Westlake Villag, CA, USA
N56484
Piper PA-34-200
The airplane collided with ground obstacles during a forced landing when the pilots could not maintain altitude after one propeller blade and part of the propeller hub from the right engine separated in flight. The CFI and the student practiced two simulated engine-out maneuvers. The CFI was reaching for the mixture control for a third simulation when he heard a loud bang. The airplane yawed violently to the right and began to lose altitude. The CFI turned toward his home field, but was having difficulty maintaining altitude. He decided that he would not be able to make it to the airport, and would have to make an off field landing. He landed in the only clear spot that he could see, which was a cemetery. The airplane collided with monuments and a wood chip pile before coming to rest next to a mausoleum. The right propeller hub fractured and separated. Investigators did not recover one blade and the corresponding part of the hub. The hub fractured as a result of fatigue cracking emanating from one of the grease-fitting holes. The fatigue cracking originated in the area of the grease-fitting threads closest to the interior surface of the hub, at the intersection between the grease-fitting hole and the chamfer. No material or manufacturing defect was found that could be identified as the source of the fatigue cracking. The fatigue cracking originated just outside the thread remnants of the grease fitting hole that were plastically deformed during a chamfering process. The grease fitting was located in a region where the wall of the hub was approximately at its thinnest, and where there was a slight change in the profile along the axis of the propeller. These geometrical factors would have led to high stresses at the edges of the hole. Chamfers were retrofitted to the grease-fitting holes by the manufacturer at both the interior and exterior surfaces of the hub in response to fatigue fractures that occurred in three-bladed propellers. This accident was the first occurrence of fatigue emanating from a grease-fitting hole for a two-bladed propeller. The chamfering was apparently introduced in an effort to smooth the stress distribution around the grease-fitting holes. The chamfer at the interior surface should have removed the stress concentration associated with the change in wall profile, but would also have decreased the amount of material in the wall, leading to higher stresses overall. The stress concentration associated with the grease-fitting hole and the chamfer, coupled with the location in a position where the hub wall was very thin, were likely sufficient to initiate fatigue cracking in this case. In more recently manufactured hubs, the grease-fitting holes have been relocated to positions where the hub wall is much thicker.
HISTORY OF FLIGHT On January 11, 2003, about 1100 Pacific standard time, a Piper PA-34-200, N56484, made a forced landing in Westlake Village, California, following a loss of power in the right engine. The Aero Club was operating the airplane under the provisions of 14 CFR Part 91. The certified flight instructor (CFI) and the private pilot undergoing instruction (PUI) sustained serious injuries; the airplane was destroyed. The local instructional flight departed Van Nuys, California, about 0945. Visual meteorological conditions prevailed, and no flight plan had been filed. The Federal Aviation Administration (FAA) accident coordinator interviewed the pilots. The CFI said that start, taxi, run up, and takeoff were uneventful. The lesson plan called for practice simulated engine-out maneuvers. The CFI said that he simulated engine loss of power by retarding the mixture control. Once the student identified the loss of power, the CFI simulated zero-thrust by setting the manifold pressure to 23 inches of Mercury, and the engine to 2,300 revolutions per minute (rpm). The PUI completed two simulated loss of power events. The CFI said that he was reaching for the mixture control for a third simulation when he heard a loud bang. The airplane yawed violently to the right and began to lose altitude. The CFI looked out of the right window at the right engine. He did not recall seeing a propeller either stopped or rotating. He was not sure that the engine was there. The PUI informed the CFI that he was unable to maintain control, and told the CFI to take the controls. The CFI turned toward his home field, but he was having difficulty maintaining altitude. He decided that he would not be able to make it to the airport, and would have to make an off field landing. He landed in the only clear spot that he could see, which was a cemetery. The airplane collided with monuments and a wood chip pile before coming to rest next to a mausoleum. PERSONNEL INFORMATION A review of FAA airman records revealed that the CFI held a commercial pilot certificate with airplane single engine and multiengine land ratings and an instrument airplane rating. The pilot held a certified flight instructor certificate with ratings for airplane single engine and multiengine land and instrument airplane. The CFI held a second-class medical certificate issued on August 30, 2003. It had the limitations that the pilot must have glasses available for near vision. No personal flight records were available for the CFI. The National Transportation Safety Board investigator-in-charge (IIC) obtained