History of the Flight
The aircraft, a Piper PA-34-200 Seneca (registration G-BETT), was conducting a Continued Airworthiness Flight Test (CAFT) from Biggin Hill Airport in Kent on 22 July 2004. The flight was part of the Certificate of Airworthiness renewal process. The flight test pilot occupied the left seat, and the aircraft owner acted as second pilot and observer in the right seat. The aircraft was prepared with full fuel tanks and underwent a thorough pre-flight inspection. All engine ground checks were completed according to the flight test schedule.
The aircraft departed Biggin Hill at 1552 UTC in conditions of light wind, good visibility, with scattered cloud at 2,500 and 5,000 feet. Temperature was 20°C, dew point 13°C. After departure, the pilot contacted the approach controller, reporting he would operate at 2,400 feet initially. He was asked to report when ready to recover to the airport, but no further transmissions were made.
The flight test schedule required an in-flight engine shutdown and single-engine climb. The right engine was shut down, and its propeller feathered. The climb rate was as expected, with the left engine behaving normally on maximum continuous power. The right engine was then restarted, and although not required by the schedule, the crew decided to repeat the exercise on the other engine. The right engine did not start as readily as expected but, once running, was allowed to warm up before the left engine was shut down and the single-engine climb repeated. The pilot reported the aircraft was at about 3,000 feet at this stage.
After about 30 to 45 seconds into the single-engine climb, the right engine experienced a sudden power loss, reducing to approximately 1,200 RPM, and did not respond to throttle movement. The pilot was uncertain whether the engine was running at reduced power or windmilling.
Attempted Restart and Crash Landing
The pilot's first action was to attempt to restart the left engine. The second pilot recalled that the left engine unfeathered but did not start, while the pilot reported the engine did not unfeather despite oil and fuel pressure and a healthy battery. The second pilot directed the pilot to focus on flying while he attempted to start the engines. Suspecting flooding, he selected throttles fully open and set mixtures off, expecting the engines to fire, but they did not. The pilot reported that fuel booster pumps were most likely selected off initially but would have been selected on when it became clear the engines were reluctant to start.
Both pilots expected to restart at least one engine. However, with the aircraft at very low altitude and reducing airspeed, the pilot realized a crash landing was imminent and warned the second pilot. There was no time or altitude to maneuver further. With landing gear and wing flaps retracted, the pilot executed a crash landing into a cornfield immediately ahead. After impact, the aircraft ran onto softer ground and came to rest against a fence after a short ground slide of about 80 feet.
Damage and Injuries
The aircraft sustained a major fuselage fracture aft of the wing and was written off. Both pilot seats and the cockpit area distorted under vertical deceleration, though the main door operated normally and was used by the crew to escape. The pilot suffered a broken eye socket and broken wrist; the second pilot suffered broken bones in his back and ribs. There was no fire. The second pilot alerted emergency services using his mobile telephone.
Flight Test Context
Pilots conducting CAFT on aircraft under 5,700 kg maximum all-up weight must be acceptable to the CAA in flying experience and recency and must be briefed prior to undertaking CAFT. The pilot of G-BETT met these requirements. The accident flight was conducted per Flight Test Schedule No. 3, issue 2, which applies to twin, piston-engined, unpressurised aeroplanes up to 5,700 kg maximum AUW. The schedule stipulates a minimum of 3,000 feet above terrain for the single-engine climb and allows the climb with either engine inoperative, but does not require repeating on the other engine.
Discussion
Both pilots believed the initial power loss suggested an interruption of fuel to the right engine. The aircraft had not been flown for some time before the flight and was fueled to full. Routine water contamination checks were performed, and post-accident inspection showed uncontaminated fuel present in fuel lines, pumps, and filters. All magnetos had recently undergone overhaul and operated correctly when tested after the accident.
The aircraft owner had previously experienced fuel flooding during warm engine starts on Seneca aircraft with Lycoming IO-360 engines. Although no obvious cause for the power loss was found, he believed the restart attempts failed due to flooding. Enquiries with other Seneca operators supported the view that flooding is possible during restart, especially with the fuel booster pump on. However, the accident pilot did not think this likely and believed he would have recognized flooded engine symptoms.