NTSB Factual Report

On September 29, 2002, about 1630 central daylight time, a Piper PA-28R-200, N2963R, registered to a private individual, collided with trees then a light pole during an emergency landing in a residential area near Irvington, Alabama. Visual meteorological conditions prevailed at the time and no flight plan was filed for the 14 CFR Part 91 personal flight. The airplane was substantially damaged and the commercial-rated pilot, the sole occupant, sustained minor injuries. The flight originated about 10 minutes earlier from the Roy E. Ray Airport, Bayou La Batre, Alabama.

The pilot stated that after landing following a 2 hour 55 minute flight from Florida, he filled both fuel tanks then performed a preflight inspection using a checklist; no contamination was noted when checking the fuel tanks. He performed an engine run-up before takeoff also using a checklist, no discrepancies were noted. He performed a takeoff with 10-degrees of flaps selected and reported that when he achieved a positive rate of climb, moved the gear selector to the "up" position. The gear remained down due to the fact that he had the automatic gear extender system installed that lowers the gear automatically when the airspeed is below a certain value. The flight continued and the engine suddenly then quit. While descending, the airplane collided with trees; the next thing he remembers is being in the airplane on the ground.

According to the FAA inspector who examined the accident site, the airplane impacted a dirt road with the landing gear extended and the left wing of the airplane separated after colliding with a light pole 6-7 feet above ground level. The airplane then spun around and came to rest upright. The inspector took a sample of fuel from the left wing fuel tank and also from the facility that fueled the airplane; visually no water or algae were noted. An API gravity test was performed on the specimen taken from the left wing tank revealing a gravity of 68.8, which is a typical value for 100 low lead fuel. (A copy of the fuel report is included).

An NTSB investigator examined the engine and confirmed crankshaft, camshaft, and valve train continuity; compression and suction were noted in all cylinders. Continuity to the accessory case including operation of the magnetos, and vacuum pump were also confirmed. The magnetos sparked to all plugs of all cylinders, and the engine driven fuel pump was operational. No blockage of any fuel lines from the tank to the engine was noted.

The servo fuel injector, Bendix Model RSA-5A D1, parts list number 2524450-2, all four fuel injectors, all four fuel injection lines, and the fuel manifold valve part number 2524397-1 were retained for further examination. Bench testing of the servo fuel injector revealed that the unit flowed 45.8 pounds per hour (pph) at full throttle. The full throttle service limits for the fuel servo injector are 82.9 to 94.0 pph. Disassembly revealed white solid contaminates, water droplets, and rust inside of the fuel regulator, fuel diaphragm, mixture control unit, and the fuel strainer and housing. Additionally, the fuel diaphragm stem was checked using a dial indicator and found to have 0.008 inch deflection. The fuel manifold valve, all four fuel injector nozzles and lines were also tested and operational checks were good. (A copy of the fuel flow test sheet is included).

Fuel samples of the right wing fuel tank were taken from the top of the tank and also from the fuel tank sump drain by the recovery crew. The sample taken from the top of the tank was clear and blue in color while the sample taken from the sump drain appeared cloudy with a white residue. The sample that appeared cloudy was submitted to Panair Laboratory, Inc. for further analysis. The white contaminant was found to be made mostly of lead, with trace amounts of aluminum and cadmium. According to the president of Panair Laboratory, Inc., the tetraethyllead in the fuel formed a metallic lead oxide from oxidation and precipitated from the fuel. (A copy of the Panair fuel analysis report is included).

A review of the maintenance records shows that the engine was remanufactured on July 19, 1973, and installed on the aircraft September 26, 1973. The engine was overhauled on May 21, 1990, and had accumulated approximately 341.09 hours since then at the time of the accident. The engine received a 100 hour inspection on August 20, 2002, and had accumulated 15.39 hours since that inspection. The servo injector had the original Lycoming lead seal, and was manufactured, April 28, 1969; there was no record that it ever had been overhauled.

Service Bulletin PRS-97 released by Precision Airmotive Corporation on November 11, 1991, establishes overhaul and calibration of the fuel injection system to be done every 10 years. Compliance with the Service Bulletin is not required by federal aviation regulations. (A copy of the Service Bulletin is included).

The aircraft wreckage minus retained components and maintenance records was released to Marshall Dean of USAIG on November 5, 2002. The retained components and maintenance records were released to Marshall Dean of USAIG on November 12, 2002. Additional maintenance records provided by the pilot were also released to Marshall Dean on November 25, 2002.

NTSB Probable Cause

The loss of engine power due to internal contamination (rust and water) of the servo fuel injector resulting in it flowing approximately half of the required fuel flow at full throttle. A contributing factor was the unsuitable terrain encountered by the pilot during the forced landing.