On September 3, 2011, at 0847 eastern daylight time, an experimental amateur-built Vans RV-6A, N241BD, was substantially damaged during a takeoff attempt at North Perry Airport (HWO), Hollywood, Florida. The airline transport pilot and the non-pilot owner sustained minor injuries. Visual meteorological conditions prevailed, and no flight plan had been filed for the flight to Palatka Municipal Airport (28J), Palatka, Florida. The personal flight was conducted under the provisions of 14 Code of Federal Regulations Part 91.

According to the pilot, preflight and before takeoff checks were all “satisfactory,” and he was subsequently cleared to take off from the 3,350-foot runway 36L. Initial lift off was "normal" at 60 knots. A few seconds later, when the airplane was about 30 to 50 feet above the ground at 70 to 75 knots, the airplane abruptly lost about 25 to 50 percent of its thrust. The pilot decided to abort the takeoff and closed the throttle. The airplane touched down about 500 feet from the end of the runway, then continued through the overrun and impacted the airport perimeter fence.

The pilot subsequently made a video playback of the airplane’s electronic flight instrument system (EFIS) display during the takeoff attempt. During the video, and after what appeared to be an addition of power, the propeller speed settled about 2,540 rpm while the manifold pressure settled at 29.3 inches.

When the airspeed increased to 74 knots, the rpm suddenly dropped significantly, followed shortly thereafter by a reduction in manifold pressure.

The airplane’s airspeed continued to climb to 83 knots before beginning to diminish. The EFIS attitude display subsequently began to shake, consistent with the airplane traveling over rough ground, until it stopped completely when the airspeed reduced to 41 knots.

A Federal Aviation Administration examination of the airplane following the accident did find any engine anomalies. The examination of the engine oil filter did not result in finding metal shavings or other contaminants.

The propeller governor was later disassembled and examined at a propeller shop with the following observed by shop personnel: “The pilot valve had to be forced out of the drive gear due to scoring on [the] pilot valve that seems to have been caused by metal ingestion.” In addition, scoring was found on the governor’s internal gear housing walls, as well as on “drive gear friction surfaces.” The governor was not reassembled.

The propeller governor components were subsequently examined at the NTSB Materials Laboratory. According to the Laboratory's Factual Report (located with photographs in the Public Docket for this accident), the components arrived already cleaned.

The report also noted that the governor was driven by the engine and contained an integral gear pump that pressurized engine oil and directed the oil to and from a variable pitch propeller via ports in a gear shaft. An examination of the components revealed scoring in the pump chamber and on the sides of the driven gear, and chipping on the pilot valve. Examinations of the drive gear shaft revealed circumferential scoring on the shaft portion, and indentations on the crest of the gear teeth and at the intersection of the crest with the gear flank. Bands of circumferential scratches were also located adjacent to the governor and propeller ports on the shaft portion of the gear shaft. A similar band of scratches was observed on the shaft between the gear and the sump port. Further examination of the pilot valve hole in the drive gear shaft revealed that the edge of the governor port nearest to the gear was slightly rounded and that a circumferential band of scratches was located adjacent to that edge.

The pilot valve had two control lands (surfaces) identified for the examination as “A” and “B”. Normally, with the propeller at an on-speed condition, land “A” would have covered the governor port in the gear shaft and land “B” would have covered the sump port so no oil flowed to or from the propeller. At an overspeed condition, land “A” would have uncovered the governor port and land while “B” still covered the sump port, allowing pressurized engine oil to enter through the governor port and out of the propeller port to the propeller. The propeller pitch would have been increased and the propeller speed would have reduced. At an underspeed condition, land “A” would have covered the governor port and land “B” would have uncovered the sump port, allowing oil from the propeller to flow through the sump port to the engine sump. The propeller pitch would have reduced and the propeller speed would have increased.

Examination of the pilot valve revealed chipping at the edge of land “A” and a circumferentially-oriented scratch that originated at the chipped edge. There were also bands of circumferential scratches located adjacent to the governor and propeller ports on the shaft portion of the gear shaft. A similar band of scratches was observed on the shaft between the gear and the sump port.

The pilot valve hole in the drive gear shaft was examined to reveal that the edge of the governor port nearest to the gear was slightly rounded and a circumferential band of scratches was located adjacent to that edge. Examination of the cap, the flyweight, and the bearing revealed no anomalies and no indications of missing material.