ERA22LA115

The pilot and a pilot-rated passenger were starting the engine before a positioning flight. The airplane’s cockpit voice recorder captured some checklist usage; however, there was no comment for the position of the power control lever (PCL) in the “Cockpit” checklist. Although the checklist specified and the pilot reported that the PCL was in the idle position at engine start, based on the flight data recorder (FDR) data, it was not in the idle position at any time during the engine start or accident sequence. After the engine was started with the parking brake engaged and the pilot’s application of normal toe brakes, the airplane accelerated on the ramp to a maximum speed of 26 knots and briefly attained a maximum propeller speed of 1,870 rpm, which was likely due to the cold engine oil temperature, but well within rpm and time limits. The pilot was unable to stop the airplane on the ramp, which likely contained ice, or to move the condition lever (CL) to stop the engine, and the airplane impacted a parked, unoccupied airplane, resulting in substantial damage. Although cockpit rigging checks of the PCL and CL showed out-of-tolerance conditions, rigging checks at the engine for those controls and for the cut-off guard of the CL in the cockpit revealed no evidence of preimpact failure or malfunction that would have precluded a normal engine start. There were no discrepancies with the fuel control unit that would have caused the engine acceleration issue at engine start, and there were no discrepancies with the parking brake or the left brake. Although testing of the right brake was not performed due to separation of the right wing, there was no evidence of preimpact failure or malfunction of the brakes or parking brake that contributed to the collision. While there was no flat-spotting of either main landing gear tire, alternating skid marks on the ramp were likely due to the ice on the ramp. Recorded data from a previous, uneventful engine start performed by the accident pilot with the PCL in the idle position revealed normal engine parameters. Testing of an exemplar airplane with the PCL in the correct position at engine start (idle) and the CL in either ground or flight idle positions during engine start revealed little forward movement with no brakes applied. The PCL was found in a forward position following the accident. Although the exact position of the PCL at engine start could not be determined from the FDR information, based on the maximum engine and propeller parameters recorded by the FDR, and the speed the airplane reached before the collision, it is likely that the PCL was nearly fully advanced during the engine start. Had the pilot performed a flow check then reviewed the checklists for any missed items or engaged the passenger to perform challenge-and-response to the entire checklist, the incorrect position of the PCL could have been detected and corrected before engine start. Though the pilot reported being unable to secure theCL while accelerating and approaching the parked airplane, there were no reported discrepancies with the cut-off guard, or of the CL during postaccident operational testing. Therefore, the pilot’s unsuccessful attempt to secure the CL were likely due to errors made while multitasking when presented with an unexpected situation (airplane acceleration during engine start and being unable to stop the airplane) while fast approaching a parked airplane.

The airplane was certificated under 14 CFR Part 23, effective February 4, 1991, including Amendments 23-1 through 23-42. Amendment 42 of 14 CFR Part 23.735, Brakes, effective February 4, 1991, specified that the, “Brakes must be able to prevent the wheels from rolling on a paved runway with takeoff power on the critical engine, but need not prevent movement of the airplane with wheels locked.” The airplane was equipped with a Pratt & Whitney Canada PT6A-67P engine. According to the Pilot’s Operating Handbook and EASA and FAA approved Airplane Flight Manual, the engine operating limits for takeoff are specified to be 1,200 shaft horsepower, 44.34 psi torque, 104% Ng, and 1,700 Np. A maximum transient Np value of 1,870 was allowed for 20 seconds. The PCL selects the required engine power (gas generator speed, specified as Ng) and in certain conditions it directly controls the propeller pitch. When the PCL is at the idle detent, Ng is at idle and the propeller is at minimum pitch. On February 2, 2022, about 0655 eastern standard time, a Pilatus PC-12/47E, N357JK, was substantially damaged when it was involved in an accident at Long Island MacArthur Airport (ISP), Ronkonkoma, New York. The commercial pilot and a pilot-rated passenger were not injured. The airplane was operated as a Title 14 Code of Federal Regulations (CFR) Part 91 personal flight. The pilot stated that the ramp conditions where the airplane was parked for engine start consisted of spots of snow, slush, “and a spot of ice,” though he felt the ramp was safe. The pilot-rated passenger, who had accrued 252 hours in the airplane, performed the exterior preflight inspection for the anticipated positioning flight, reporting no airplane discrepancies. He then entered the airplane and closed the door, later reporting there were no PIREPs or NOTAMs for any adverse ramp condition(s). After the passenger joined the pilot in the cockpit, they used the manufacturer’s checklist to prepare for engine start. According to a summary