ERA10LA370

The accident flight was the first flight after an overhauled fuel servo was installed on the airplane. During initial climb, about 600 feet above ground level, the engine experienced a partial loss of power. The pilot performed a 180-degree turn and attempted to land in the opposite direction of takeoff; however, the engine lost all power during final approach, and the airplane impacted a fence about 150 feet prior to the runway. Examination of the wreckage did not reveal any anomalies with the airframe or engine, and adequate fuel was onboard. A subsequent teardown examination of the fuel servo revealed that small, clear, spherical particles were found on the threads of the inlet fitting and plug of the fuel inlet screen. The particles were consistent with glass beads and were also found in the manual mixture control valve chamber and on the unmetered side of the fuel diaphragm. One particle was found underneath the diaphragm. According to the Federal Aviation Administration principle maintenance inspector of the company that overhauled the fuel servo, the glass beads found in the accident fuel servo were similar to the glass beads used by the company. Although the company did not bead-blast their fuel servos, bead-blasting was performed in other locations in the factory, including areas approximately 20 feet from where the fuel servos were overhauled.

On July 20, 2010, about 1630 eastern daylight time, a Piper PA-28R-200, N8900E, operated by Tradewind Aviation International, was substantially damaged during impact with a fence, following a loss of engine power while departing Coastal Carolina Regional Airport (EWN), New Bern, North Carolina. The certificated commercial pilot and passenger were not injured. The maintenance test flight was conducted under the provisions of 14 Code of Federal Regulations Part 91. Visual meteorological conditions prevailed and no flight plan was filed for the local flight. According to a Federal Aviation Administration (FAA) inspector, the engine was recently overhauled and the airplane flew uneventfully during several subsequent flights, for a total of approximately 6 hours. During that time, pilots reported difficulty starting the engine. Once the engine was started, there were no other problems during the flights. Through trouble-shooting, a certificated airframe and powerplant mechanic diagnosed the starting problem as the result of a faulty fuel servo. The mechanic replaced the fuel servo with an overhauled fuel servo. The accident flight was the first flight after the overhauled fuel servo was installed. During takeoff from runway 22, the engine experienced a partial loss of power, about 600 feet above ground level. The pilot performed a 180-degree turn and attempted to land in the opposite direction of takeoff, on runway 4; however, the engine lost all power during final approach, and the airplane impacted a fence about 150 feet prior to the runway. During the impact, the nosegear, fuselage, and both wings were substantially damaged. Examination of the wreckage by the FAA inspector revealed fuel system continuity and fuel present to the fuel servo. That fuel was recovered from the fuel lines and no contamination was noted. The fuel strainer located at the firewall was also absent of visible contamination. The right fuel tank was compromised during impact and evidence of a fuel spill was observed near the accident site. The left fuel tank remained intact and contained fuel. When the FAA inspector moved the fuel selector from "RIGHT" to "LEFT," fuel flowed freely from the selector. Following recovery, further examination of the airplane by a mechanic did not reveal any evidence of catastrophic engine failure. The fuel servo and fuel flow divider were retained for further examination. A teardown examination of the fuel servo and fuel flow divider was performed at the manufacturer's facility, under the supervision of an FAA inspector. During the examination, small, clear, spherical particles were found on the threads of inlet fitting and plug of the fuel inlet screen. The particles were consistent with glass beads and were also found in the manual mixture control valve chamber and on the unmetered side of the fuel diaphragm. One particle was found underneath the flow divider diaphragm. The fuel servo and flow divider flowed within limits when tested after theaccident. According to the FAA principle maintenance inspector (PMI) of the company that overhauled the fuel servo, the glass beads found in the accident fuel servo were similar to the glass beads used by the company. Although the company did not bead-blast their fuel servos, bead-blasting was performed in other locations in the factory, including areas approximately 20 feet from where the fuel servos were overhauled. The PMI did not know how the glass beads entered the accident servo and did not discover any other servos with contamination; however, he planned additional inspections of the company. The recorded weather at EWN, at 1654, included clear skies and wind from 230 degrees at 8 knots.