N5626DBeech A36 · 1990

On June 28, 2015, at 1745 eastern daylight time, a Beech A36, N5626D, impacted a residence and terrain following a total loss of engine power near Plainville, Massachusetts. The commercial pilot and two passengers were fatally injured, and the airplane was consumed by postcrash fire. The airplane was being operated as a 14 Code of Federal Regulations Part 91 personal flight. Instrument meteorological conditions existed near the accident site about the time of the accident, and an instrument flight rules flight plan had been filed. The flight originated at Lancaster Airport, Lancaster, Pennsylvania, about 1612, and was destined for Norwood Memorial Airport (OWD), Norwood, Massachusetts.

Review of radar and voice communication data provided by the Federal Aviation Administration (FAA) revealed that the pilot was preparing to conduct an instrument approach procedure for landing at OWD. The flight was about 15 miles from OWD at an altitude of 3,300 ft mean sea level (msl) when the pilot declared an emergency to air traffic control, stating that the airplane was experiencing an "engine problem." The pilot asked the controller about the nearest airport, and the controller provided him with radar vectors to an alternate airport. About 30 seconds later, the pilot advised that he was unable to maintain altitude, and he subsequently stated, "we got a real bad vibration we're losing engine." The controller advised that there was a highway right of the airplane's position and about 2.5 miles away, and the pilot responded, "we have no engine we're [in instrument meteorological conditions] I need help." The controller provided vectors toward the highway, which the pilot acknowledged. Shortly thereafter, the pilot stated, "we're gliding." At this time, radar data showed the airplane about 1,450 ft msl and 2 miles south of the highway. The last recorded radar return, about 40 seconds later, showed the airplane in a right turn about 700 ft msl and at a groundspeed of 66 knots about 0.1 mile from the accident site.

Several witnesses reported hearing an airplane engine making noise and then stopping or going silent. They could not see the airplane due to the low cloud ceiling. One witness, who was located across the street from the accident site, described first hearing a "low moan buzzing sound" and then looking up and seeing the airplane over the trees at the rear of his property with its wings "wagging back and forth." He added that, as it passed over his house, it was flying in a straight line and "wobbling." He then lost sight of it, and shortly after, heard a "boom." Another witness, located adjacent to the accident scene, described hearing what sounded like "a broken fan" before the airplane impacted the house.

According to FAA records, the pilot held a commercial pilot certificate with airplane single-engine land, airplane multiengine land (limited to center thrust), and instrument airplane ratings.

The pilot's logbook was damaged by fire and was not legible. His most recent FAA third-class medical certificate was issued on April 17, 2014, at which time, he reported 1,500 hours of total flight experience.

The six-seat, low-wing airplane was manufactured in 1990 and powered by a Continental IO-550-B6A, 300-horspower, six-cylinder engine driving a three-bladed propeller. The airplane's most recent 100-hour inspection was completed on June 5, 2015, at which time, the airplane had accumulated 2,143 total hours in service and 2,143 total engine hours, 410 hours of which were accumulated since the engine was overhauled in June 2011.

The engine maintenance logs noted that the oil filter was examined for metal particles during the most recent inspection as well as during five of the seven oil changes performed since the engine overhaul, and no metal or debris was found. The engine oil was also sampled and analyzed by a test laboratory three times since overhaul with normal results.

When the engine was overhauled, an overhauled primary alternator and a new parts manufacturer approval (PMA) alternator drive coupling were installed. The alternator was removed and replaced twice in October 2011, at 21 and 23 hours since engine overhaul, respectively. No logbook entries specifically mentioned the alternator drive coupling (a gear assembly installed on the alternator shaft that mates with a gear on the engine crankshaft to turn the alternator). According to a representative from the engine overhaul facility, the serial number of the coupling in service at the time of the accident matched the one that was installed at the time of overhaul. The coupling did not have a specified life limit; however, the instructions for continued airworthiness provided by the coupling manufacturer recommended that it be replaced at engine overhaul. The airplane was also equipped with a standby alternator installed in accordance with a supplemental type certificate.

The 1735 recorded weather observation at North Central State Airport (SFZ), Pawtucket, Rhode Island, located about 8 miles southwest of the accident site at an elevation of 440 ft msl, included overcast ceiling at 800 ft above ground level, wind from 010° at 9 knots, visibility 9 statute miles, temperature 14°C, dew point 11°C, and an altimeter setting of 29.70 inches of mercury.

The six-seat, low-wing airplane was manufactured in 1990 and powered by a Continental IO-550-B6A, 300-horspower, six-cylinder engine driving a three-bladed propeller. The airplane's most recent 100-hour inspection was completed on June 5, 2015, at which time, the airplane had accumulated 2,143 total hours in service and 2,143 total engine hours, 410 hours of which were accumulated since the engine was overhauled in June 2011.

The engine maintenance logs noted that the oil filter was examined for metal particles during the most recent inspection as well as during five of the seven oil changes performed since the engine overhaul, and no metal or debris was found. The engine oil was also sampled and analyzed by a test laboratory three times since overhaul with normal results.

