CEN11FA508

The pilot's improper fuel management, which resulted in fuel starvation and a partial loss of engine power while maneuvering at a low altitude.

The passenger reported that the purpose of the local flight was to take low-altitude aerial photographs of nearby ground structures and that, shortly before the accident, the pilot remarked that the engine was not operating normally and asked him to look for a suitable landing area. He believed that the engine was still running at that time. The pilot located an open landing area and was in the process of landing when the right wing collided with terrain. During the postaccident examination, about 18 gallons of usable fuel was recovered from the left wing fuel tank, and no usable fuel was recovered from the right wing fuel tank. The observed damage to the propeller blades was indicative of the engine producing at least idle power at the time of the accident. A postaccident examination of the airplane revealed no evidence of mechanical malfunctions or failures that would have precluded normal operation. Additionally, the engine demonstrated the ability to produce rated horsepower during an operational test run. A review of fueling records and recorded engine data from previous flights established that the partial loss of engine power was likely due to fuel starvation and not from a mechanical malfunction of the engine. Although the fuel selector was found positioned to draw fuel from the left wing fuel tank, it is likely that the pilot repositioned the selector from the right wing fuel tank when he detected the engine problem. Because he was maneuvering at a low altitude, the airplane likely had insufficient altitude and time to reestablish fuel flow and restore engine power.

