Back to Search

NTSB investigation record

ERA17FA327

Completed

Mooney M20C· N53CP

Date
September 16, 2017
Location
North Branford, CT
Conditions
VMC
Record
Published September 25, 2020

Primary finding

Probable cause

A total loss of engine power due to fuel starvation as the result of foreign object debris in the fuel selector valve. Contributing to the accident was the pilot's selection of a low cruising altitude, which reduced the available time to troubleshoot the loss of engine power and afforded fewer forced landing site options, and improper maintenance of the airplane, which allowed a portion of a shop towel into the fuel system.

Investigator assessment

Analysis narrative

The airline transport pilot, who was the owner of the airplane, and one passenger departed on a day visual flight rules cross-country flight. The airplane came to rest in a wooded area near an open field about 24 miles from the departure airport. The pilot was not in contact with air traffic control during the flight. Review of radar information revealed radar targets that were coincident with the accident flight on a south-southeast track at altitudes between 900 and 1,300 ft above ground level until radar contact was lost about 1 mile northwest of the accident site. Several individuals near the accident site reported that they heard the sound of the impact, but there were no witnesses to the accident. The propeller exhibited signatures consistent with a lack of engine power at the time of impact. The fuel selector was found in the left tank position and the landing gear was extended. There was evidence of fuel in both tanks at the accident site. Examination of the fuel system revealed that air would not pass through the fuel selector valve with the valve selected to the left fuel tank position. The handle was operated by hand and could be moved normally between the settings. Air passed freely through the valve when selected to the right tank position. Disassembly of the fuel selector revealed a piece of red, fibrous material consistent with a shop towel that likely inhibited fuel flow to the engine and resulted in fuel starvation and a total loss of engine power. The airplane's maintenance logs were not found and when the shop towel debris may have been introduced to the fuel system could not be determined. Additionally, a homemade tool constructed of PVC pipe and connection fittings was found in the wreckage that appeared to be designed to manipulate the fuel selector; however, the reason for its fabrication and use during the accident flight could not be determined. The device was broken at its handle. Following the loss of engine power, the pilot may have attempted to switch the fuel selector from the left tank to the right tank and was unable to do so, either due to a failure of his homemade tool or to the inadequate time afforded to troubleshoot the loss of engine power due to his selection of a low cruising altitude, or a combination of the two factors. The airplane's low cruising altitude also reduced the pilot's available forced landing site options after the engine lost power. It is likely that the pilot was attempting to reach an open field that was about 1,500 ft beyond the accident site and had lowered the landing gear in preparation for landing, but due to the airplane's low altitude, it was unable to reach the field and impacted trees. The pilot's medical certificate was denied nearly 10 years before the accident and never re-issued. Autopsy identified severe cardiac disease, which placed the pilot at risk for sudden symptoms such as chest pain, shortness of breath, palpitations, or fainting; however, it is not likely that this condition contributed to the accident.

