WPR17LA097

A total loss of engine power during cruise flight due to the failure of an engine piston rod for reasons that could not be determined due to extensive damage.

The commercial pilot was conducting a local sightseeing flight in a helicopter with paying passengers. During cruise flight, the engine lost partial power. The pilot performed troubleshooting steps, and while he was maneuvering to conduct a forced landing, the engine lost all power, and he immediately initiated an autorotation during which the helicopter struck a building and then landed hard. The helicopter was subsequently destroyed by fire. Multiple witnesses reported seeing an object fall from the helicopter as it flew over a highway just before the accident. Postaccident examination of the engine revealed that the No. 3 cylinder head assembly and piston were missing; they were found the following day. Examination of all the studs and through-bolts for the detached cylinder head revealed that they exhibited failure signatures consistent with ductile overstress fracture. Although some slight fretting and oxidation were observed on the cylinder flange, machining marks were also present, indicating that the movement was relatively limited and not consistent with a fatigue fracture of the cylinder attachments. Examination of the No. 3 cylinder's rod, piston, cylinder skirt, and flange revealed damage signatures indicating that the cylinder assembly was forced out by the end of the piston after the rod separated from the piston boss as the crankshaft continued to rotate. The rod end and its bushing were obliterated as the engine continued to operate, and fire damage precluded a full examination of the rod and, thus, a determination of what caused it to fail. Bushings in several of the engine's connecting rods were found displaced, and their ends were tinted blue, indicating that they were abnormally heated, which could have led to the bushings shifting. However, the abnormal heating likely happened after the connecting rod failed because heat tinting was also present on all of the wrist pins except for the pin from the failed rod. The engine oil filter, which sustained extensive thermal damage, contained charred remnants of bronzelike material, which was similar in appearance to the bushing material. This evidence indicates that a bushing failure had either taken place before the event or that it occurred secondary to another failure event to the piston rod. Other failure scenarios, such as a defect in the rod assembly or mechanical damage, could also have led to the observed failure; however, because the bushing and rod end were not located, it could not be determined what led to the rod's failure. About 3 1/2 months after the accident, the engine manufacturer published a mandatory service bulletin (SB), compliance of which was mandated by a Federal Aviation Administration airworthiness directive (AD), for the detection and replacement of connecting rods with nonconforming small end bushings because continued use of the nonconforming bushings could lead to a connecting rod failure. However, maintenance records showed that the engine was rebuilt by the manufacturer during a period not applicable to the SB. The manufacturer also issued an SB that recommended the inspection of the bushings any time a cylinder head was removed; however, a cylinder had never been removed from the engine since the rebuild. Although the shifted bushing condition observed on several of the connecting rods may have been secondary to the connecting rod failure, bushing shifting can cause connecting rod fractures. According to an AD issued by the Civil Aviation Safety Authority of Australia (CASA), all the manufacturer's engines that were new, factory rebuilt (such as the accident engine), or factory overhauled during the same calendar year as the accident engine were susceptible to premature wear of the connecting rod bushings. According to CASA, the development of this premature wear condition was relatively slow and could be detected through regular oil and oil filter inspections in accordance with the manufacturer's recommendations. Maintenance records indicated that these inspections had been regularly conducted on the accident engine at the required times. During these inspections, maintenance personnel should have been able to detect the excessive wear associated with bushing shift before the connecting rod fractured but failed to do so.

