N816FF

On July 18, 2017, about 1710 central daylight time, an Air Tractor AT-802A airplane, N816FF, was substantially damaged in an accident near Reyno, Arkansas. The commercial pilot was not injured. The airplane was operated as a Title 14 Code of Federal Regulations Part 135 aerial-application flight.

The pilot reported that, after an uneventful spray flight, he was maneuvering to land on private airstrip when he noticed an increase in engine torque shortly after he made a power reduction. The pilot recalled that the engine torque increased to about 6,000 ft-lbs and the propeller speed decreased to about 710 rpm. The pilot stated that he “looked at my levers and pushed the prop lever all the way forward.” The pilot stated that there was insufficient thrust to maintain altitude and that he made a forced landing in an open soybean field. The airplane’s left wing impacted the ground during landing and the airplane skidded sideways to a stop. The right wingtip, right aileron, and right elevator were substantially damaged.

Examination of the airplane at the accident site confirmed flight control continuity. The power lever was at idle, the propeller lever was in feather, and the start control (condition lever) was in cut-off. Engine control continuity was confirmed from the cockpit levers to their respective engine components, and the engine controls moved smoothly through their full range of motion. The bleed air reference line between the fuel control unit (FCU) and the constant speed unit (CSU) was properly installed and secured. A fuel sample from the header tank did not exhibit any contamination and was consistent with Jet-A aviation fuel. Both wing tanks contained fuel, and the engine monitor indicated that there were 55.5 gallons and 142 gallons remaining in the left and right fuel tanks, respectively. The airframe fuel lines were intact and did not exhibit any leaks. The FCU filter housing was removed, and it contained fuel. The propeller remained attached to the engine and appeared to be undamaged. The propeller beta system was found properly rigged; the beta plunger valve was flush with the cap nut, and the beta block was properly seated under the guide pin in the propeller beta ring.

The engine was removed from the airplane and sent to the manufacturer for additional examination. A teardown inspection of the engine did not reveal any evidence of mechanical malfunction or failure that would have precluded normal engine operation. The clearance between the propeller shaft and the propeller sleeve met the manufacturer’s design limits and the dimensions between the propeller shaft and the internal oil transfer tube/oil jet were within specifications. Bench testing of the engine-driven fuel pump, flow divider, FCU, compressor bleed valve, CSU (propeller governor), and overspeed governor did not reveal any anomalies that would have precluded normal engine operation.

An analysis of the airplane’s engine monitor data (see figure 1) revealed that toward the end of the flight, between 1708:44 to 1709:19, the propeller speed (Np) decreased at three separate rates, consistent with the cockpit propeller lever being moved aft at three different rates. The aft movement of the propeller control increased propeller pitch which, in turn, decreased Np. The data analysis also indicated that as Np decreased, there was a corresponding increase in engine torque. Between 1708:45 and 1708:58, the Np decreased from 92% to 84% and there was a 10% decrease in thrust. Between 1708:58 and 1709:14, the Np continued to decrease from 84% to 65% and there was another 10% decrease in thrust.

At 1709:14, the Np had decreased to 62% (1,060 rpm) at which point the CSU feather valve dumped oil pressure from the propeller causing it to feather. Postaccident bench testing of the CSU confirmed that the feather valve would dump oil pressure at 1,060 rpm. When the propeller went into feather, the engine torque rapidly increased above the 5,100 ft-lbs limitation. At 1709:25, there was a rapid increase in fuel flow and engine torque, consistent with the pilot moving the power lever full forward with the propeller still feathered. At 1709:28, the engine torque reached 6,920 ft-lbs. The gas generator speed (Ng) and fuel flow remained relatively constant and the engine torque remained above 6,600 ft-lbs until the engine was shut down after the accident.

Figure 1 – Engine Monitor Data