the aeronautical experience listed in this report from a review of the FAA airman medical records on file in the Airman and Medical Records Center located in Oklahoma City, Oklahoma. These records indicated a total time of 3,500 hours. A review of FAA airman records revealed that the PUI held a private pilot certificate with an airplane single engine land rating. The PUI held a second-class medical certificate issued on June 22, 2001. It had no limitations or waivers. A review of the PUI's logbooks indicated an estimated total flight time of 145 hours. He had an estimated 6 hours of dual flight instruction in this make and model. AIRCRAFT INFORMATION The airplane was a Piper PA-34-200, serial number 34-7350338. The logbooks contained an entry for an annual inspection dated July 26, 2002. The left tachometer read 257.0 at the last inspection; the right tachometer read 156.6 at the last inspection. The left tachometer read 513.4 at the accident scene; the right tachometer read 411.3 at the accident scene. The left engine was a Textron Lycoming IO-360-C1E6 engine, serial number L20922-51A. Total time on the engine was approximately 6,740 hours since a factory overhaul. The right engine was a Textron Lycoming LIO-360-C1E6 engine, serial number RL-1115-67A. Total time on the engine was approximately 6,400 hours since a factory remanufacture. The FAA accident coordinator obtained records, which indicated that Sky Trails at Van Nuys fueled the airplane with 51.9 gallons of 100 LL aviation fuel at 0935 on the morning of the accident. The coordinator interviewed the refueling operator who said that he filled the tanks. TESTS AND RESEARCH The IIC, the FAA accident coordinator, and investigators from Piper, Textron Lycoming, and Hartzell Propellers examined the wreckage at Aircraft Recovery Service, Littlerock, California, on January 15, 2003. The left engine separated from the airplane during the collision sequence. Retrievers cut wires and cables to remove the right engine. LEFT ENGINE Investigators slung the left engine from a hoist, and removed the top spark plugs. All spark plugs were clean with no mechanical deformation. The spark plug electrodes for all cylinders were gray in color, which corresponded to normal operation according to the Champion Aviation Check-A-Plug AV-27 Chart. A borescope inspection revealed no mechanical deformation on the valves, cylinder walls, or internal cylinder head. Investigators rotated the engine manually, but encountered mechanical damage and it bound after approximately 30 degrees of rotation. Investigators manually rotated the magnetos, and both magnetos produced spark at all posts. The fuel injector screen contained a fluid that smelled like aviation gasoline. The fuel manifold lines and nozzles were unobstructed. The oil filter and the oil suction screen contained shiny debris that was not attracted to a magnet. The No. 1 and No. 2 main bearings were unremarkable. The main bearing for the No. 3 journal did not have a tang and had a smooth polished appearance. It exhibited rotational scoring on the outer surface where it seated in the bearing saddle. The bearing had pits and was delaminating with areas that were missing material. The main bearing for journal No. 4 had pits. Left Propeller The left propeller was a Hartzell model HC-C2YK-2CGUF, serial number AU1440. Both propeller blades bent aft and exhibited trailing edge gouges. Both tips curled aft, and toward the low pitch position. The Hartzell representative examined the left propeller. He determined that the left propeller was rotating, and not feathered at the time of impact. He could not determine the amount of power. He noted no discrepancies that would have precluded normal operation. He did not inspect the propeller vibration dampers. RIGHT ENGINE Investigators slung the right engine from a hoist, and removed the top spark plugs. All spark plugs were clean with no mechanical deformation. The spark plug electrodes for cylinders No. 1 and No. 3 were gray in color, which corresponded to normal operation according to the Champion Aviation Check-A-Plug AV-27 Chart. The spark plugs for cylinders No. 2 and No. 4 were oily. The engine lay on its left side during recovery. A borescope inspection revealed no mechanical deformation on the valves, cylinder walls, or internal cylinder head. Investigators manually rotated the engine using a tool in an accessory drive socket. The crankshaft rotated freely, and the valves moved approximately the same amount of lift in firing order. The vacuum pump drive gear remained unbroken, and the vacuum pump turned freely. The fuel pump plunger moved up and down, and the gears in the accessory case turned freely. Investigators obtained thumb compression on all cylinders in firing order. The fuel pump's rubber diaphragm was unbroken and investigators blew air through the lines. The plunger in the fuel distribution valve moved freely, the rubber diaphragm was unbroken, and investigators did not observe any contamination. The fuel nozzles were open, and the screens were clean. The right magneto sustained mechanical damage, and investigators did not test it. Investigators manually rotated the left magneto and obtained spark at all posts. The oil suction screen and the oil filter contained shiny debris. A magnet attracted about 90 percent of the material in the oil filter. Debris filled about 50 percent of the