N978AFPilatus Aircraft PC-12/47E · 2008

On January 6, 2016, about 0835 eastern standard time, a Pilatus PC-12/47E, N978AF, collided with a ditch during a precautionary landing after takeoff from Savannah/Hilton Head International Airport (SAV), Savannah, Georgia. The pilot and copilot sustained minor injuries, and the airplane was substantially damaged. The airplane was registered to Upper Deck Holdings, Inc. and was being operated by PlaneSense, Inc,. as a Title 14 Code of Federal Regulations Part 91 positioning flight. Visual meteorological conditions prevailed, and an instrument flight rules flight plan was filed for the flight to Blue Grass Airport (LEX), Lexington, Kentucky.

The pilot in the left seat was the pilot monitoring and the copilot in the right seat was the pilot flying. The crew had the full length of the runway 1 available (7,002 ft) for takeoff. The pilots reported that the acceleration and takeoff was normal and after establishing a positive rate of climb, the crew received an auditory annunciation and a red crew alerting system (CAS) torque warning. The engine torque indicated 5.3 pounds per square inch (psi); the nominal torque value for the conditions that day was reported by the crew to be 43.3 psi. With about 2,700 ft of runway remaining while at an altitude of 200 ft msl, the copilot elected to land immediately; the copilot pushed the nose down and executed a 90° left descending turn and subsequently landed in the grass. Although he applied "hard" braking in an attempt to stop, the airplane impacted a drainage ditch, resulting in substantial impact damage and a postimpact fire.

The pilot reported that, after takeoff, he observed a low torque CAS message and the copilot told him to "declare an emergency and run the checklist." The pilot confirmed that the landing gear were extended and the copilot turned the airplane to the left toward open ground between the runways and the terminal. About 60 seconds elapsed from the start of the takeoff roll until the accident.

The airport was equipped with security cameras that captured the airplane from its initial climb through the landing and collision. One camera, pointed toward the west-southwest, recorded the airplane's left descending turn and its landing in the grass, followed by impact and smoke. A second camera, mounted on the control tower, pointed toward the southeast and showed the airplane during the initial climb before it leveled off and entered a descending left turn; it also showed the airplane land and roll through the grass before colliding with the ditch.

A review of Federal Aviation Administration (FAA) airman records revealed that the pilot held an airline transport pilot certificate with airplane single- and multi-engine land ratings. His most recent FAA first-class airman medical certificate was issued on November 19, 2015. He reported 23,141 total hours of flight experience with 534 hours in the accident airplane make and model at the time of the accident. He reported 154 hours and 35 hours of flight experience in the 90 days and 30 days before the accident, respectively. His most recent flight review was completed on September 14, 2015, in the PC-12.

The copilot held an airline transport pilot certificate with airplane single- and multi-engine land ratings. He held instructor ratings for airplane single engine, airplane multi-engine, and instrument airplane. His most recent FAA second-class medical certificate was issued on January 16, 2015. He reported 7,900 total hours of flight experience at the time of the accident, and 5,100 hours in the accident airplane make and model. He reported 168 hours and 37 hours of flight experience in the 90 days and 30 days before the accident, respectively. His most recent flight review was completed on September 14, 2015, in the PC-12.

According to FAA records, the low wing, T-tail, retractable-gear airplane was powered by a Pratt & Whitney Canada PT6A-67P, 1,200 shaft horsepower turboprop engine equipped with a Hartzell, four-bladed, hydraulically-actuated, constant-speed propeller assembly. The airplane was issued a normal category airworthiness certificate on December 19, 2008.

Maintenance records indicated that the airplane's most recent Approved Aircraft Inspection Program (AAIP) inspection was conducted on December 31, 2015, at an airframe total time of 4,189 hours and engine total time of 4,189 hours, with 3,976 cycles since new.

According to the operator, the airplane's estimated takeoff weight at the time of the accident was 9,100 lbs, which was 1,350 lbs below its maximum gross weight of 10,450 lbs. Takeoff distance performance calculations from the pilots operating handbook (POH) indicated that the runway required to safely take off and climb over a 50 ft obstacle using 15° of flaps at this weight was 2,000 ft; the runway being used was 7,002 ft long and according to the aircraft condition monitoring system (ACMS) data, the airplane rotated for takeoff about 2,100 ft down the runway.

