Primary finding
Probable cause
A loss of engine power due to a malfunction of the turbocharging system likely due to contaminated oil. Also causal were the pilot's decision to continue the takeoff although the airplane was not performing normally and his failure to maintain adequate airspeed following the loss of engine power, which resulted in the airplane exceeding its critical angle of attack and an aerodynamic stall. Contributing to the accident was the engine manufacturer's inadequate guidance regarding inspection and maintenance of its turbocharged engines.
Investigator assessment
Analysis narrative
The commercial pilot was departing in the turbocharged airplane to go to another airport and pick up the owner of the airplane. He contacted the air traffic control tower and received instructions from the controller to taxi to the active runway and hold short. The airplane taxied to the designated location and remained there for about 5 minutes. During this time, a student pilot heard the airplane's engine cycle from near idle to full power about five times and reported that the engine did not "sound right." The pilot requested and received clearance to takeoff, and, shortly after becoming airborne, advised that he had a "problem," declared an emergency, and requested to "return to the field immediately." The controller cleared the pilot to land on any runway, and the pilot reported that he was unable to maintain engine power. There were no other communications from the airplane. Review of security camera video revealed that the airplane was slow to accelerate and did not rotate until about 1,800 ft down the 4,552-foot-long runway from the point where the pilot initiated the takeoff roll. Once airborne, the airplane began to pitch slightly up and down while remaining in ground effect. Considering that the pilot was the only occupant of the six-seat airplane, the airplane should have become airborne much sooner. Further, there was adequate runway remaining at the point of rotation for the pilot to abort the takeoff and stop on the remaining runway. However, the pilot elected to continue the takeoff. The airplane climbed slowly, momentarily reaching an altitude that was just above the trees that surrounded the airport, then began to lose altitude, and turned left about 90°. The airplane then disappeared from view of the camera, and a smoke cloud was observed to rise from behind a tree line. Witnesses who observed the airplane just before impact saw the airplane gliding toward the ground "in slow motion" and heard no noise coming from the airplane. The witnesses reported that the airplane then rolled into a steep left bank, entered a nose dive, and exploded when it hit the ground. The witness observations were consistent with the pilot failing to maintain adequate airspeed, resulting in the airplane exceeding its critical angle of attack and an aerodynamic stall. Examination of the wreckage revealed signatures indicating that the propeller and the turbocharger's turbine wheel were not rotating during the impact sequence, which is indicative of a loss of engine power. The spark plug electrodes displayed evidence of black sooty deposits indicative of carbon fouling. The carbon fouling could have been the result of failure of the turbocharging system, which can result in an overly rich mixture condition so severe as to cause a total power failure. Examination of the turbocharging system revealed that it had been heavily damaged by the postcrash fire, and only the turbocharger and wastegate were recovered. Examination of the turbocharger revealed that the turbine and compressor wheels, which were interconnected by a shaft, could not be rotated by hand as the shaft had partially fused to the bearings likely as a result of exposure to the postcrash fire. The bearing radial holes were clear, and there were no excessive or abnormal scoring marks on the bearings as would be expected if they were contaminated, distressed, or subject to prolonged oil starvation. There was also no coking of oil in the turbocharger body that would have prevented lubrication of the bearings, and no definitive rotational rub marks that would have suggested excessive bearing wear or imbalance. Examination of the wastegate also did not reveal any anomalies, and the wastegate valve was free and could move through its full range of motion. The wastegate actuator body had been mostly consumed by the postcrash fire; only the valve housing assembly, actuator shaft assembly, springs, and retainer remained. X-ray examination of the oil supply line check valve, which was located upstream from the turbocharger and regulated the supply of oil that it received, showed that instead of being straight, the internal spring was slightly cocked about 5°. Review of the manufacturer's specifications revealed that no check valve leakage was allowed below 8 psi of oil pressure. However, flow testing of the check valve revealed that oil leaked from the check valve exit hole before 1 psi of pressure was reached, which indicated that the check valve was likely not preventing oil from draining into the turbocharger after shutdown and was pooling in the turbocharger body. During further examination of the check valve using computed tomography scanning and radiography, a small gap was found between the ball and the internal channel along the neck. Sectioning of the check valve revealed that the angled spring and the small gap between the ball and the internal channel were due to