North American T-28B N787AS accident near Fallbrook, California

Probable cause

A loss of power due to an unmaintained carburetor diaphragm and inoperative electric fuel pump that resulted in fuel starvation to the engine.

NTSB analysis

The pilot flew the airplane earlier in the day and made a stop at the accident airport. He performed a pre-flight inspection and positioned himself in the front seat, with the pilot-rated passenger in the rear seat. After takeoff, with the airplane about 200 feet above ground level (agl), the engine sustained a loss of power. The airplane collided with a plant nursery.

The fuel system was designed where fuel flowed by gravity from all internal cells into the sump tank where an electric fuel boost pump was located. The airplane was also equipped with an engine-driven fuel pump. When the pilot turned the fuel “ON”, the electric fuel pump would simultaneously be turned on. An electric fuel-boost-pump test switch was in the cockpit and, as part of the start-up checklist, the switch had to be activated to momentarily interrupt power to the electric pump, allowing the pilot to confirm that the engine-driven fuel pump pressure is adequate (and that the electric fuel pump is operational). There was a 20-amp circuit breaker in series with the switch and the pump. If the electric pump fails below 10,000 feet pressure altitude, fuel drawn by the engine-driven pump is designed to bypass through the electric pump and sustain approximately normal fuel flow to the carburetor (with a slight drop in fuel pressure indication that may be noted).

Disassembly of the electric fuel pump revealed that metal shavings were lodged in the armature creating a short in the system. The blades on the impeller showed wear consistent with instability of the shaft during rotation. The upper commutator was cracked and showed wear on the upper portion. The pump-end bearing was worn on the outside with the labyrinth seal, washer, and shims deformed. The bearing cage was determined to be the metal pieces shorting out the armature.

Continuity was established from the electric pump test switch in the cockpit to the electric pump and to the circuit breaker. Attempts to trip the breaker were unsuccessful and further testing of the unit revealed it was functional. It could not be determined why the breaker did not trip when the electric pump shorted, but if the pilot had used the test switch, he likely would have been able to see the pump had failed.

A flow test of the carburetor revealed that numerous parameters were out of limits. At low power settings, the carburetor ran rich (more fuel flow than required for normal operation), and at high power settings the carburetor ran lean (less fuel flow than required for normal operation). Disassembly revealed that the enrichment valve’s diaphragm was stiff/rigid, consistent with it not being submerged in fuel for long durations. The carburetor manufacturer recommended that it should be overhauled at least every ten years and be pressurized regularly (if in a hot and dry climate, it should be done monthly). The internal diaphragms become brittle and can fail if they are not wetted with fuel regularly. The carburetor was overhauled over 12 years before the accident, equating to about 400 hours of flight time; during that time the pilot stated he had not completed any maintenance on it.

At takeoff power, the carburetor was not able to provide enough fuel to the engine because the enrichment diaphragm was brittle from inactivity. Additionally, because the electric fuel pump was inoperative, the fuel flow pressure was diminished providing less fuel to the carburetor.

Text from the NTSB. Open the NTSB docket for WPR22FA265 for the full record.