ERA15FA129

The left engine propeller's uncommanded travel to the feathered position during takeoff for reasons that could not be determined due to impact damage. Contributing to the accident was the flight crew's failure to establish a coordinated climb once the left engine was shut down and the left propeller was in the feathered position.

The accident flight was a repositioning flight being operated by two airline transport pilots, and it was the multiengine turboprop airplane's first flight after an aviation maintenance technician (AMT) had replaced the left engine propeller with an overhauled propeller. The AMT subsequently performed an engine run, which included verifying correct power settings and corresponding blade angles. A review of flight data recorder (FDR) data revealed that, about 2 seconds after rotation, the left engine propeller rpm decreased to 60 percent, and the left engine torque increased off-scale (beyond 5,000 ft-lbs), which is consistent with the left propeller traveling to the feathered position and the engine torque increasing in an attempt to maintain propeller rpm. About 30 seconds later, the flight crew shut down the left engine and attempted to return to the departure airport. Postaccident examination of the rudder trim actuator revealed that the rudder trim was at its full-right limit, which would have occurred to counteract the left engine drag before its shutdown. Based on this evidence, it is likely that the flight crew did not readjust the trim when the drag was alleviated, which resulted in the airplane being operated in a cross-controlled attitude for about 50 seconds with a left bank and full-right rudder trim. Although the airplane should have been able to climb about 500 ft per minute with one engine operating, it slowed and descended from 300 ft in the cross-controlled attitude until it stalled, as indicated by a stall warning recorded by the cockpit voice recorder, and subsequently impacted terrain. Examination of the wreckage, including teardown examination of the left engine and propeller, did not reveal any preimpact mechanical anomalies. Review of the airplane maintenance manual revealed instructions to check the propeller reversing linkage on the front end of the engine, which controlled the beta valve, for proper rigging during propeller installation. The manual also contained a warning that misadjustment of the beta valve can cause unplanned feathering of the propeller and result in a possible hazard to airplane operation and overtorque damage to the engine; however, the beta valve rigging could not be verified postaccident due to impact damage. Additionally, the ground/flight idle solenoid energizes when weight becomes off wheels and further opens the beta valve, which could exacerbate an existing misrigged condition as soon as the airplane becomes airborne, which is when the airplane experienced the uncommanded propeller feathering. The FDR data were consistent with the flight crew not performing the Before Takeoff (Runup) checklist. One of the items on that checklist was a low-pitch solenoid test, which would have energized the solenoid and possibly driven the left propeller uncommanded to feather during ground operations rather than in flight. A similar test during the postmaintenance engine-run would have had the same results.

