History of the Flight
The aircraft departed from Lands End Airfield and was cruising at approximately 2,000 ft when the left engine failed. The pilot feathered the propeller and performed the engine failure checks, then decided to divert to St Mawgan. The aircraft's hydraulic system, which operates both the landing gear and flaps, became inoperative. Consequently, the pilot executed an orbit on final approach to manually lower the landing gear and carried out a successful flapless landing without further damage or injury.
Aircraft Description
G-BBHF was a Piper PA-23-250 fitted with two Lycoming IO-540-C4B5 fuel-injected, six-cylinder, direct drive, horizontally opposed, air-cooled engines. The aircraft had a hydraulically actuated retractable tricycle landing gear system; each landing gear leg is individually actuated by a hydraulic actuator powered by a pump driven by the left engine. In the event of hydraulic system failure, an emergency hand pump or CO2 bottle can be used to lower the landing gear.
Aircraft and Maintenance History
Both engines underwent overhaul to zero hours in January 2002 and were refitted in July 2002. In November 2002, the aircraft was grounded awaiting embodiment of Airworthiness Directive AD 2004-05-24, which required replacement of the crankshaft gear retaining bolt on both engines. This work was completed in January 2003, performed in-situ by hoisting the engines without complete removal. Following this work, pilot reports indicated the aircraft 'would not fly straight,' and in May 2003 it was discovered that the left engine mounts had been incorrectly fitted; rectification was carried out. Since overhaul, both engines had completed approximately 130 hours. The most recent maintenance was a 50-hour check on 23 January 2003, 30 hours prior to the failure. At that time, a log book entry noted 'very small amount of alloy particles found in oil filter, considered fit to continue and to be re-inspected at next 50-hour inspection.'
Engine Strip Examination
The left engine was removed and stripped in the presence of the AAIB. A hole was found in the crankcase caused by the No 4 cylinder connecting rod being forced through it. The connecting rod and bolt remains were retained within the engine cowling. Pieces comprising the complete connecting rod assembly and bolts were found, but the No 4 crankshaft bearing shell was missing. All fractures of the recovered components were consistent with overload failure, with no evidence of fatigue.
The oil filter was removed and found to contain a large amount of metallic debris. The oil pump had seized due to debris; after clearing, it rotated freely. The engine oil sump filter was nearly full of similar debris. The crankshaft and main bearings exhibited wear consistent with debris circulating in the oil. Connecting rods from the other five cylinders were undamaged, with bolts correctly torque-tightened. Piston crowns showed normal combustion deposits; piston skirts had some scoring consistent with circulating debris. Piston rings were intact and showed normal wear. The crankshaft bolt, subject to the AD, was correctly torque-tightened.
Debris from the oil was examined using a Scanning Electron Microscope and found consistent with crankshaft bearing shell material. The right engine oil filter was clean. Both magnetos tested satisfactorily; spark plugs were serviceable with clean electrodes and light deposits.
Discussion
The final uncontained failure of the connecting rod could be attributed to the break-up of the No 4 crankshaft bearing. Other connecting rods were correctly torqued; although No 4 torque could not be checked, fracture surfaces were consistent with overload, making loose bolts unlikely. Wear in other bearings was consistent with debris circulating in the oil, suggesting that some debris was already circulating, leading to the No 4 bearing breakup. At the last 50-hour check, the presence of some particles had been noted. It was not possible to identify the source of the initial contamination from the large amount of debris present in the oil.