History of the Flight
The aircraft, a Piper PA-28-161 Cherokee Warrior II registered G-BYKR, was operated by a flight training organisation at Oxford Airport. On 30 August 2006 at 0810 UTC, it departed on the first flight of the day, a training detail with an instructor, a student, and a passenger seated in the rear.
Shortly after start-up, the instructor observed symptoms of carburettor icing and set the engine speed between 1,300 and 1,400 rpm instead of the recommended 800–1,200 rpm to expedite warm-up. After taxiing approximately 1,000 m, the pre-takeoff checks were performed satisfactorily. Carburettor heat was applied for about 15 seconds at the holding point, with no signs of icing noted.
Following a short backtrack on the active runway, the student applied full power for takeoff. The engine indications appeared normal, but shortly after rotation, the engine lost power and was heard to cough. The instructor took control, radioed, and landed back on the runway. Heavy braking was applied, but the aircraft overran the paved surface, passed through the airport boundary fence and hedge, and came to rest inverted on Langford Lane, a public road. The left wing separated completely, and the right wing nearly did, allowing fuel to leak. No fire occurred. The occupants, wearing lap and shoulder belts, exited unaided after turning off the master switch. The cabin door required forceful opening. Emergency services arrived within two to three minutes.
Aircraft Information
The Piper PA-28-161 is a single-engine low-wing monoplane with a 160-hp Lycoming O-320-D3G carburetted piston engine. It has dual ignition systems and a total fuel capacity of 190 litres in two wing tanks. The aircraft was equipped with a conventional carburettor air heat system. No deferred defects relevant to the accident were carried, and no maintenance was performed on the engine or fuel system immediately prior to the flight.
Wreckage and Component Examination
The wreckage was examined at Oxford Airport. The wing spars were severed at the roots, disrupting fuel pipes. The fuselage, empennage, and nose landing gear sustained severe damage. Chord-wise scrape marks on the propeller blades and circumferential scoring on the spinner indicated the propeller was turning when the aircraft struck the fence, but the lack of blade distortion suggested low power. The engine contained sufficient oil and had not seized; valve train continuity was confirmed. Engine controls operated correctly except for the mixture control, which was stiff due to cable trapping during impact. Spark plugs were in good condition, and magnetos were secure. The fuel selector was on the right-hand tank, and the primer was locked. The gascolator was clean, and filters were clear. The electric fuel pump and ignition switch tested satisfactory. Bench tests of the carburettor confirmed proper adjustment and good internal condition; fuel filler caps were in good condition.
Engine and Fuel Testing
The engine, including the carburettor, sustained minor damage and was installed in a test facility without adjustment. After warming, high-power runs were completed satisfactorily with no hesitation. At full throttle, the engine developed near nominal rated power. Fuel samples from the gascolator, pump, and hoses were free of water and exhibited the characteristic blue colour of Avgas 100LL. A sample from the bowser used for refuelling also met specification. Fuelling records showed the aircraft received 62 litres of Avgas 100LL from Bowser AV3 the previous day; the bowser had refuelled about 40 aircraft that day with no other reported engine problems.
Discussion
The examination and testing did not identify any technical defects that could account for the loss of engine power. An intermittent ignition problem was considered unlikely due to the dual ignition system and satisfactory test runs. Water contamination was also considered unlikely given the bowser's usage and the condition of the filler caps. The conditions at the time—temperature +9°C, dewpoint +8°C (88% relative humidity)—were highly conducive to carburettor icing, and the instructor had noted icing symptoms after start. The prolonged low-power taxi may have increased the likelihood of ice forming in the carburettor. Although carburettor heat was applied at the holding point, it may not have been completely effective in removing all accumulated ice, or ice may have formed during the short taxi from the holding point to takeoff.