When the engine was overhauled, an overhauled primary alternator and a new parts manufacturer approval (PMA) alternator drive coupling were installed. The alternator was removed and replaced twice in October 2011, at 21 and 23 hours since engine overhaul, respectively. No logbook entries specifically mentioned the alternator drive coupling (a gear assembly installed on the alternator shaft that mates with a gear on the engine crankshaft to turn the alternator). According to a representative from the engine overhaul facility, the serial number of the coupling in service at the time of the accident matched the one that was installed at the time of overhaul. The coupling did not have a specified life limit; however, the instructions for continued airworthiness provided by the coupling manufacturer recommended that it be replaced at engine overhaul. The airplane was also equipped with a standby alternator installed in accordance with a supplemental type certificate.

The airplane impacted the corner of the roof and the backyard deck of a residence and came to rest upright, parallel to and about 15 ft from the rear wall of the home, oriented on a magnetic heading of 320°. The airplane and most of the residence were consumed by postcrash fire.

Flight control continuity was established from the flight control surfaces to the cockpit area; the left rudder cable and the left aileron cable turnbuckle exhibited fractures consistent with overload. The right flap actuator position was consistent with the right wing flap being in the retracted position. The left flap actuator was consumed by fire. The landing gear and actuators were found in the extended position. First responders reported detecting a strong odor of fuel when they arrived on scene.

Examination of the engine revealed that the crankcase was breached over the No. 6 cylinder barrel. Two additional puncture holes were found in line with the No. 1 cylinder connecting rod, located between the left and right magnetos. The oil sump was fractured and partially melted away. Fragments consistent with bearing material, connecting rods, lifters, and crankcase material were found in the oil sump and outside the engine crankcase.

During a follow-up examination of the engine, the alternator drive coupling (also commonly referred to as the alternator clutch, drive hub or clutch gear assembly) was found loose on the alternator shaft with its housing separated into two pieces (see figure 1). The coupling exhibited abnormal wear on the inside of the shaft collar and on the gear teeth. The shaft collar section was separated from the remainder of the housing on the back of the coupling (alternator side), around the circumference of the hole, which also exhibited wear consistent with the coupling rotating on the alternator shaft. Both sides of the thrust washer exhibited abnormal wear. The woodruff key was not found.

Figure 1. Accident alternator and coupling

The face gear that drove the coupling was separated from its crankshaft flange, and all four of the attachment bolts and remnants of their locking tabs were found in the oil sump. A portion of the crankcase adjacent to the face gear was damaged and worn away, consistent with contact from the back side of the face gear ring. Metal particles; bearing, piston, connecting rod, and crankcase fragments; and orange particles consistent with the alternator drive coupling's elastomer were found throughout the engine. Metal and elastomer particles were also found in the oil filter element. Four of the piston connecting rods were found separated from the crankshaft, three of which exhibited thermal discoloration and mechanical damage consistent with lack of lubrication and overheating. The two connecting rods that remained attached to the crankshaft (Nos. 3 and 4) would not rotate. Upon disassembly, the Nos. 3 and 4 connecting rod bearings were found mostly intact, and displayed discoloration and deformation consistent with thermal distress and lack of lubrication. The oil passages in the crankshaft that supply oil to the connecting rod bearings were absent of debris, however the No. 6 oil passage was partially obstructed by smeared journal surface material. All of the connecting rod journals sustained thermal discoloration, the most significant of which were Nos. 1, 2, and 6.

Coupling Inspection and Maintenance Procedures

According to the engine manufacturer's maintenance manual, the coupling and the face gear that drives it should be examined as part of the alternator inspection during the 500-hour engine inspection. The procedure included examining the coupling for damage or missing material and performing the "Alternator Drive Hub Slippage Inspection" or slip test. The OEM coupling may be reused indefinitely if it satisfactorily passes inspection and slip testing. The PMA coupling manufacturer recommended that it be replaced during major engine overhaul.

Alternator Failure Indication System

The airplane was equipped with a warning annunciator light labeled "LOW BUS VOLTS" that was designed to illuminate in the event of an alternator failure. Although the alternator may not fail, a coupling failure may affect or stop the rotation of the alternator, which would result in an indication of alternator failure in the cockpit. There is no method to differentiate a mechanical failure of the coupling, which could lead to catastrophic engine failure, from an electrical failure of the alternator system.

Alternator Failure Emergency Procedures

According to the Beech A36 pilot's operating handbook, the last step in the emergency procedures for an alternator failure with or without a standby alternator was to "land as soon as practical."

The Office of the Chief Medical Examiner, Commonwealth of Massachusetts, conducted an autopsy on the pilot. The manner of death was determined to be "accident."

The FAA Bioaeronautical Sciences Research Laboratory, Oklahoma City, Oklahoma, conducted toxicological testing of specimens from the pilot. The testing was negative for carbon monoxide, ethanol, and drugs of abuse. Ranitidine, an antihistamine used in the treatment of gastric acid secretion, was detected in the urine. This medication does not pose a hazard to flight safety.

A review of maintenance guidance materials and testing in the National Transportation Safety Board (NTSB) Materials Laboratory revealed two conditions that could result in the alternator drive coupling failure identified during examination.

