History of Flight
The aircraft, a Gulfstream AA-5B Tiger (G-BFZR), had been parked outside for two months with each fuel tank less than three-quarters full. On 15 October 2004 at 1545 UTC, the pilot, a licensed engineer holding a Private Pilot's Licence with 1,020 hours total flying experience (5 on type), was demonstrating the aircraft to two potential buyers. He conducted a thorough pre-flight inspection, taking fuel samples before and after refuelling and again just before departure, with no water visible in any sample. A satisfactory power check was performed using both fuel tanks, which were approximately three-quarters full for departure.
The aircraft lined up for takeoff with the right fuel tank selected and the fuel pump on. Normal fuel pressure of 5.1 psi was observed, and all engine indications were satisfactory. During the climb at approximately 800 feet agl, the engine suddenly lost power and stabilised at about 1,500 RPM. The pilot selected the left fuel tank and applied carburettor heat but could not restore power. During the subsequent forced landing in a large field of recently sown crop, the nose landing gear leg collapsed. The uninjured occupants, wearing lap and diagonal harnesses, exited the aircraft without difficulty using the aft-sliding canopy. Local fire services were called by ATC but not required.
Weather Conditions
The Oxford ATIS reported surface wind 270°/10 kt, visibility 10 km, cloudbase at 2,900 feet, temperature +11°C, dew point +6°C, and QNH 995 mbs.
Engineering Inspection
The aircraft was recovered to Oxford Airport and inspected. Damage to the propeller, cowlings, and landing gear was consistent with the forced landing forces. Part of the damaged engine cowling had impinged upon the oil sump 'quick drain' valve, allowing most engine oil to drain. The fuel system contains two fuel tanks outboard of wing root fairings feeding two sump tanks in the wing root fairings, with four drains (one per fuel tank, one per sump tank). Post-accident samples from these drains revealed no water. However, the carburettor bowl, electric fuel pump filter, and fuel lines aft of the firewall contained significant amounts of rusty water. Visual inspection of the fuel tanks also showed several large puddles of water in the remaining fuel. The inspecting engineer suggested the sump drains might not have been at the lowest point when parked.
Maintenance records showed the fuel system was flushed on 15 March 2004. The aircraft then flew without incident for 30 minutes on 29 March and 40 minutes on 12 August 2004. From then until the accident flight, it was parked outside with fuel tanks less than three-quarters full.
Previous Occurrences
An accident report in AAIB Bulletin No 11/2004 concerning a Gulfstream AA-5A, G-BGVW, describes a similar occurrence. An operator with considerable experience of the AA-5 reported several instances of rough running where, despite draining fuel tanks pre-flight with no water found, subsequent examination revealed water in the fuel pump filter and carburettor bowl. On another occasion, three gallons of water were drained from tanks of an aircraft left outside for a long time, attributed partly to poorly seated or perished fuel cap seals admitting water during wet weather.
Conclusion
The sudden loss of power was most likely caused by the large amount of water present in the fuel system. The quantity of water, and the type's reported history of fuel contamination, suggests that considerable water can collect before causing engine failure, and such an amount may not be detected by sampling from tank or collector drains alone.