NTSB Factual Report

HISTORY OF FLIGHT

On February 17, 2008, at 1638 Pacific standard time, an experimental Wilson Vans RV-7, N331KM, experienced a loss of engine power and collided with trees approximately 1.7 miles northwest of Nevada City, California. The private pilot operated the airplane under the provisions of Title 14 Code of Federal Regulations Part 91. The pilot was killed, and the airplane was substantially damaged. Visual meteorological conditions prevailed, and no flight plan had been filed. The personal flight originated at Nevada County Air Park, Grass Valley, California, approximately 1610.

At 1637, the pilot of N6375Q was in communications with the accident pilot and made a mayday distress call on guard (121.5 MHz) for the pilot. The pilot of the accident airplane had made a distress call on the Nevada County Air Park common traffic advisory frequency (ctaf) stating that the engine had quit and he could not get it restarted.

Another pilot who was airborne at 5,500 feet msl (mean sea level) at the time of the accident said that he had radio communications with the pilot of the accident airplane. The accident pilot stated that he could not get the engine to restart, that he had tried everything, and he was going down. The pilot then observed the accident airplane in a controlled descent into an area populated by large trees.

The elevation of Nevada County Air Park, Grass Valley, California, is 3,152-feet msl.

The airplane wreckage was located in a wooded area. It rested inverted against the ground, with wing leading edge damage, and broken pine tree branches in the vicinity. One witness and two sheriff's deputies reported the distinct smell of fuel around the wreckage. Photographs taken of the airplanes instrument panel, immediately after the accident, show the fuel pump switch in the 'on' position.

PERSONNEL INFORMATION

A review of the Federal Aviation Administration airman records revealed that the 57-year-old pilot held a private pilot certificate dated May 9, 2005, and a third-class medical certificate dated August 2007. The pilot's logbook recorded that he had 871.6 hours in single engine airplanes as of February 14, 2008. He had logged 42.2 hours in RV-6A's and RV-7's since November 8, 2007, and 27.3 hours were in the accident airplane.

AIRCRAFT INFORMATION

The two seat, low wing, fixed landing gear, experimental category airplane, serial number 71967, was constructed by the pilot. It was powered by an experimental Xtreem 360 engine, model IO-360-B1AA4, serial number G06933. It was equipped with a Hartzell constant speed propeller, model F7496-2. The engine was installed on the airplane and the first test run dated in the engine logbook was November 12, 2007. Review of copies of the airframe logbook show that the aircraft engine was tested again on December 10, 2007. On December 13 an FAA Aviation Safety Inspector issued the airplane a Special Airworthiness Certificate.

A November 13, 2007, engine logbook entry stated the engine was test run for 0.5 hours. There was "no fuel pressure indication, but all other parameters checked out." On December 10, the engine was test run to full power, achieved 26 inches of manifold pressure at 2,600 rpm, and "all parameters checked ok engine ready for flight." On January 4, 2008, at total engine time of 13.0 hours, the engine driven fuel pump was removed and replaced. During the wreckage examination the hobbs hour-meter displayed 29.1 in the instrument window.

A colleague of the pilot, who was also building an RV-7, stated that early in January the pilot had discussed with him concerns about a fuel leak around the engine driven fuel pump on the accident airplane. The pilot removed and replaced the engine driven fuel pump but could not get the engine to run afterwards. The pilot enlisted the help of another friend, who was a mechanic, to assist him in installing the fuel pump, which was successful.

The fuel pump was a reciprocating type that is operated by riding on a cam lobe off of the engine cam shaft. As the engine rotates, the push rod extends and retracts, normally acting against the fuel pump lever arm. To install the fuel pump, the push rod should be in its retracted position to allow the fuel pump lever arm to be placed on to the engine accessory section without difficulty. The manufacturer of the fuel pump stated that if during the fuel pump installation, the push rod is not aligned under the fuel pump lever arm, but instead is beside an extended push rod, when the bolts are torqued down the lever arm bends or cracks. Usually in this case the engine will not run. The pump is removed and the lever arm may be at a slight angle or the lever arm may be cracked. The crack may not be visually detectable because it may be somewhat hidden inside the pump casing.

