History of the Flight
The flight was planned to position the aircraft to Coventry Airport for an annual inspection. Before departure, the aircraft was refuelled and the pilot completed a pre-flight inspection. Weather conditions at Rochester were good, with a surface wind from 200° at 6 kt and scattered cloud at 2,000 feet.
The engine was started with the auxiliary fuel pump selected to 'LO'; once running, the pump was switched 'OFF'. The aircraft taxied to the holding area where power and pre-takeoff checks were completed satisfactorily. Takeoff was from grass Runway 20L using a short field technique with 25° of flap. At 250 feet above aerodrome level, the pilot retracted the flap and noticed the aircraft was not accelerating as quickly as usual. He realised there was a reduction in power and detected a slight smell of fuel. A complete loss of engine power followed, prompting a radio call to ATC: "ZULU YANKEE ENGINE FAILURE".
The pilot lowered the nose to achieve a glide speed of 95 kt and began immediate actions, including changing the selected fuel tank. Unable to identify a clear area ahead, he scanned to the sides and noticed a small clearing in woodland about 20° to his left. He turned toward it, reduced speed to clear trees, and selected landing gear down. As the aircraft crossed the trees, he turned right to align with the clearing and landed. The aircraft touched down before the landing gear fully extended and slid approximately 40 metres on its lower fuselage before coming to rest.
The pilot switched off the electrical system and evacuated without assistance. Witnesses arrived within minutes. The slight fuel smell noted earlier persisted after the accident.
Aircraft Fuel System
Either wing fuel tank can be selected via a tank selector valve. Fuel is supplied to the engine through an engine-driven pump. An auxiliary electric fuel pump has three switch positions: 'HI', 'LO', and 'OFF'. 'HI' is for engine-driven pump failure; 'LO' is for vapour suppression. The auxiliary pump is used for priming but must be 'OFF' for takeoff. The Pilot's Operating Handbook states: "The auxiliary pump has no standby function. Actuation of the HI switch position when the engine is operating normally may cause engine roughness and/or power loss."
Examination of Engine-Driven Fuel Pump
The maintenance organisation removed the engine-driven fuel pump and conducted a rudimentary test by manually spinning it while connected to a fuel container. It failed to operate and was forwarded to the AAIB for further investigation. An overhaul agent bench-tested the pump per manufacturer's specifications; it performed adequately, though maximum RPM fuel flow was approximately 16% below the test schedule value. Two fuel flow adjustments (for maximum and idle rpm) were provided, possibly legitimately altered from factory settings. Slight fuel seepage was noted from a seal near the mixture control lever. Disassembly revealed no corrosion, contamination, or mechanical defect affecting operation.
These findings contradicted the earlier test. However, this pump type will not operate unless fully primed; it is possible that not all air was purged during the first test.
Background
The engine was replaced in August 2001 and had flown 275 hours since. The aircraft has a priming system for starting, but pilots found the replacement engine difficult to start. The maintenance organisation checked the system and found it operating correctly; the reason for the starting difficulty was never determined. To address this, an additional procedure was developed: the auxiliary electric fuel pump was set on 'LO' until the engine started.
Analysis
The pilot achieved a landing in difficult circumstances without injury. The cause of the loss of engine power has not been determined, but it is unlikely that the engine-driven fuel pump was responsible. If fuel supply was the reason, a loose fuel union was considered possible, though the maintenance organisation found no evidence of a loose connection. Further investigation was left to the maintenance organisation; if a cause becomes apparent, the AAIB will publish an addendum.