According to the POH, red CAS warning messages required immediate corrective action by the pilot and amber caution messages required the pilot's attention, but not an immediate action. Whenever a red or amber message illuminated, the master warning or caution lamp would illuminate. A continuous chime would sound with red messages. A single chime would sound with all amber messages.

The torque and engine temperature indications in the cockpit were depicted with pointer-type display gauges and digital values (see figure 1). During normal operations, all digital values were displayed in white against a black background. The torque analog scale range was from 0 to 55 psi and the digital scale range was from 0 to 70 psi. Two white torque tick marks were on the analog scale at 15 psi and 25 psi. An amber tick mark at 36.95 psi indicated maximum climb/cruise torque, and a red tick mark at 44.34 psi indicated maximum continuous torque. A green arc on the torque gauge from 0 psi to the amber tick mark indicated normal operating torque range and beyond the amber tick mark, the gray arc is the maximum torque range. hen the excessive torque warning occurred, the digital indicator and pointer were displayed in red with the associated CAS message. There were no associated CAS messages for low torque values, only high torque exceedances activated a CAS caution or warning.

Figure 1. Primary Flight Display (PFD) and engine instruments window showing torque conditions examples.

The POH section 3, Emergency Procedures, Section 3.6.4, outlined the procedures for an engine torque CAS warning or caution:

1. TORQUE – Check torque indication

2. If torque above 44.3 psi, reduce power

If torque CAS warning or caution remains,

3. Aircraft – Land as soon as possible, using minimum power. If possible always retain glide capability, to the selected landing airfield, in case of total engine failure

The POH defined "Land as soon as possible" as landing without delay at the nearest airport where a safe approach and landing could be reasonably assured.

At 0853, the reported weather at SAV included wind from 020° at 9 knots, 10 statute miles visibility, broken clouds at 4,300 ft, temperature 1°C, dew point -7°C, and an altimeter setting of 30.46 inches of mercury.

According to FAA records, the low wing, T-tail, retractable-gear airplane was powered by a Pratt & Whitney Canada PT6A-67P, 1,200 shaft horsepower turboprop engine equipped with a Hartzell, four-bladed, hydraulically-actuated, constant-speed propeller assembly. The airplane was issued a normal category airworthiness certificate on December 19, 2008.

Maintenance records indicated that the airplane's most recent Approved Aircraft Inspection Program (AAIP) inspection was conducted on December 31, 2015, at an airframe total time of 4,189 hours and engine total time of 4,189 hours, with 3,976 cycles since new.

According to the operator, the airplane's estimated takeoff weight at the time of the accident was 9,100 lbs, which was 1,350 lbs below its maximum gross weight of 10,450 lbs. Takeoff distance performance calculations from the pilots operating handbook (POH) indicated that the runway required to safely take off and climb over a 50 ft obstacle using 15° of flaps at this weight was 2,000 ft; the runway being used was 7,002 ft long and according to the aircraft condition monitoring system (ACMS) data, the airplane rotated for takeoff about 2,100 ft down the runway.

According to the POH, red CAS warning messages required immediate corrective action by the pilot and amber caution messages required the pilot's attention, but not an immediate action. Whenever a red or amber message illuminated, the master warning or caution lamp would illuminate. A continuous chime would sound with red messages. A single chime would sound with all amber messages.

The torque and engine temperature indications in the cockpit were depicted with pointer-type display gauges and digital values (see figure 1). During normal operations, all digital values were displayed in white against a black background. The torque analog scale range was from 0 to 55 psi and the digital scale range was from 0 to 70 psi. Two white torque tick marks were on the analog scale at 15 psi and 25 psi. An amber tick mark at 36.95 psi indicated maximum climb/cruise torque, and a red tick mark at 44.34 psi indicated maximum continuous torque. A green arc on the torque gauge from 0 psi to the amber tick mark indicated normal operating torque range and beyond the amber tick mark, the gray arc is the maximum torque range. hen the excessive torque warning occurred, the digital indicator and pointer were displayed in red with the associated CAS message. There were no associated CAS messages for low torque values, only high torque exceedances activated a CAS caution or warning.

Figure 1. Primary Flight Display (PFD) and engine instruments window showing torque conditions examples.

The POH section 3, Emergency Procedures, Section 3.6.4, outlined the procedures for an engine torque CAS warning or caution:

1. TORQUE – Check torque indication

2. If torque above 44.3 psi, reduce power

If torque CAS warning or caution remains,

3. Aircraft – Land as soon as possible, using minimum power. If possible always retain glide capability, to the selected landing airfield, in case of total engine failure

The POH defined "Land as soon as possible" as landing without delay at the nearest airport where a safe approach and landing could be reasonably assured.

The airplane came to rest in a drainage ditch about 2,300 ft west of runway 1 and 1,800 ft south of the main terminal. The airplane was oriented on a 270° heading. There were three distinct landing gear ground tracks that measured about 700 ft in length in the grass leading up to the airplane. The airplane displayed substantial airframe damage caused by the impact and fire. The engine compartment sustained impact and fire damage, the cockpit was charred, and the airframe was buckled in multiple locations. The cockpit area was nearly separated from the fuselage and both wings sustained significant structural damage.

The engine was examined at Pratt & Whitney Canada Corp., St. Hubert, Quebec, Canada. The engine was covered in soot. The compressor section showed no evidence of pre-impact anomalies. Aside from the thermal and impact damage, there was no evidence of pre-impact anomalies noted with the gearboxes, governor, torque limiter, or fuel control unit (FCU) bellows that would have prevented normal operation.

Testing of the FCU bellows was conducted at Woodward, Inc. The bellows were installed in an exemplar unit and placed on the test stand. The compressor discharge pressure (CDP) was initially set to 125 psi absolute (psia), resulting in a fuel flow of 816 lbs/hour, which was maintained for 6 minutes without the fuel flow changing. The CDP pressure was then set to 100 psia. The control sustained a fuel flow of 475 lbs/hour, which was maintained for 8 minutes without a change in fuel flow. There were no anomalies observed during the test.

The modular avionics unit (MAU) contained data from the custom I/O module (CSIO). The CSIO module was examined and downloaded by Honeywell. According to Honeywell, examination revealed a small amount of debris on "a couple of pins" of the unit internal to the CSIO. The source of the debris could not be determined; however, all tests associated with the reading of the engine torque signals resulted in values within the required specifications and showed no faults or anomalies relevant to the accident.

The torque transducer did not show any signs of excessive mechanical damage, bent connections, or damage to the pressure port or sleeve. Pin continuity was confirmed. The harness was attached to the transducer. Testing of the wire resistances revealed no anomalies. The connector pins were clean and free of oxidation, dirt, and dust; some small scratches did not affect the connection. Additional testing of the transducer, which included input, output, voltage output over temperature range, and visual inspection yielded normal results. Testing of the torque transducer did not show any mechanical anomaly.

An actuator input/output processor module and an advanced graphics module were installed in the MAU. These components are part of the Honeywell Apex avionics system which captures data in the ACMS. The ACMS contains engine trend, cockpit and flight specific non-volatile memory files was downloaded. In addition, the fault history database (FHDB) that contained discreet airplane system condition data was downloaded.

Takeoff power was applied about 0834:46, at which time the torque value was 10.6 psi, the fuel flow was 353 lbs fuel used per hour (lbs/hour), the gas generator (Ng) speed was 80.2%, and the inter turbine temperature (ITT) was 603°.

About 1,500 ft down the runway while at 90 kts, the torque was at 40.9 psi, the fuel flow was 713 lbs/hour, the Ng speed was 94.5%, and the ITT was 711°.

At 0835:03, during rotation, the airplane was at 93 kts indicating a torque value of 44.9 psi, fuel flow 691 lbs/hour, the Ng speed was 94.2%, and the ITT was 709°.

Six seconds later, while on initial climb at 107 knots at 127 ft msl, the torque indication climbed to 71 psi while the fuel flow, Ng speed, and ITT remained within normal limits. The fault history data confirmed that the engine torque caution and warning were displayed during this time.

Four seconds later, the torque, fuel flow, Ng, and ITT values decreased to 47.3 psi, 683 lbs/hour, 93.6%, and 705°, respectively. As the copilot initiated the left descending turn, the torque and fuel flow were at 6.6 psi and 335 lbs/hour, respectively, the Ng was 80.4%, and the ITT was at 607° and decreasing. As the airplane descended and continued the turn to a heading about 260° on final approach for landing, the torque value increased to 18.9 psi and the fuel flow, Ng, and ITT all increased simultaneously until touchdown. The recorded parameters did not include flight control positions or power control lever position.