of the cockpit voice recorder (CVR) recording, some under-the-breath comments similar in cadence to reading a checklist to oneself were heard, and portions of checklists were read out loud and challenged. There was no audible comment or challenge-and-response concerning the position of the power control lever (PCL) listed in the “Cockpit” checklist that immediately preceded the “Before Starting Engine” checklist. Although the PCL was specified to be in the idle position at engine start, and the pilot reported that it was in the idle position during the engine start, the data from the airplane’s flight data recorder (FDR) revealed that the PCL was not in the idle position for the entirety of recorded data; although the exact position could not be determined. The pilot applied the toe brakes and the parking brake while he engaged the engine starter switch and verified that the oil pressure was rising, and the compressor turbine (Ng) speed had increased to 13%. At 13%, basedon the engine oil temperature and a note for a cold start in the “Engine Starting” checklist, he moved the condition lever (CL) to the flight idle position. He observed Ng increase and felt the airplane “lurching forward.” He pressed harder on the toe brakes, but the airplane continued forward while the engine, “spooled up.” FDR data indicated that the engine Ng, propeller speed (Np), and torque increased to maximum values of 95.1%, 1,870 rpm maximum for about 6 seconds, and 43.2 psi, respectively. While moving forward, the pilot attempted to secure the CL, but his thumb slipped off the cut-off guard. He looked down at the CL, then back outside, and noted the airplane was fast approaching a parked airplane, reaching a maximum groundspeed of about 26 knots based on FDR data. Surveillance video showed that the right wing of the Pilatus separated at the wing root after impacting the nose section of the parked, unoccupied airplane. The passenger assisted the pilot in securing the engine after impacting the airplane. Postaccident examination of the cockpit by a Federal Aviation Administration (FAA) airworthiness inspector revealed that the parking brake was engaged, the PCL was in a forward position, the CL was in the cut-off position, and the manual override lever was in the off position. Postaccident operational testing of the left brake and parking brake revealed that both were operational with no discrepancies noted. Both main gear tires showed minor scuffing, but no flat-spotting. Because the right wing was separated, no testing of the right brake was performed. Postaccident examination of the ramp performed by an FAA airworthiness inspector about 1 hour after the accident revealed that it was wet with patches of snow and ice. Alternating skid marks consistent with the width of the main landing gear tires began about 50 ft from the reported parking spot and continued for 75 ft toward the nose of the impacted airplane. Inspection and examination of the engine controls, the throttle friction, the cut-off guard of the CL, and rigging checks of the engine controls were performed by the FAA airworthiness inspector using the appropriate sections of the airplane maintenance manual (AMM). No discrepancies were noted during the inspection of the engine controls, throttle friction device, or inspection and testing of the cut-off guard of the CL. A check of the engine control rigging revealed the maximum reverse stop at the center console was between 0.005 inch to 0.015 inch more than the clearance specified in the AMM, and the forward stop, also at the center console, was about 0.200 inch more than the value specified in the AMM. The rigging checks of the PCL and CL at the engine were within tolerances. The engine fuel control unit (FCU) was removed from the engine and shipped to the manufacturer’s facility for operational testing with NTSB oversight. According to a report from the FCU manufacturer, although multiple test points in the start andacceleration schedule were out of acceptance test procedure limits during testing as received, the reported out-of-limit tests points did not preclude normal operation of the FCU and did not contribute to the reported engine acceleration. According to the maintenance records, the most recent inspections were performed on January 20, 2022. An adjustment of the overspeed governor and inspection of the FCU linkage were performed during this time. The airplane was approved for return to service, and there was no maintenance performed to the airframe or engine between the numerous inspections and the accident date, accruing 16 hours over 10 flights. Recorded data from the FDR for the previous uneventful engine start, on January 30, 2022, performed by the pilot of the accident flight, revealed that with the PCL positioned to idle, the Ng, Np, and torque values increased to maximum values of 66.7%, 1,036 rpm, and 7.59 psi, respectively. Engine start testing was performed of an exemplar airplane, witnessed and video recorded by personnel from the FAA. During an engine start with the PCL in the idle position (correct per checklist) and the CL moved to the ground idle position at the appropriate percent Ng, the Np increased to about 963 rpm and with no brakes applied the airplane was “barely starting to move a little bit.” Additional testing was performed with the PCL at idle and the CL moved to the flight idle position, and with no brakes applied, the airplane moved forward slightly.