the presence of contamination in the internal channel on the upstream (inlet) side of the check valve and the presence of foreign material between the ball and the internal channel along the neck. The presence of contamination in the check valve indicated that contamination was likely present in other components of the turbocharging system. Because the controller and the wastegate use engine oil and pressure for operation and control of the turbocharger, if either one is contaminated, system performance can be compromised. Maintenance records indicated that two repairs requiring replacement of major components of the engine took place about 2 years before the accident. The first repair occurred following a report by the owner of high oil consumption, and it entailed replacing a cracked air/oil separator, leaking oil dipstick gaskets, a leaking fitting on the turbocharger wastegate actuator, and the turbocharger "due to oil leaking past shaft seal intake system." The second repair occurred about 4 months later, when the owner again reported high oil consumption. This resulted in replacement of the Nos. 3, 5, and 6 cylinders because the oil control rings stuck in the pistons of these cylinders, which indicated debris had been deposited in the ring grooves. Although these repairs provided evidence that suggested the oil system was contaminated, the maintenance records did not show that any reused oil lines, the turbocharger oil supply line check valve, or turbocharger system components such as the controller and wastegate were flushed. Further, review of the engine manufacturer's guidance revealed that it did not include instructions for checking or replacing the check valve during inspections, flushing of any reused oil lines, the check valve, and components such as the turbocharger, controller, and wastegate whenever a turbocharger leak was detected, following an engine test run after cylinder replacement, after replacing lubrication system components, or when doing any type of maintenance where contamination or foreign debris could be introduced into the system. If the engine manufacturer had included these instructions and the mechanics had performed actions such as flushing the check valve and turbocharger system components following either of the two engine repairs, it is likely the contamination found in the check valve (and likely present in other components of the turbocharging system) would have been removed. The presence of contamination in the check valve, the airplane's maintenance history, and the carbon fouling of the spark plugs, strongly suggest that the engine lost power due to contaminated oil compromising the performance of the turbocharger system. The National Transportation Safety Board asked the Federal Aviation Administration in 2008 to require manufacturers to amend their pilot operating handbooks (POHs) to include emergency procedures for turbocharger failures (Safety Recommendation A-08-21). However, the FAA did not take this action, and review of the POH for the airplane revealed that i
Source record
Factual narrative
The Office of the State Medical Examiner, State of New Jersey, performed an autopsy on the pilot. The listed cause of death was blunt impact injuries. The FAA Bioaeronautical Sciences Research Laboratory, Oklahoma City, Oklahoma conducted toxicological testing of the pilot. The specimens were negative for carbon monoxide. Carvedilol, which is used to treat high blood pressure, was detected in urine and blood, and quinapril, which is used to treat hypertension, was detected in urine. Use of these two non-impairing drugs was previously reported by the pilot to the FAA. Salicylate (aspirin) was detected in urine. Review of FAA medical certificates and supporting documentation indicated that the pilot had a history of high blood pressure and a myocardial infarction with stent placement in 2007. Based on clinical reports, his conditions were stable, and no significant conditions were identified during his FAA physical examinations. The recorded weather at CDW, at 1012, about 10 minutes after the accident, included: variable winds at 3 knots, 10 miles visibility, clear skies, temperature 28°C, dew point -17°C, and an altimeter setting of 30.13inches of mercury. According to FAA and pilot records, the pilot held a commercial pilot certificate with ratings for airplane single-engine land and instrument airplane, a flight instructor certificate with a rating for airplane single-engine, and a ground instructor certificate with an advanced ground instructor rating. His most recent FAA third-class medical certificate was issued on March 30, 2015. He had accrued about 1,941 total hours of flight experience, 16 hours of which were in the accident airplane make and model. The accident site was located in a wooded area about 0.3 nautical mile from the departure end of runway 22 on a magnetic heading of 156°. Examination of the accident site revealed that the airplane impacted terrain on a 20° magnetic heading after striking several trees. Further examination revealed that the airplane impacted in a nose-down, inverted attitude. During the impact sequence, the engine separated from its mounting location. The empennage was displaced about 20° to the left of the fuselage centerline and was partially separated from the aft fuselage. The wings were separated from their mounting locations. The fuselage came to rest upright on a 100° magnetic heading against the base of a tree. The majority of the