position on the grass adjacent to the runway - reproduced from the Official Report

The First Officer subsequently stated that the landing was initially normal but the landing gear then collapsed and the propellers hit the surface of the runway before it left the runway to the right. It came to a stop on a heading of 055° with the fuselage in direct contact with the grass, approximately 1560 metres beyond the runway threshold. The landing gear was subsequently found to be retracted, although the landing gear doors were open.

A weather report issued at about the time of the accident - 0200 local time on 26 April - gave 130°/13 knots, visibility 10km, thunderstorm with rain and cloud FEW CB at 1800 feet, SCT Towering CU at 2800 feet and OVC at 10,000 feet with a coded remark indicating that there had been 7.1mm of rain in the previous hour.

It was found that none of the five abnormal procedures documented for the aircraft electrical system in the Aircraft Flight Manual (AFM) were relevant to the failure which occurred.

The Investigation sought to establish the nature of the electrical system failure and the reason for the landing gear retraction at touchdown. It was found that that electrical failure had been confined to the "no 2 secondary circuit" which supplied power to various equipment and lighting systems including the windscreen wipers and some flight deck lighting as well as the landing gear control unit and the landing gear status lights. It did not supply power to the landing gear warning horn.

It was found that the reason for the electrical system failure was that two 70 amp diodes which were inaccessible to the crew, at least one of which was required to allow the "no 2 secondary circuit" to function, had failed. It was considered that one had almost certainly already been inoperative before the other having failed when the landing gear was selected down on the approach to Nadi. Once the two diodes had been replaced, the landing gear system was found to work normally. None of the CBs visible to the crew had tripped.

The aircraft manufacturer provided statistics which indicated that the in-service failure rate for the type of diodes which failed when fitted to both the accident aircraft type and to other aircraft types was "extremely low". It appeared that these diodes were an 'on condition'; component with no maximum service life but subject to "periodic inspection" every 600 flying hours.

When the aircraft was lifted, the landing gear doors were found to be in the open position and when the battery was reconnected, the warning horn for the landing gear could be heard. The landing gear locking system was checked and found to be working normally. It was found during ground testing of an aircraft of the same type supported on jacks that selection of gear down associated with an almost simultaneous cut-off of the electrical power supply to the "no 2 electrical circuit" led to the complete opening of the landing gear doors as well as the beginning (only) of the extension of the landing gear. The position of the doors and the landing gear was then similar to that noted when the accident aircraft had been lifted. During this test, the three green gear status lights remained unlit and landing gear actuation using the gear selector was no longer possible. When either one of the power levers was retarded to a low power position, the gear warning horn sounded to indicate that the landing gear was not locked.

After carefully considering all possibilities, the Investigation could not find a way to explain the sequence of events described by the crew, i.e. the illumination of the green landing gear down status lights followed by the unlocking of the landing gear with no action taken by the crew on the associated controls.

It was concluded that the decision to continue the approach without performing a proper evaluation of their situation after the obvious electrical malfunction was probably reinforced by:

  • The fact that they stated that they had seen three greens indicating that the landing gear was down after actuating the landing gear selector
  • The absence of any tripped circuit breakers on the various flight deck panels;
  • The turbulent flying conditions at night with rain with the windscreen wipers no longer operating due to the electrical failure;
  • The fact that the necessary visual references for landing had almost certainly been acquired by the time the gear warning horn sounded.

The Causes of the Accident were summarised as follows:

At night in adverse meteorological conditions, the crew decided to continue with the approach and to land, while the landing gear was not locked in the down position due to the failure of the secondary electric circuit.

The origin of this failure was probably the increase in the number of operating cycles of the hydraulic power pack which delivers the hydraulic energy required to extend and retract the landing gear. This led to the failure of the diodes in the protective secondary circuit, which could not be checked in flight.

The decision to continue the approach resulted from incorrect comprehension of the electrical failure and the aural warning that sounded during the final approach. The conviction that the landing gear was locked down, associated with the meteorological conditions and the failure of the circuit-breakers to trip, did not prompt the crew to correctly assess the risk associated with the electrical failure."

The Final Report was published in English translation on 24 July 2014 following publication in French in June 2014. No Safety Recommendations were made.

Further Reading

SKYbrary Partners:

Safety knowledge contributed by: