Accident Overview
On July 30, 2021, at approximately 1423 Pacific daylight time, a Grumman G-164B airplane, registered as N3629F, was substantially damaged during an accident near Tracy, California. The airplane was being operated as a Title 14 Code of Federal Regulations Part 137 aerial application flight. The pilot reported no injuries.
Pilot Actions
The pilot was on the eighth pass over a hay field located beneath high-tension power lines. The flight path had previously taken the airplane south below the power lines, which ran diagonally across the field from northeast to southwest. Midway through the spray pass, the pilot experienced a total loss of engine power accompanied by a vibration, though he could not recall whether the vibration occurred before or after the power loss. Airspeed was about 130 mph at the time of power loss but decayed rapidly. Unable to climb due to overhead power lines, the pilot flew about 10–15 ft above ground level under another set of power lines at the south end of the field to an adjacent field. He began a shallow right turn to orient the airplane parallel to raised dirt beds, but during the turn the airplane contacted the ground and several trees, causing substantial damage to the wings and fuselage.
In a subsequent statement, the pilot noted that he was likely running the engine at 70–80% of full power as he was nearing the end of the load and did not require a high power setting. He could not recall the exact position of the power lever.
Postaccident Examination
Postaccident examination of the engine revealed that the rotating group rotated with some resistance but was free to rotate after removal of the planetary gear assembly from the nose cone assembly. Continuity of the direct drive fuel control drive train was confirmed from the fuel control drive shaft to the power shaft. Rotational scoring was observed in several locations, including the propeller shaft, first and second stage compressor impellers, and the entire turbine rotating group. Metal spray deposits from the compressor shroud were found on the first, second, and third stage turbine rotor blades and vanes.
Engine Components
Testing of the fuel control unit showed it operated outside the manufacturer’s specifications in several areas, including fuel flow, flight idle setting, and various test points. However, according to the manufacturer, none of these deviations would have been contributing factors in a loss of power event. The fuel pump tested without anomalies. The propeller governor exhibited maximum and minimum rpm beyond normal limitations, and arm travel range was out of limits; the manufacturer indicated the test data reflected typical field adjustments and that the unit functioned properly at the time of the accident.
Fuel System Findings
The engine was equipped with a fuel solenoid valve with a manual override to control fuel delivery. Examination of the valve showed that the scribe mark on the override shaft was at the top, consistent with an OPEN position, but low-pressure air applied to the fuel inlet port did not pass through, consistent with the ball valve being in the CLOSED position. According to the manufacturer, a closed ball valve would prevent fuel flow to the engine nozzles. However, the closed position is inconsistent with metal spray in the turbine section, which indicates the engine was rotating and fuel was delivered at impact. The manufacturer stated that these opposing findings can only exist if the fuel shutoff valve closed at impact due to pilot intervention, an electrical short, or inertial movement during impact. Removal of the beta tube required 18–18.5 turns, consistent with normal alignment.
Propeller Examination
The propeller manufacturer noted no visible discrepancies that would have degraded normal operation before impact. Blades remained in their clamps, and all three pilot tubes were intact. Blade No. 2 exhibited S-bend characteristics and an approximately 3-inch tip fracture (not recovered). Blades No. 2 and 3 showed bending opposite the direction of rotation and twisting toward low pitch. All three blades had chordwise/rotational scoring on the camber side near the tips. Maintenance records indicated the propeller’s most recent overhaul in 2012, with blade angles set at 30 inches radius, consistent with published guidance. Lock brackets were fractured, and start lock plates showed signatures consistent with forceful impact. The counterweight puncture in the spinner dome was consistent with a low blade angle when the spinner impacted terrain. Only blade No. 2 would intermittently lock near the start lock angle when actuated manually. Piston marks corresponded to blade angles of approximately -1°, 15.6°, and -1° to 0.7°, the latter near the start lock angle.