fuselage was consumed by a postimpact fire. The wing flaps were found in the up position. The elevator trim was near neutral. The fuel selector valve was in the "BOTH" position, and there was no evidence of fuel blockage. Control continuity was established from the ailerons, elevator, and rudder to the flight controls in the cockpit. There was no evidence of any inflight structural failure. Examination of the propeller revealed that one of the three blades separated during the impact sequence and came to rest about 30 ft from the rest of the propeller assembly. The propeller blades did not display evidence of propeller rotation during the impact sequence. Examination of the engine revealed that the rear of the engine had been heavily damaged by the postcrash fire, and the magnesium oil sump was destroyed by fire. The engine's fuel system was heavily damaged by the postcrash fire. The engine-driven fuel pump was destroyed by the postcrash fire. The fuel servo inlet fuel screen was free of contaminants, and the diaphragm displayed thermal damage. The fuel flow divider was thermally damaged. The left magneto's internal windings were found in the molten metal beneath the engine. The right magneto was found loosely attached to the rear accessory housing and was thermally damaged. Attempts to rotate the engine drive train by hand were unsuccessful. There was no evidence of any type of blockage in the intake or exhaust systems. The spark plug electrodes displayed evidence of black sooty deposits indicative of carbon fouling. A portion of the No. 3 piston's skirt was missing; metal fragments were present in the crankcase, and metal was found extruded from the edges of the No. 3 main engine bearing. The turbocharger system was partially destroyed by the postimpact fire; a majority of the damage was to the compressor housing and compressor wheel. The turbine housing exhaust port did not display evidence of turbine wheel rotation during the impact sequence. The turbocharger was secured to its mount with the exhaust pipes separated from the exhaust bypass valve on both ends. The slope controller and the pressure relief valve were destroyed by the post impact fire, and the exhaust bypass valve was partially destroyed but remained attached to the turbocharger. The airplane was a 6-place, single-engine, high-wing monoplane of conventional metal construction. It was equipped with fixed-tricycle-type landing gear and was powered by a turbocharged, 310-horsepower, Lycoming TIO-540-AJ1A engine, driving a three-blade, McCauley, controllable pitch propeller. According to FAA and maintenance records, the airplane was manufactured in 2009. Its most recent annual inspection was completed on April 17, 2015. At the time of the inspection, the airplane and engine had accrued 1,155.4 total hours of operation. According to the maintenance provider who had maintained the airplane since December 2011, anything that bothered the owner about the airplane would get fixed. Most of the items that were addressed by the maintenance provider were cosmetic or routine maintenance, such as oil and filter changes, gauges, starter replacement, lights, accessories, battery replacement, and compliance with airworthiness directives and service bulletins. The maintenance provider reported that the owner's landings could be a little rough, so they had also replaced some tires as he had experienced a few flat tires, and, as a result, the owner would keep a spare set in the airplane in case he blew a tire on landing. Review of maintenance records revealed that the airplane's engine had been receiving regular oil changes since new as well as spectrometric oil analysis. Review of oil analysis reports provided by the maintenance repair organization indicated that a sample of the engine's oil that was taken on March 5, 2012, contained elevated levels of iron, nickel and chromium. Another sample taken on December 19, 2013, contained elevated levels of aluminum, chromium, iron, and nickel. In a report dated March 9, 2015, the laboratory commented about an oil sample that had been taken on March 4, 2015, stating that: "These numbers are a lot easier to take than the high aluminum, chrome, iron, and nickel we saw last time. The shorter oil run obviously helped, but most of the metals are lower on a ppm/hour basis too, meaning that the engine really did wear better. If anything, nickel could still stand to be lower. 13 ppm is almost high enough to get a mark, so that's one we'll be monitoring next time. There's a trace of fuel to report this time, but that's not anything to worry about. It's likely just from normal use. Much better at 1,151.6 hours S[ince ]New." In a report dated August 12, 2015, for an oil sample that was taken on August 4, 2015 (11 days before the accident), the laboratory commented that: "Steady as she goes for this sample out of N63TV. If we're being picky you could say that iron should have come down as a result of the shorter oil run, but 39 ppm isn't bad at all for one of these engines after 20 hours on the oil. Everything else is in good shape, so we'd be surprised if the extra iron on a per-hour basis turned out to be an issue. No problems with the oil itself were found, making for a very nice report overall." Maintenance records indicated that two repairs requiring replacement of major components of the engine had been accomplished. The first repair followed a report from the owner that the engine was experiencing high oil consumption. According to a m