MIA99LA016

The instructor simulated the failure of the left engine by reducing the manifold pressure to 11 inches Hg., while on the downwind leg. She stated they did not clear the engine during the visual approach and that the engine had been running rich. On short final approach, the instructor took control of the aircraft and applied full power to both engines for a go-around. The left engine did not accelerate and the aircraft rolled rapidly to the left. The instructor reduced engine power and the aircraft rolled to a level attitude and touched down in a side slip on the runway, collapsing the landing gear. Postcrash examination of the left engine showed the spark plugs were fouled with combustion deposits and the fuel control was operating at a rich mixture in the idle range due to sand contamination.

On October 27, 1998, about 1241 eastern standard time, a Piper PA-34, N45SA, registered to International Aeronautical Consultants, Inc., crashed during a go-around at Ormond Beach Municipal Airport, Ormond Beach, Florida, while on a Title 14 CFR Part 91 instructional flight. Visual meteorological conditions prevailed at the time and no flight plan was filed. The aircraft received substantial damage and the commercial-rated flight instructor, commercial-rated dual student, and one passenger were not injured. The flight originated from Daytona Beach, Florida, the same day, about 1130.

The instructor stated they had performed an instrument approach to runway 17 at Ormond Beach Municipal Airport followed by a circle to land on runway 08. They performed a go-around and entered the traffic pattern for a visual approach to runway 08. On the downwind leg she simulated failure of the left engine for the student by moving the throttle to 11 inches of manifold pressure. The student continued the approach with the simulated failure of the left engine and on final approach the student allowed the aircraft to drift to the right of runway centerline. The aircraft's nose was pointing to the left and the aircraft was approaching the runway at an angle. The flight instructor took control of the aircraft and moved both throttles to the full power position to perform a go-around. The left engine did not respond and the aircraft's left wing dropped down. The flight instructor closed both throttles and the aircraft returned to a wings level attitude. The aircraft then impacted hard on the runway while angled to the left. The landing gear collapsed and the aircraft came to rest upright. The instructor stated further that the left engine had been "running rich" in the recent past, and that they did not clear the engine during the simulated engine out approach to runway 8.

Postcrash examination of the left engine was performed by a FAA inspector and a representative of Lycoming Engines. Fuel was found in the engine fuel system. The engine-driven fuel boost pump and electric driven fuel boost pump operated normally. The magneto timing was checked and the left magneto was found set to 24 degrees before top center (BTC) and the right magneto was set to 25 degrees BTC. Lycoming Service Instruction No. 1325A, dated May 14, 1976, states the previously approved 25 degrees BTC setting should be changed to 20 degrees BTC. Each magneto produced spark through the respective leads when the engine was rotated by hand. Continuity of the engine crankshaft, camshaft, valve train, and accessory drives was confirmed. Each cylinder produced normal compression. The spark plugs were removed and found to have black to brown soot deposits on them. The spark plug models found in the engine were not approved by Lycoming Engines for use in the engine. (See attached Lycoming Engine Report)

The engine fuel system was tested under NTSB supervision at a fuel system overhaul facility. The inlet screen of the fuel servo was checked before testing and found to contain sand. The sand was removed and the unit was tested. The servo operated normally except at the idle setting. The fuel flow as tested at the idle setting was 37.6 pounds per hour. The specification calls for a fuel flow of 22 to 28 pounds per hour. The servo was disassembled and sand was found in the internal areas of the unit, which prevented the diaphragm ball from seating and giving a normal idle fuel flow.

The fuel manifold operated normally when tested and the fuel injector nozzles operated normally. Each of the fuel injector nozzles was found to have cracks in the fuel line attach areas. Two of the injectors had small leaks due to these cracks.