ERA24LA223

At the conclusion of a training flight that included two uneventful landings using ground fine/beta or reverse thrust settings, the pilots of the multi-engine, turbo-propeller equipped airplane returned to the home base airport. The pilot flying (PF) reported that he moved each propeller control full forward. He reported that the airplane crossed the runway threshold at 110 knots indicated airspeed (KIAS), or landing reference speed (Vref) plus 22 knots (kts). The airplane touched down with both power levers at flight idle and the “REV NOT READY” annunciator extinguished. After touchdown, the PF moved both power levers aft of the idle gate into beta, then into reverse. Recorded GPS data revealed that the airplane began deviating slightly to the left while decelerating, then about 8 seconds after touchdown, while at 66 kts groundspeed (about 72 KIAS), the airplane quickly veered to the left. The flight crew applied right rudder and right brake, but were unable to correct the airplane’s trajectory. The pilot monitoring (PM) stated that it felt like the airplane was accelerating as it departed the left side of the runway. The GPS data showed that the airplane continued to slow while rolling off the runway. The airplane continued through a 5-ft-deep depression in the grass off the side of the runway before it came to rest with the left main and nose landing gear collapsed, resulting in substantial damage to the fuselage and left wing. Flight control continuity was confirmed from the cockpit to each control surface, while the rudder and aileron trims were neutral. No discrepancies were noted with either engine power lever or propeller control lever. There was no evidence of pre-impact mechanical failures or malfunctions that would have precluded normal operation of either propeller constant speed unit (CSU) or its attached beta valve; either propeller; or either fuel control unit. No propeller-related issues were reported by the accident flight crew during the two previous landings earlier that day using beta or reverse, nor were any propeller discrepancies reported by other pilots who had recently flown the airplane and employed beta or reverse. Flight testing of an exemplar airplane equipped with the same engines and propellers revealed that, with each power lever at flight idle and each propeller control full forward, the propeller CSUs were still governing to 120 KIAS, while flight testing with the propeller controls set to 1,700 rpm revealed that the propeller CSUs were still governing to 90 KIAS; safety concerns prevented flight testing below that airspeed. By design, propeller blade movement into beta and reverse cannot occur unless its respective propeller CSU is in an underspeed condition. Based on the examination of the airplane, there was no evidence of any mechanical anomaly with either propeller CSU or its beta valve that would have prevented the propellers from entering beta and then reverse as commanded if either CSU had been in anunderspeed condition. Flight testing confirmed that if the propeller controls were full forward, any airspeed below 120 KIAS would have resulted in both propeller CSUs being in an underspeed condition; the airplane was well below that speed at touchdown. When the PF moved each power lever aft of the flight idle gate and into beta, then reverse, after touchdown, the propellers should have produced symmetrical reverse thrust. However, the circumstances of the accident flight are consistent with production of asymmetric reverse thrust, which resulted in the flight crew’s inability to maintain directional control and a runway excursion. Based on the postaccident examination of the airplane and propeller systems, the reason for the asymmetric reverse thrust could not be determined.

The two-place airplane was powered by two Pratt & Whitney Canada PT6A-41 engines equipped with four-bladed Hartzell HC-D4N-3A/D9515 constant-speed, single acting, hydraulically-controlled, manually-reversable propellers. The airplane was not configured with a ground idle reset system or an annunciator that illuminated when either propeller was in the beta or ground fine positions; however, it was equipped with a propeller reverse not ready annunciator on the annunciator panel that illuminated only when the landing gear handle was down, and the propeller levers were not at HIGH RPM (full forward) position. Each propeller constant speed unit (CSU) was equipped with an integral beta valve that controlled propeller blade angle when operating in the beta control range, which corresponded to a range of operation where the blade angle was between the primary blade angle (PBA) and reverse. In beta mode, control of the propeller pitch was a direct function of the position of the beta valve. Changing propeller blade angle on the ground below idle or PBA into beta mode required movement of the power lever aft of the idle detent and the CSU flyweights must be in an underspeed, or non-governing, condition. According to the engine manufacturer representative, there is a “deadband range” when moving the power lever toward maximum reverse or forward, where fuel flow, gas generator speed (Ng), and torque remains at idle values. As the power levers are moved further toward the maximum reverse or forward positions, fuel flow, Ng, and torque will increase at a similar rate. The deadband range was determined by the cambox and rigging. A review of the airplane’s “Flight Log List” revealed no recorded discrepancies for the 4 flights that accrued 12 landings and 4.4 flight hours during the 90 days before the accident. A pilot who had flown the airplane on two separate flights, one of which was 3 months before the accident and the other 2 months before the accident, reported that he experienced no discrepancies using beta or reverse. The chief pilot, who had also flown the airplane 5 days before the accident, reported no discrepancies with either propeller while taxiing using beta, or during multiple landings using reverse. He also reported that the Vref airspeed at the airplane’s landing weight and with full flaps extended was 88 KIAS. On May 20, 2024, about 1008 eastern daylight time, a Beechcraft TC-12B airplane, N762MC, was substantially damaged when it was involved in an accident near Lehigh Acres, Florida. The pilot and the copilot were not injured. The airplane was operated as a public aircraft. The pilots departed on a training flight, which included two uneventful full-stop landings at a different airport. During one of the landings, they used ground fine/beta thrust, and during the second, they used reverse thrust. At the conclusion of the flight, the pilots returned for landing on runway 32 at their home base airport. The pilot monitoring (PM) reported thateach condition lever was in the high idle position and the pilot flying (PF) stated that he moved the propeller controls full forward on the base leg of the airport traffic pattern. The PF also stated that the airplane crossed the runway threshold at 110 knots indicated airspeed (KIAS), or landing reference speed (Vref) plus 22 knots. The PF placed the power levers to flight idle, and reported a normal touch down between 90 