19980426007679A

Lost power on climbout. Ditched in bay. Throttle disconnected. (.4)on April 26, 1998, about 1215 eastern daylight time, a Piper PA-18, N225NY, registered to a private individual and operated by advertising air force, inc., crashed into the tampa bay, near st. Petersburg, Florida, shortly after takeoff from the Albert Whitted Airport, st. Petersburg, Florida. Visual meteorological conditions prevailed at the time and no flight plan was filed for the 14 CFR part 91 banner tow flight. The airplane was substantially damaged and the commercial-rated pilot, the sole occupant, was not injured. The flight originated about 1212 from the Albert Whitted Airport. The pilot stated that before takeoff, he performed a static run-up of the engine which included applying carburetor heat. He noted after removing carburetor heat, that the handle was "loose." He then initiated the takeoff and during the initial climb, he heard a loud "popping" sound from the engine and the engine experienced a loss of power accompanied by black smoke. He called "mayday" and while attempting to return to the airport, the airplane was inadvertently stalled and impacted the water nose and left wing low. He furthe r stated that at full rated power, the mixture had to be leaned, and that condition had existed for about 7 years. According to the operator, he had been employed there as a pilot on and off for the past 8 years. Examination of the airplane by an FAA airworthiness inspector revealed that the left main landing gear was separated from the airframe and the engine firewall was damaged. The flight controls were inspected with no evidence of failure or malfunction. Fuel was found in the carburetor bowl and fuel lines; 35 gallons drained from the fuel tanks. Examination of the engine revealed crankshaft, camshaft, and valve train continuity. The magnetos were determined to be properly timed to the engine and were found to operate normally. The carburetor heat control cable was found separated at the control lever at the airbox assembly, but the attach hardware was in place. No portion of failed cable was found in the hole of the attach bolt. The carburetor was removed for further examination. Disassembly of the carburetor revealed that the float level was too high due to the worn condition of both the head of the float needle valve, and the adjustment tab of the float assembly. Evidence of previous fuel leakage was noted from the carburetor bowl vent due to cleaning of the accumulated deposits upstream of the vent opening in the throat area of the carburetor. Additionally, the nozzle assembly which was correct by part number, was bench tested and found to be out of limits (high) in terms of fuel flow. This was caused by the worn condition of the nozzle assembly. A date stamp on the float bracket for the metal float indicates that the bracket was manufactured by July 1984. According to the carburetor manufacturer, it was manufactured on September 24, 1962, with no record of it ever being sent to the manufacturers facility for overhaul. At manufacturer, a composite type float was installed. In April 1984, the manufacturer issued a service bulletin which specified changing the composite float to a metal type float. Review of the provided aircraft maintenance records revealed no record of the installed carburetor being overhauled. The records indicate that on July 29, 1997, a one-piece venturi was installed in accordance with airworthiness directive (AD 93-18-03). The records also indicat that on April 18, 1998, the airbox shaft bushings were repaired with clamps. The maintenance records also indicate the accident carburetor was installed on the engine on June 3, 1996, and had accumulated 1,246 hours since installation in 1996, at the time of the accident. (-23) see narrative #MIA98LA144