Cockpit Voice Recorder The airplane was equipped with a Fairchild model A-100S CVR. The CVR recorded a minimum of 30 minutes of digital audio stored on solid state memory modules. Four channels were recorded: one channel for each flightcrew, one channel for a cockpit observer, and one channel for the cockpit area microphone. The channel for a cockpit observer did not record any audio, nor was it required to. The other three channels recorded audio of poor quality; however, a CVR Group convened at the NTSB Vehicle Records Laboratory, Washington, DC, on April 28, 2015 and was able to prepare a transcript of the recording. The transcript included 12 minutes, 26 seconds of the audio recording, which began as the flightcrew was preparing to start the engines and ended after impact (for more information, and a copy of the transcript, see CVR Group Chairman's Factual Report of Investigation in the NTSB public docket). Flight Data Recorder The airplane was equipped with a Loral/Fairchild Model F1000 flight data recorder (FDR). Data were successfully downloaded at the NTSB Vehicle Recorders Laboratory, Washington, DC. The following parameters were recorded and plotted: autopilot engage; engine No. 1 propeller reverse; engine No. 1 propeller speed; engine No. 1 torque; engine No. 2 propeller reverse; engine No. 2 propeller speed; engine No. 2 torque; indicated airspeed; longitudinal acceleration; magnetic heading; microphone No. 1 keying; pressure altitude; and vertical acceleration. Review of the plotted data revealed that engine torque and propeller rpm increased slightly about 10 seconds after the flight was cleared for takeoff, consistent with the airplane taxiing onto the runway for takeoff. At 1437:20, the engine torque increased to approximately 3,000 foot-pounds (ft-lbs) and the propeller rpm increased to 90 percent, consistent with takeoff power. At 1437:54, which was 2 seconds after one of the pilots stated "rotate," the left engine propeller rpm decreased to 60 percent while the left engine torque increased off-scale (beyond 5,000 ft-lbs), consistent with the left propeller travelling to the feathered position and the engine torque increasing in an attempt to maintain propeller rpm. The left engine propeller rpm and torque decreased to 0 at 1438:28 and 1438:34, respectively, consistent with the engine being shut down. At that time, the right engine propeller rpm remained at 90 percent and right engine torque remained between 3,000 to 5,000 ft-lbs until the end of the data, which was approximately 50 seconds later. During that time, the indicated airspeed decayed from about 110 knots to 100 knots and the magnetic heading indicated a 40-degree left turn (for more information, see FDR Specialist's Factual Report of Investigation in the NTSB public docket). Additionally, FDR data recorded between the end of the previous flight and the accident flight were consistent with an approximate 12-minute postmaintenance engine run; however, there was no evidence of a torque check, overspeed governor test, low pitch solenoid test, primary governor check or autofeather check. Checklists The "Before Takeoff (Runup)" checklist included: "…7. Cockpit Voice Recorder (if installed)..……….Check a. Headset………………………………..……..Check b. CVR Test Button…………………..Press And Hold • Observe meter needle in green band. • Listen for test tone in headset… …16…c. Power Levers………......Increase Until Props Are Stabilized at 1520 To 1610 RPM… …17. Low Pitch Solenoid…………….………………TEST a. Power Levers………………...Idle (Note Prop RPM) b. Prop Test Switch……………..…Hold To Low Pitch c. Prop RPM……………Stabilized Approx 200 RPM Below Value in Step a…. …20. Autofeather………..…………………………..Check a. Power Levers...Set Approximately 700 FT-LBS Torque b. Autofeather Switch…………………...…...Hold In Test (Both Autofeather Annunciators – Illuminated) c. Power Levers……………………....Retard Individually 1) At approximately 550 ft-lbs- Opposite Annunciator……………....Extinguished 2) At Approximately 320 ft-lbs- Both Annunciators………………….Extinguished…" Data consistent with execution of these checklist items were not captured on the CVR or FDR. Performance Review of the Beech 1900C Airliner Emergency Procedures revealed: "Engine Failure During Takeoff (At or Above V1) – Takeoff Continued 1. V1 Speed……………………………………………………………….Rotate 2. Landing Gear (when positive climb established)………………………….Up 3. Airspeed…………………………………….…Maintain V2 to 400 Feet AGL 4. Propeller (inoperative engine)………………………………Verify Feathered 5. Airspeed (at 400 feet AGL minimum)……………Increase To Flaps Up Vyse 6. Flaps………………………………………………………………………..Up 7. Climb to 1000 feet AGL and Accomplish the Following Cleanup Procedures:…" The airplane's maximum gross weight was 17,610 pounds. A weight and balance could not be computed due to the destruction of baggage/cargo; however, there was no evidence to indicate the airplane was at or over gross weight. The airplane's published takeoff safety speed (V2) was 120 knots (at 16,600 pounds and flaps up) and the single-engine best rate of climb speed (Vyse) was 125 knots. The published single-engine rate of climb under the given conditions was approximately 500 feet per minute. Propeller Review of the Beech 1900C airliner maintenance manual, chapter 61, Propeller Maintenance Practices, C. Installation, revealed, "…(13) Check the propeller reversing linkage on the front end of the engine for proper rigging…" Ground Fine Stop System The airplane's ground fine stop system was designed to prevent the propeller from operating at too fine (flat) a pitch during flight, whereas such a pitch might be utilized during ground operations. It used an electrical solenoid mounted on the front of the reversing push/pull cable to limit the propeller blade angle to 0 to 7 degrees. The solenoid was connected to the propeller reversing lever by means of a slotted clevis, which allowed the reversing lever to be pulled aft, resetting the beta valve. The electrical solenoid could be energized by two ground paths. One was through the right landing gear squat (weight on wheels) switch. The other occurred by pulling the power levers to the ground idle fine switch. The solenoid in each case energizes and pulls the reverse lever of the beta valve aft to reset the blade angle. Review of the Beech 1900C airliner maintenance manual, chapter 76, Controls – Maintenance Practices, 8 Propeller Ground-Fine Solenoid, A. Rigging revealed: "(1) Install the solenoid in the supporting bracket with the aft surface of the solenoid flush with the aft portion of the support bracket. (2) Connect the solenoid arm to the propeller reversing lever (3) Position the solenoid bracket on the beta cable housing so that the distance between the forward surface of the solenoid and the center of the clevis pin through the end of the solenoid plunger is 0.5-inch. (4) Assure that all hardware is tight and that the safety wired are installed at the beta cable end as appropriate." The section also contained a warning, "Warning: Misadjustment of the beta valve can cause unplanned feathering of the propeller. Resulting in a possible hazard to airplane operation and overtorque damage to the engine" The rigging on the accident airplane could not be verified due to impact damage. Autopsies were performed on the pilot and copilot by the Miami-Dade County Medical Examiner Department, Miami Florida, on February 12, 2015. Toxicological testing of the pilot and copilot was performed on the pilot by the FAA Bi