MEDICAL AND PATHOLOGICAL INFORMATION

An autopsy was performed on the pilot February 20, 2008. The report listed the cause of death as a result of multiple blunt force trauma injuries.

Forensic toxicology was performed on specimens from the pilot by the FAA Forensic Toxicology Team CAMI, Oklahoma City, Oklahoma. The toxicology report stated that the specimens tested negative for carbon monoxide, cyanide, ethanol, and listed drugs. Quinine and ibuprofen were detected in blood.

TESTS AND RESEARCH

On March 27, 2008, the Safety Board investigator examined the airplane and engine at the Plain Parts facility in Pleasant Grove, California. The number 2 cylinder fuel injection line was observed in a slightly crimped condition, and both leads for the number 1 cylinder had been severed. Four quarts of oil remained in the engine and was measured using the dipstick. All four top were removed, examined, and observed to exhibit a whitish-gray color on the electrode elements, all gaps were similar, and no mechanical damage was identified. The engine was rotated by hand and thumb compression was achieved on all four cylinders. The throttle, propeller condition, and mixture controls were observed to be connected and moved without binding. Electrical power was applied to the airplane allowing the electric fuel boost pump to be energized by manipulating the boost pump switch on the instrument panel. All the electrical switches on the instrument panel had been damaged, but could function if held in the ‘on' position. Fuel from an external fuel can was plumbed into the fuel selector valve. The boost pump was momentarily engaged and the fuel pressure gage on the instrument panel indicated over 30 psi. The engine was successfully started using the airplane's ignition switch and external power but would stop running after a few seconds. Additional starts were attempted. The engine was successfully started and ran continuously, at approximately 1,100 rpm, as long as the fuel boost pump was held in the on position.

The engine driven fuel pump was removed from the engine. A visual examination of the fuel pump revealed that the pump arm that drives the reciprocating motion of the pump was not aligned squarely with the centerline of the pump. The pump was disassembled and a crack was identified on the pump arm below the pivot hole where the arm is attached to the pump casing. The lever arm pad face, where the push rod contacts the lever arm, exhibited areas of wear on the center of the pad and to the left of the pad face. The fuel pump components were sent to the Safety Board's Materials Laboratory for examination.

A bench binocular microscope examination of the lever arm revealed the exposed crack face contained fine granular features all around the edges of the lever arm wall consistent with a hardened case. The core of the lever arm wall showed smooth textured fracture features. The crack extended through approximately 75 percent of the lever arm cross section. The lever arm was reinserted into the pump casting into a position that simulated the condition of an installed assembly. In the installed position, the arm showed evidence of lateral bending deformation, in a direction consistent with the gaping crack at the hole for the attachment pin. This lateral displacement measured approximately 0.25 inch.

The lever arm was again removed from the upper casting, and the crack face at the engine cam end of the lever arm was excised. Scanning electron microscope examination of the excised fracture face revealed striation features typical of fatigue cracking that emanated below the surface of the hardened case at the edge of the channel. The fatigue crack origin area was located approximately 0.3 inches from the hole for the attachment pin and was located longitudinally in line with the position of the hole. The edges of the wall for the lever arm showed granular features that were typical of a hardened case, and the areas adjacent to and below the hardened case showed a band of ductile dimple features. The fatigue crack originated from the ductile dimple band region. The fracture features at the core showed mechanical damage consistent with relative movement between mating faces of the crack. The core portion of the fracture face outside of the fatigue crack region and the mechanical damage region showed ductile dimple features typical of over stress separation.

The complete Metallurgical Laboratory Factual Report is located in the official docket of this investigation.

ADDITIONAL INFORMATION

The Pilot's Handbook of Aeronautical Knowledge (FAA-H-8083-25) describes basic principles and design elements of fuel injection systems. Listed as disadvantages of fuel injection are difficulty in starting a hot engine, and problems associated with restarting an engine that quits because of fuel starvation.

NTSB Probable Cause

The improper installation of the fuel pump, leading to a fatigue failure of the pump's lever arm and subsequent loss of engine power. Contributing to the accident was the lack of suitable terrain for a forced landing.