On May 5, 2017, at 1402 Pacific daylight time, a Robinson R44 helicopter, N981RR, lost engine power and landed hard following an autorotation near Santa Barbara, California. The flight instructor and two passengers sustained serious injuries, and the helicopter was destroyed by post impact fire. The helicopter was registered to Spitzer Helicopter LLC, and operated by Santa Barbara Helicopter Tours, as a revenue sightseeing flight under the provisions of Title 14 Code of Federal Regulations Part 91. Visual meteorological conditions prevailed, and no flight plan had been filed. The local flight departed Santa Barbara Municipal Airport, Santa Barbara, California, about 1345. The flight was planned to be a standard 20 minute, "City Tour" for the two passengers, which included a roundtrip flight from the airport to the Santa Barbara Zoo area, about 10 miles east. The outbound flight was uneventful with the pilot requesting a special VFR (visual flight rules) clearance due to low cloud ceilings. On the return leg, the weather had not improved, and the pilot requested a special VFR clearance to land. The airport controllers directed the pilot to hold outside of the airport's airspace due to landing traffic, and the pilot circled the city for about 10 minutes until the landing request was granted. He then proceeded to follow the 101 Highway west towards the airport, and a brief time later he noticed that the engine began to lose partial power, coincident with the clutch actuator light illuminating. He pulled the clutch actuator circuit breaker while evaluating his landing options. He began to maneuver the helicopter for landing at a golf course, rather than the highway or congested areas below, and a few seconds later the engine lost all power. He immediately initiated an autorotation, with the intention of landing on the golf course. During the final stage of the descent, he realized he would not be able to reach the grass due to a wall, so he landed just short in a parking lot. During the landing flare, the helicopter's main rotor blades struck the roof of a building, and the helicopter landed hard, spreading both skids (See Figure 1). All occupants egressed from the helicopter, while the golf course superintendent, who heard the impact, attempted to extinguish a fire which had developed at the rear of the helicopter's fuselage. Within a few minutes the fire had spread, ultimately engulfing the main cabin as the local fire department arrived about 5 minutes later. Multiple witnesses reported seeing an object fall from the helicopter as it flew over the highway, and post-accident examination revealed that the engines number 3 cylinder head assembly and piston were missing (See Figure 2). A search was conducted by volunteer search and rescue personnel from the Santa Barbara County Sherriff's Department, and the assembly was located the following day, in a field about 1/4 mile north of the wreckage location (See Figure 3). Figure 1 - Helicopter at the Accident Site Figure 2 - Helicopter at the Accident Site with Cylinder #3 Missing Figure 3 - Cylinder #3 Engine Examination The engine was disassembled following the accident by technical representatives from Lycoming Engines under the supervision of the NTSB investigator-in-charge, and inspectors from the Federal Aviation Administration (FAA). Complete examination reports are contained in the public docket, the following is a summary of findings. The engine remained attached to the aft frame support assembly and the forward engine mounts. It sustained extensive thermal damage, partially consuming most of its ancillary components. The entire number 3 cylinder assembly had separated from the engine, and neither its pushrods were recovered. The lower section of the crankcase deck adjacent to the underside of cylinder 3, along with the entire oil sump was consumed by fire. The crankshaft could be viewed through the opening, and was intact. The number 3 connecting rod remained attached to the crankshaft, and protruded from the opening. The outboard (piston) end of the rod was not present. The engine was examined for evidence of signatures typically associated with overspeed, as defined in Lycoming Service Bulletin SB369N. The valves, springs, and rocker arm assemblies for each cylinder were intact and undamaged. The valves slid easily out of their valve guides, and did not exhibit any indications of valve head "mushrooming" or burs to the areas in contact with the valve keys. The valve keys and exhaust valve caps were all undamaged, and the remaining pushrods were straight. The camshaft was intact, and all lobes and bearing surfaces were shiny in appearance and free of score marks. The oil filter had sustained thermal damage and was removed from the housing and disassembled. The filter element was black and charred, and contained bronze-colored non-ferrous metallic debris similar in appearance to connecting rod (piston-end) bushing material. Materials Laboratory Examination The engine crankcase, 6 connecting rod wrist pins, 12 wrist pin plugs, the through bolts, main crankshaft bearings, and the number 3 and 4 cylinder and piston assemblies were sent to the NTSB Materials Laboratory for examination. Examination revealed that the number 3 cylinder attachment studs and throughbolts all exhibited varying combinations and degrees of necking damage, smearing deformation, and rough and matte gray fracture features, all of which were consistent with ductile overstress. The damaged number 3 connecting rod had fractured just short of the bushing, which was missing. The corresponding piston was mostly intact, and the wrist pin was in place. The piston had multiple impact marks on the underside surfaces and piston skirt. The skirt was deformed inward at the forward and aft sides, trapping the piston within the cylinder barrel. A segment of the upper side of the piston skirt was fractured and missing, and a similar shaped segment of the skirt at the lower side of the barrel was fractured and flattened against the cylinder attachment flange. The damage to both segments was consistent with contact with the number 3 connecting rod. At the upper side of the cylinder flange, the attachment holes were elongated. The location of the hole elongation was consistent with the direction of shear deformation observed in the corresponding studs. At the lower side of the cylinder, the flange was bent, and cracked though the middle two attachment holes. The cracks were associated with deformation consistent with ductile overstress fracture. Most of the number 3 cylinder attachment flange around the attachment holes appeared dark gray, but portions of the flange were either coated in an orange oxidized layer, or appeared light gray and smeared. Machining marks were visible around most of the surfaces, including many of the oxidized areas. The crankcase was separated to reveal the mating sides of the crankcase halves. The mating surfaces had a smooth, polished appearance with scattered pits. Similar pit features were also observed on other machined surfaces including the saddle surfaces supporting the main journal bearings and the camshaft journal bore surfaces. The connecting rod for cylinder 3 was fractured where the wrist pin bore intersected the beam section of the rod. A curved impact mark was observed on the outboard end of the connecting rod. The shape of the impact mark was consistent with the inboard end of the cylinder barrel skirt at the "6 o'clock" position. The rod end surfaces were obliterated by post-fracture contact damage, and fine features of the fractures were mostly obliterated by oxidation. Remaining fracture surfaces were visually smooth with sliding contact marks, consistent with ductile overstress fracture in compressive shear loading. Elongated dimple features were observed on remaining fracture surfaces, also consistent with