and 100 KIAS. The PM stated that the “REV NOT READY” annunciator was not illuminated; the PF moved each power lever aft of the idle gate into beta and then into reverse. Recorded GPS data revealed that the airplane began deviating slightly to the left while decelerating, then about 8 seconds after touchdown, while at 66 knots groundspeed or about 72 KIAS, the airplane quickly veered to the left. The PM stated that it felt like the airplane seemed to be accelerating as it departed the left side of the runway. GPS data showed that the airplane continued to slow while rolling off the runway. The airplane continued through a 5-ft-deep depression in the grass off the side of the runway and came to rest upright with the left main and nose landing gear collapsed. Flight testing was performed using another of the operator’s aircraft equipped with the same engines and propellers to determine the airspeed at which, when at idle power, the propellers are windmilling and no longer being governed when the propeller controls were set to 2,000 and 1,700 rpm. The testing was performed between 130 KIAS and 90 KIAS, at 10-knot intervals with both power levers at flight idle. The flight testing at 2,000 rpm revealed that each propeller remained at that set point up to and including 120 knots then began to decrease at each of the following test points. Testing at 1,700 rpm revealed that the propellers remained at that rpm to the lowest tested airspeed of 90 knots. Safety concerns prevented testing below that airspeed. Examination of runway 32 first revealed marks from the left main landing gear, while marks from the right main landing gear were noted farther along the runway; both tire marks were noted left of the runway centerline. Using satellite imagery, the marks from the left main landing gear were measured to begin about 1,716 ft from the approach end of the runway, while the marks from the right tires began about 1,795 ft from the approach end of the runway. The first marks showed a gentle drift to the left, then an aggressive veer to the left that began about 1,900 ft from the approach end of the runway. The airplane came to rest in a dry fish pond/depression about 2,450 ft from the approach end of the runway. Examination of the airplane following recovery revealed compression buckling at fuselage station 348, on the upper left wing skin at wing station (WS) 195, and also just outboard of the wing fuel filler cap at WS 306. Examination of the left main landing gear tires revealed that the outboard tire had no pressure (which wasattributed to postaccident recovery), while the inboard tire pressure was 44 psi, 18 psi less than specified inflation pressure. Examination of the right main landing gear revealed that both tires exhibited flat spots consistent with sliding. The inboard tire also had no air pressure and exhibited 2 layers of cord showing, while the outboard tire was at 46 psi and had 1 layer of cord showing. Flight control continuity was confirmed from the cockpit to each control surface. Examination of the cockpit revealed that both power levers were at the flight idle position, both propeller controls were about 1 inch forward of the feather position, and both condition levers were past the low idle detent. The rudder and aileron trims were neutral. Continuity was confirmed for each power lever from the pedestal to each respective attach point. Each power lever moved freely aft of the idle stop into beta range but was not moved farther aft into reverse (normal) because the engines were not running. Each propeller control moved forward to the full forward, or high rpm, position with no discrepancies noted. Examination of the left engine revealed that the beta ring was pulled forward and jammed in the forward position consistent with impact damage. The carbon block was in the guide and the beta lever was under the retention arm. Because of the position of the beta valve, the power lever rigging check could not be performed. After removing the low pitch stop sleeves, which allowed the beta ring and beta valve to relax to the “home” position, the beta valve was 3/32 inches forward of the beta valve cap when it should have been flush with the cap nut. Operational testing of the fuel control unit (FCU) and CSU as received revealed minor out-of-tolerance conditions consistent with field adjustments, but there was no evidence of preimpact mechanical failure or malfunction of either unit that would have precluded normal operation. Examination of the right engine revealed the beta ring was in the home position, the carbon block in the guide, the beta lever was under the retention arm, but the beta valve was positioned slightly below the normal flush point. Testing of the power lever revealed the beta valve was 4/32 inches forward of the beta valve cap when it should have been flush with the cap nut. The air conditioning compressor, fuel control unit, and the starter generator were fractured from the accessory gearbox, but remained partially attached by cables, hoses, and electrical cables. Four of the six inlet case struts were also fractured. Operational testing of the CSU as received revealed minor out-of-tolerance conditions, but there was no evidence of preimpact mechanical failure or malfunction of the unit or beta valve that would have precluded normal operation or prevented the unit from going into an underspeed condition. Operational testing of the FCU as received could not be performed because of impact damage to the drive body, which also fractured thedriveshaft coupling. A slave drive body and missing preformed packings were installed to test functionality of the fuel section. Because the slave drive body was not calibrated with the flow body, not all test points were performed. Of the test points that were performed, minor out-of-tolerance conditions consistent with field adjustments were noted. There was no evidence of preimpact mechanical failure or malfunction that would have precluded normal operation. Examination of the left propeller revealed that all blades were bent aft midblade and bent opposite direction of rotation, twisted toward low pitch with chordwise rotational scoring on the cambered side of the blade. All blades exhibited leading edge nicks. The L2 and L4 blades were found rotated beyond the reverse position, indicating internal damage. Blades L1 and L3 were at approximately flight idle/ground idle position. The pitch change rod was displaced forward 1.87 inches from the feather stop while the low pitch stop sleeves were displaced 0.26 inches forward of the home position. There was no evidence of preimpact failure or malfunction of the left propeller. Examination of the right propeller revealed that all four blades appeared to be in the feathered position. All blades were bent aft and also bent opposite the direction of rotation in the outboard 1/3 of the blade. All blades were twisted toward low pitch and exhibited rotation chordwise scoring on the cambered side of the blade. The pitch change rod was displaced 0.11 inch forward of the feather stop. All blades exhibited leading edge nicks. There was no evidence of preimpact failure or malfunction of the right propeller.