WPR23FA083 · Piper Aircraft Inc PA-34-220T

The pilot departed on a cross-country visual flight rules (VFR) flight on a route with which he was familiar. On the day of the accident, visual meteorological conditions (VMC) prevailed along most of the route; however, the mountain pass just north of the departure airport was obscured by low clouds and light rain. The pilot had obtained weather information before departure that would have reflected the instrument meteorological conditions (IMC), and he would have been able to see the mountain obscuration from the departure airport.

Recorded data from onboard avionics revealed that, during previous flights on the accident route, the pilot typically departed and flew direct toward the destination through the mountain pass north of the departure airport, engaging the autopilot shortly after takeoff. The accident takeoff and departure were uneventful, but rather than taking the northern route through the pass as usual, the pilot chose an initial heading to the northeast, toward rising terrain. It is likely that the airplane entered IMC at this point and continued in IMC for the remainder of the flight. The airplane descended as it approached the terrain, and the pilot appeared to make corrective control inputs via the autopilot using the heading mode (HDG) function to turn left back toward the pass. The airplane responded appropriately, and once the airplane had reached the entrance of the pass, the pilot provided a series of small heading corrections toward the north.

The airplane then continued on a track similar to previous flights, but veered to the left of the pass, then continued the left turn toward rising terrain. Recorded data indicated that, around this time, power was applied to both engines and the airplane started to climb, likely as a result of the terrain awareness system alerting the pilot to the rising terrain ahead. The airplane then began a series of extreme pitch and bank excursions while climbing and descending in a left turn. The airplane ultimately impacted the ground in a steep left wing- and nose-low attitude.

Based on the recorded basic autopilot mode parameters, which included commanded (but not selected) altitude and bank values, the directional commands the pilot issued the autopilot could not be definitively determined; however, it is likely that, based on his use of HDG mode, he was guiding the airplane incrementally through the pass in an effort to avoid terrain and/or inclement weather. The pilot was likely attempting to negotiate the surrounding terrain by reference to the terrain awareness features of the airplane’s integrated flight display (IFD). This feature was not designed to be used as the sole source of navigation.

Review of the data during the final stages of the flight indicated that the pilot was most likely disoriented and fighting against the autopilot. The autopilot was equipped with a return-to-level (LVL) safety feature, which can be activated by the pilot at any time. The LVL system works to return the airplane to a straight-and-level flight attitude. About 40 seconds before ground impact, the autopilot was turned off, and LVL mode was activated, likely by the pilot. The autopilot immediately issued commands to regain control, and although it appeared to respond, the descent and roll continued, possibly due to the pilot physically overpowering the autopilot’s control inputs via the control column.

Examination of the flight control system did not reveal any anomalies that would have precluded normal operation. The recorded data indicated that both engine power levels mirrored each other during the entire flight, and damage signatures to the engine and propellers were consistent with them both producing power at the time of impact. Damage to the airplane prevented a complete assessment of the operational status of the IFD system; however, the recorded data confirmed that the autopilot was being used throughout the accident flight. The airplane was also equipped with a set of independent backup flight instruments; therefore, the pilot had the capability to disengage the autopilot and fly the airplane by reference to those instruments if needed; however, it could not be determined if he had practiced such procedures before or was current and capable of hand-flying the airplane in IMC.

The weather conditions at the accident site were conducive to the accretion of airframe icing, and although the airplane was equipped with deicing equipment, whether it was in use during the flight could not be determined. The system’s operation was not automatic and would have required manual activation by the pilot at regular intervals, which would have represented additional workload and a diversion of his attention as he attempted to maneuver through the terrain.

The pilot had no urgent personal activities that would have required him to return that day, but he was aware of the incoming weather system (which persisted for a few days after the accident), and he had a vacation planned for the following week. Returning by car would have resulted in a relatively long drive of over 300 miles.

Although the pilot held an instrument rating, no pilot logbooks were available to review, and his recency of instrument flight experience could not be determined. Given his familiarity with the accident route of flight and the isolated nature of the IMC along the route, it is likely that the pilot became overconfident in the airplane’s automation to navigate through the relatively small area of mountain obscuration. The airplane’s erratic pitch and bank excursions just before the accident are consistent with the known effects of pilot spatial disorientation, and the pilot’s susceptibility to the development of spatial disorientation would have increased with a lack of recent IFR flight experience. Based on the available information, it is likely that the pilot’s decision to continue the visual flight rules flight into IMC resulted in his spatial disorientation, a subsequent loss of airplane control, and impact with terrain.

Verbatim NTSB analysis and probable cause from the NTSB dataset