On July 27, 2011, an autopsy was performed on the pilot at the North Dakota State Forensic Examiner's Office, located in Bismarck, North Dakota. The cause of death for the pilot was attributed to multiple blunt-force injuries sustained during the accident. The FAA Civil Aerospace Medical Institute (CAMI), located in Oklahoma City, Oklahoma, performed toxicology tests on samples obtained during the pilot's autopsy. No carbon monoxide, cyanide, or ethanol was detected in samples tested. Ibuprofen was detected in urine. Ibuprofen, sold under multiple brand names, is a nonsteroidal anti-inflammatory analgesic drug. --- Engine Operational Test Run --- The engine was sent to the manufacturer for an operational test run under the supervision of the National Transportation Safety Board (NTSB) Investigator-In-Charge. The engine was installed in a test cell and outfitted with a test club propeller. The engine started on the first attempt and idled without excessive fluctuations in engine rpm. The engine speed was increased incrementally to full throttle (2,575 rpm) over a period of 25 minutes. The engine ran at each incremental power setting for a period of 5 minutes without anomaly. The engine throttle was cycled several times between idle and maximum power settings in quick succession. The engine did not experience any hesitation, stumbling, or interruption in engine operation during the test run. The engine demonstrated the ability to produce rated horsepower during the operational test run. --- Engine Monitoring System --- The airplane's engine monitoring system, a JPI model EDM-700, was examined at the NTSB Vehicle Recorders Laboratory in Washington, D.C. The device contained about 25 hours of engine data that had been recorded over 18 power cycles. The accident flight was the last dataset recorded and contained 1,194 seconds (about 20 minutes) worth of engine data. A review of the recovered engine data revealed an average increase of 225 degrees in exhaust gas temperatures (EGT) and the turbine inlet temperature (TIT) during a 24 second period before there was significant decrease to all EGT, TIT, and engine oil temperature values. All of the engine temperature values continued to decrease until there was a power interruption to the device, which occurred about 55 seconds after the temperature values had peaked. --- Electronic Fuel Flow Indicator --- The electronic fuel flow indicator, a Shadin model Microflo-L, was examined at the NTSB Vehicle Recorders Laboratory. The fuel flow indicator did not interface with the airplane's fuel quantity indicating system. In order to properly indicate the amount of fuel available, the unit required the pilot to correctly enter or verify the actual fuel load before each engine startup. All subsequent fuel readings were based on fuel flow to the engine. The fuel flow indicator contained non-volatile memory for the amount of fuel remaining and the amount of fuel used since the last time the device had been programmed. These retained values were obtained by applying electrical power to the device and viewing the display. The values obtained for fuel-used and fuel-remaining were 3.9 gallons and 70.0 gallons, respectively. Fueling records established that the airplane had not been refueled before the accident flight. Additionally, the airplane fuel system did not contain 70 gallons of fuel at the accident site. The fuel flow indicator was equipped with a toggle-switch that upon deflection would automatically reset the fuel quantity to a "FULL" indication. The retained value for fuel-used (3.9 gallons) was consistent with the amount of fuel that was likely consumed during the accident flight. --- Fuel Consumption Rate Calculations --- The accident airplane had an 80 gallon fuel capacity (74 gallons usable) that was distributed evenly between two 40 gallon wing fuel tanks. During the on-scene investigation, 18 gallons of useable fuel was defueled from the left wing tank. The right wing tank was void of useable fuel. According to available flight tracking and engine monitoring system data, the airplane engine had been operating at least 3.4 hours since the last refueling. Considering the useable fuel recovered from the airplane, the average fuel consumption since the last refueling was calculated to be about 16.5 gallons per hour (gph). A review of the previous 18 flight legs established that the airplane had an average fuel consumption rate of about 17.4 gph. Additionally, the airplane's co-owner further corroborated that the airplane's average fuel consumption rate was about 17.4 gph. According to engine performance data, the expected fuel consumption rate during takeoff and cruise-climb was 35 gph. The expected fuel consumption rate during cruise flight, at a lean-of-peak mixture setting, could vary between 13.5 and 17.0 gph depending on the selected engine rpm. The nearest aviation weather reporting station was located at the Sloulin Field International Airport (ISN), Williston, North Dakota, about 19.5 miles west of the accident site. At 2052, the ISN weather observing system reported: wind 110 degrees at 10 knots, 10 miles visibility, clear sky conditions, temperature 26 degrees Celsius, dew point 14 degrees Celsius, and an altimeter setting of 29.84 inches-of-mercury. A postaccident investigation confirmed that all airframe structural and flight control components were located at the accident site. The accident site consisted of open, rolling terrain. The wreckage debris path began about 109 feet west of the main wreckage and consisted of navigational lens cover fragments. There was a large ground depression located about 20 feet from the initial debris. The main wreckage consisted of the entire airframe, flight controls, engine, and propeller. The main wreckage was facing a heading of south. The topography of the accident site and orientation of the wreckage resulted in the right wing being higher than the left wing. Flight control continuity was established between the individual flight control surfaces and their respective cockpit controls. The flaps and landing gear were fully retracted. The throttle, propeller, and mixture controls were found in the full forward position. Mechanical continuity was confirmed from the engine components to their respective cockpit engine controls. The magneto-ignition switch was found in the LEFT position. The electrical fuel pump was found in the OFF position. The fuel selector valve was positioned to draw fuel from the left wing fuel tank. About 21 gallons of fuel (18 gallons useable) was recovered from the left wing fuel tank. The left wing tank appeared undamaged with no visible fuel leaks. The right wing tank contained minimal fuel, estimated to be about 1 gallon, all of which was found in the fuel tank's inboard baffle compartment. The integrity of the right wing fuel tank had been breached by several punctures that were located about midspan. The fuselage fuel-sump strainer contained a liquid consistent in appearance and odor with 100-low lead aviation fuel that was void of water and particulate contamination. The engine remained partially attached to the firewall and the propeller remained attached to the crankshaft flange. Internal engine and valve train continuity was confirmed as the engine crankshaft was rotated. Compression and suction were noted on all cylinders in conjunction with crankshaft rotation. The upper spark plugs were removed and exhibited features consistent with normal engine operation. Both magnetos provided spark on all leads when rotated. The throttle and fuel control assembly had separated from the engine during the accident. The intake ducting from the air filter to the throttle control was not obstructed by debris. The engine-driven fuel pump functioned when rotated. The fuel supply line to the fuel manifold valve contained a liquid consistent in appearance and odor with 100-low lead aviation fuel. The turbine, compr