Source record

Factual narrative

At 1653, the weather conditions reported at Tweed-New Haven Airport (HVN), New Haven, Connecticut, located 9 miles southwest of the accident site, included variable wind at 3 knots, 10 statute miles visibility, broken clouds at 1,400 ft above ground level, temperature 24°C, dew point 19°C, and an altimeter setting of 30.16 inches of mercury. According to Federal Aviation Administration (FAA) airman records, the pilot, age 81, held an airline transport pilot certificate with a rating for airplane multi-engine land and commercial privileges for airplane single-engine land and sea. He also held a flight instructor certificate with ratings for airplane single-engine and instrument airplane, a flight engineer certificate with a rating for turbojet-powered airplanes, and a mechanic certificate with airframe and powerplant ratings. The pilot's most recent application for an FAA second-class medical certificate was dated October 16, 2006. On that date, he reported 31,300 total hours of flight experience. Examination of the accident site revealed that the airplane's first point of impact was in 75-ft-tall pine trees in a nose-down attitude before coming to rest against trees in a nose-down position on its right side. The wreckage path was 175 ft long and was oriented on a heading of about 010° magnetic. There were an open field about 1,500 ft north of the accident site. The right wing separated from the fuselage at the wing root during impact and was the first piece of wreckage discovered at the start of the wreckage path. Three feet of the outboard left wing was found 75 ft north of the right wing and was wrapped around a tree. The fuselage, left wing, and tail assembly remained together at the main wreckage site, where they came to rest against a tree. The landing gear was extended, and the landing gear selector was in the "DOWN" position. The wing flaps were in the retracted position. The primary flight instruments on the pilot's (left) side remained intact. The magneto switch was found in the "BOTH" position. The airplane was not equipped with any instruments that contained non-volatile memory. The right wing fuel tank was breached during the impact sequence and evidence of fuel was found on the trees and vegetation near the initial impact point. The left wing fuel tank contained about 7.5 gallons of fuel. The fuel selector was inaccessible due to cockpit crushing and floor buckling, but a visual examination through the firewall indicated that it was in the left tank position. The airframe and engine were removed from the site to facilitate further examination. The propeller remained attached to the crankshaft flange. The spinner was partially crushed on one side. For the examination and visual reference, the three propeller blades were labeled A, B, and C. Blade A was bent aft about 30° about 6 inches outboard of the hub and could be rotated in the hub by hand. Blades B and C appeared straight and undamaged, with no rotational scoring, s-bending, or chordwise scratching. The propeller governor was impact-damaged and partially separated from the engine. The governor control cable remained attached to the governor control arm but was impact damaged. The governor oil screen was absent of debris. Continuity of the crankshaft to the rear gears and to the valve train was confirmed and each cylinder produced suction and compression. The interiors of the cylinders were examined using a lighted borescope and no anomalies were noted. The No. 1 cylinder was removed to facilitate inspection of the engine crankcase. No anomalies were noted to the crankcase interior components or to the No. 1 cylinder, piston or valves. Oil was present in the engine and the pistons, valves, and crankshaft appeared lubricated. Both left and right magnetos were undamaged and no anomalies were noted. Both magnetos produced sparks at regular intervals when rotated by an electric drill. The ignition harness remained attached to the magnetos, and the leads remained secured by their terminal ends to their respective spark plugs. The spark plugs remained secured to their respective cylinders. The top spark plugs were removed and examined. They all displayed little wear and no evidence of carbon or lead fouling in accordance with the Champion Check-A-Plug chart. The vacuum pump remained attached to the engine and no damage was noted. The pump was removed and partially disassembled. The composite drive assembly, carbon rotor, and carbon vanes were intact. The alternator remained attached to the engine and was undamaged. The alternator was rotated easily by hand and the drive belt was in place and unbroken. The fuel injector servo was fractured across the throttle bore and separated from the engine oil sump. The throttle and mixture control cables remained attached to their respective servo control arms. The control cables and associated brackets were impact-damaged and the positions of the controls could not be determined. The induction air box and air filter were present and impact-damaged but did not exhibit any preimpact anomalies. The fuel injector servo was partially disassembled and no damage to the rubber diaphragms or other internal components was noted. The servo fuel inlet screen was absent of debris. The fuel flow divider remained attached to the engine and no damage was noted. The flow divider was partially disassembled, and no damage was noted to the rubber diaphragms or other internal components. No debris was noted inside the flow divider. The fuel nozzle lines and the two-piece nozzles remained in place and were unobstructed. The engine-driven fuel pump remained attached to the engine. No damage was noted and it operated normally when actuated by hand. The pump was partially disassembled, and no damage was noted. Liquid with an odor consistent with aviation fuel was observed in the engine-driven fuel pump, the hose from the pump to the servo, in the servo, and in the fuel selector. The fuel selector valve was removed from the airframe and air pressure applied to the valve fuel outlet port. Air did not pass through the selector valve when the handle was in the position marked "LEFT." The handle was moved to the "OFF" position, then back to the "LEFT" position, and it remained blocked. Air did not pass through the valve when the actuator handle was placed in the position marked "OFF" or in the rearward, unmarked position. Air passed freely when the handle was placed in the position marked "RIGHT." When the handle was returned to the position marked "LEFT," no air passed through the selector valve. The selector handle moved normally with no unusual resistance between the settings. The valve was disassembled and a spongy mass of reddish fibers consistent in appearance with red cotton shop towel fibers were observed in the selector cavity (see figure 1). The rounded mass was about 5/8 inches in length and about 3/8 inches in width. Fibers also covered about 5% of the fuel drain screen. Figure 1-Debris discovered in the fuel selector. A section of PVC similar to plumbing or electrical conduit was discovered in the wreckage (see figure 2). It was made up of five individually-threaded, male-to-female connections which, when threaded together, measured about 9 inches long. On the top of the device was a PVC pipe in the shape of a handle. The entire device was in three separate pieces when discovered; the top of the t-handle was broken from the device and the bottom section was unscrewed. On each side of the handle was a label indicating "LEFT" and "RIGHT." The top of the handle was labeled "FUEL." On the bottom of the T-handle connection, the vertical pipe appeared to be hand carved/shaved so that it would fit into the top section of the devise There was a 3/4-inch notch cut out on the bottom of the device. When the device was reassembled during the examination, it fit into the airplane fuel selector handle, and appe

Continue research

Find similar accidents

Continue with the strongest shared characteristics.