First, using improper assembly torque values or procedures when tightening the alternator shaft nut will cause the coupling to progressively loosen over time due to abnormal wear. This may occur by not using a calibrated torque wrench, applying improper torque values, or using inadequate maintenance guidance materials. Six different sets of installation instructions for the coupling were found during the investigation: two from the alternator manufacturer, one from an unidentified repair station, one from the engine overhaul facility, and two (the overhaul maintenance manual and a superseding service bulletin) from the engine manufacturer. Although similar, each set of instructions had different steps and different levels of detail. For example, specific steps in the engine manufacturer's set of instructions were omitted in other instructions. Further, some steps were more clearly prescribed in one set of instructions than in another set. For example, the alternator manufacturer's set of instructions advised to torque the nut to within the range of 300 to 450 inch-pounds and install the cotter pin. Both the engine manufacturer's and the overhaul facility's set prescribed to torque the nut to precisely 300 inch-pounds and then to align the cotter pin holes. If the holes did not align at 300 inch-pounds, the instructions advised to continue applying torque until the holes aligned, up to the maximum value of 450 inch-pounds. The engine manufacturer's instructions also specifically stated not to reverse the nut when attempting to align the holes.

Only the engine manufacturer's set of instructions advised that the shaft/nut threads be lubricated with engine oil. That guidance was available peripherally in the preamble to an appendix in the engine maintenance manual that contained tables of torque values. Lubricated threads will result in a much higher "clamping force" holding the coupling against the alternator and resisting rotation on the shaft for a given torque value. None of the sets of instructions noted that the torque procedure and the resulting clamping force are the primary means of preventing the coupling from rotating on the shaft. Although the design of the coupling includes a woodruff key and channels, which are commonly used in other applications to prevent collars from rotating on shafts, that was not the intended purpose for this coupling. In this case, the woodruff key was intended to aid in the coupling installation process by arresting the alternator shaft during the torqueing procedure. The woodruff key was not designed to resist the loads on the coupling during normal engine operation.

Figure 2. Exemplar alternator, coupling, and components.

The second condition that could result in a coupling failure like that identified during examination is an out-of-tolerance coupling assembly. Specifically, if the gear section of the assembly is not fully seated in the housing and protrudes above the top of the shaft collar, the thrust washer will contact the gear instead of the shaft collar during installation of the coupling (see figure 2). As the nut is tightened against the washer, some of the applied torque will be consumed as the washer presses the gear back into the housing. As a result, less of the applied torque will be converted into the clamping force holding the coupling against the alternator. If that clamping force is insufficient, the coupling will progressively loosen over time due to abnormal wear.

This out-of-tolerance condition was observed in the NTSB Materials Laboratory on an exemplar PMA coupling and an exemplar original equipment manufacturer (OEM) coupling that had undergone improper "slip testing," which is a procedure the engine manufacturer prescribed to be performed anytime the coupling was removed and installed on an alternator (slip testing is also performed on all couplings at the manufacturer before delivery). The purpose of the test was to ensure that the coupling's elastomer section, which was designed to slip (allow the gear to rotate independently from the housing) in the event of an alternator shaft seizure, will not slip under normal conditions. The test required that the housing be held fixed and that a specified torque be applied to the gear using a special tool. During this test, if too much torque was applied and the elastomer started to slip, the gear section would slide up and away from the housing a small distance. If this improper testing occurred repeatedly and/or severely, the coupling will become out of tolerance. Currently, there are no published procedures to inspect or measure the coupling for this out of tolerance condition. The engine overhaul facility found two new PMA couplings that were out of tolerance as delivered from the manufacturer; the cause of their condition was not determined.

Similar Failures

A representative from the engine overhaul facility reported that he had observed several couplings from 2005 to 2015 that exhibited some type of failure or abnormal wear. The damage he observed was consistent with the coupling rotating on the shaft, slipping at the elastomer, or both. Specifically, he observed worn thrust washers, damaged or missing woodruff keys, damage to the key channels on the alternator shaft and the coupling, abnormal wear to the shaft collar section of the housing, separation of the shaft collar section from the housing, and disintegration of the elastomer. He was not aware of any effort to systematically track these failures or estimate the time in service of the damaged couplings; however, he notified the FAA principal maintenance inspector who conducted oversight of the overhaul facility whenever a failure was discovered. He indicated that most of the failures he had observed occurred in engines that were installed in twin-engine airplanes. He noted that the failures were often in the Continental Motors GTSIO series engines, in which the alternator rotated in the reverse direction from the type of engine installed on the accident airplane. He added that the facility only installed PMA couplings during overhaul but that he had observed failures in both the PMA and OEM couplings.

A review of records available in the FAA service difficulty reporting system database revealed 10 entries that may be related to an alternator coupling failure; however, the descriptions of the issue and level of detail available in those records varied widely and did not include any photographs.

A review of the engine manufacturer's warranty records for the 5 years preceding the accident revealed six claims in which the coupling was described as "failed," "damaged," or as having "come apart." The warranty records did not contain any photographs or sufficient detail to determine if the reported problems were similar to the failure of the accident coupling.