Nepalese investigators believe an instructor captain inadvertently moved the propeller condition levers, instead of the flap selector, during an ATR 72-500’s approach to Pokhara, causing the propellers to feather and the aircraft to stall fatally.

The instructor had been supervising another captain, the flying pilot, during the Yeti Airlines service from Tribhuvan on 15 January.

After joining the downwind track for Pokhara’s runway 12, the crew extended the flaps to 15°. The captain disengaged the autopilot at about 720ft above ground and then called for a further flap extension to 30°.

But while the instructor confirmed the ‘flaps 30’ call, the flight-data recorder “did not record any flap surface movement”, says the Nepalese aircraft accident investigation commission.

The rotation speed of both propellers, however, simultaneously decreased to less than 25%, and torque started falling to zero. The inquiry says this was “consistent” with both propellers feathering, although the feather condition is not a parameter captured by the flight-data recorder.

Yeti ATR crash Pokhara-c-Nepalese aircraft accident investigation commission

Source: Nepalese aircraft accident investigation commission

Video footage captured the ATR 72 banking moments before striking the ground

Cockpit-voice recorder information picked up a master caution chime, but the crew conducted the ‘before landing’ checklist without identifying that the flaps had not extended to 30° before carrying out a left-hand turn onto the base leg.

While the throttle levers slightly advanced at this point, the inquiry states: “When propellers are in feather, they are not producing thrust.”

The ATR’s bank angle reached a maximum of 30° during the left turn.

Some 20s after the captain had called for 30° flap, the flight-data recorder registered the flaps extending to this position.

But within a few seconds, as air traffic control granted landing clearance, the captain twice remarked that the engines were not delivering power. The throttle levers were quickly advanced to their maximum position, as the aircraft initiated its final left turn at 368ft.

The inquiry says the captain handed over control to the instructor and again stated that there was no engine power.

As the ATR descended through 311ft the stick-shaker activated, and triggered again a couple of seconds later as the aircraft abruptly banked to the left. A radio-altitude alert sounded at 200ft, just before the aircraft struck the ground.

None of the 68 passengers and four crew members survived the crash.

Yeti ATR wreckage Pokhara-c-Nepalese aircraft accident investigation commission

Source: Nepalese aircraft accident investigation commission

Wreckage of the aircraft located near a gorge close to the Seti Gandaki river

The inquiry says the instructor “likely” misidentified the propeller condition levers and moved them to the ‘feather’ position after the captain had called for 30° flap.

On the ATR 72-500 the condition levers are situated immediately left of the flap lever, and right of the throttles. Although the captain had only 186h on ATRs, the instructor had around 3,300h.

“Following the unintentional feathering of both engine propellers, the flightcrew failed to identify the problem and take corrective actions despite the [crew alerting panel] cautions,” the inquiry says, adding that the captain did not crosscheck the commanded configuration change.

Pokhara airport was a new facility, and had only opened on 1 January, two weeks before the accident. The airport had not published instrument approach procedures, and was operating as a visual-only airport.

Procedures for visual approaches, however, had not been developed and Yeti Airlines had internally created a visual-circuit pattern for operating into the airport, intended to avoid both the old airport and surrounding terrain.

“This resulted in an approach that required tight turns during the descent and would result in the aircraft being at a lower altitude once aligned [with runway 12],” says the inquiry. “This did not meet the requirements for a stabilised visual approach.”

Investigators point out that this “challenging” operation to an unfamiliar airport generated high workload, and a loss of recognition by the crew of the aircraft’s state.