oil suction screen. A magnet attracted some of the debris. The governor screen was clean. The Safety Board Materials Laboratory examined a sample of the debris. The specialist's report is part of the public docket. Pertinent parts of the report follow. The material from the right engine oil suction screen consisted for the most part of thin fragments with lateral dimensions ranging from about 0.02 inch to 0.5 inch. The fragments were generally of three different types: a shiny light-colored material of irregular shape, a darker material of irregular shape, or a darker material with a roughly rectangular shape. The specialist labeled three representative pieces showing the three different types as Sample A, Sample B, and Sample C. X-ray energy dispersive spectroscopy of Sample A revealed spectra having a major peak for aluminum with minor peaks for copper, tin, silicon, carbon, iron, and nickel. The spectra from Samples B and C were nearly identical, with a major peak for iron and minor peaks for lead, bromine, and carbon. Right Propeller The right propeller was a Hartzell HC-CYK-2CLGUF, serial number DN2543. The right propeller hub fractured and separated. Investigators did not recover one blade and part of the hub. The Hartzell representative examined the right propeller. Missing parts included: one complete blade assembly, part of the hub, the piston, feather spring, cylinder, start lock mechanism, and forward half of the pitch change rod. The remaining blade had no indications of rotation at the time of impact. He did not inspect the propeller vibration dampers. The Safety Board Materials Laboratory examined the hub remnant and prepared a factual report. Pertinent parts of the report follow; the complete report is part of the public docket. The propeller blades were Hartzell model FJC7666A, with serial numbers reported as J55500 and J55503. A fracture through both the forward and aft pieces of the hub led to the separation of blade J55500. Features on the fracture surface indicated that the fracture initiated as fatigue emanating from a grease-fitting hole in the aft piece of the hub. The fatigue cracking propagated from the grease-fitting hole in the forward direction to the boundary between the forward and aft pieces, and arrested. The fatigue cracking propagated in the aft direction from the grease-fitting hole just past the centerline of the hub. In total, the fatigue crack extended over approximately 60 percent of the aft piece of the hub. Fatigue reinitiated on the forward piece from the interior of the bolt hole inboard of the initial fatigue fracture plane, and propagated approximately 0.5 inch in the forward direction. The remainder of the fracture had features consistent with overstress. There are two grease fittings in each piece of the hub, and the overstress portion of the fracture also ran through a grease-fitting hole on the forward piece of the hub. The grease-fitting holes were threaded and chamfered at both the exterior and interior surfaces of the hub. Optical microscopy and scanning electron microscopy indicated that the fatigue originated at the surfaces of the threads closest to the interior of the hub, adjacent to the interior surface chamfer. The chamfering procedure caused some plastic deformation of the last partial thread remaining at the surface at both the forward and aft edges of the hole. The chamfer intersection crossed from one thread to the next in passing from the forward edge of the hole to the aft edge of the hole. The hub specifications required AMS-4133 (2014-T6) aluminum, with a minimum hardness of Brinell 125 (HB/10/500) and an electrical conductivity of 35 percent to 42 percent IACS at 68 degrees Fahrenheit. Hartzell reported that they manufactured the hub in 1982, and it would have been subject to a Hartzell service bulletin (HC-SB-61-213) requiring chamfering of the grease-fitting holes on both the interior and exterior surfaces. The chamfers were to be cut with a 45 degree tool to a diameter of 0.429 to 0.500 inch on the interior surface and a diameter of 0.304 to 0.357 inch on the exterior surface. The service bulletin required a surface finish of 63 RMS or better for the chamfers. Hartzell indicated that service bulletin HC-SB-61-213 is now obsolete, and the chamfering procedure has been incorporated into Manual 202A. X-ray energy dispersive spectroscopy at several locations on the fracture surface revealed spectra with a major peak for aluminum and minor peaks for copper, silicon, magnesium, manganese, and possibly iron, consistent with a 2014 aluminum alloy. Rockwell B hardness measurements averaged 84.5 HRB; conversion from the Rockwell B scale to the Brinell hardness scale results in an average hardness of 141 HB. The electrical conductivity averaged 36.5 percent IACS. The wall thickness at the location of the grease-fitting hole where the fatigue cracking initiated was approximately 0.41 inch. The center of the grease-fitting hole was approximately aligned with a change in profile in the wall of the hub (along the axis of the propeller blade). The specialist used an optical gauging microscope system to fit circles to the chamfer diameters at the interior and exterior surfaces of the hub. Measurements of the chamfer diameter at the interior surface of the hub averaged 0.426 inch, slightly below the specified range. Measurements of the chamfer diameter at the exterior surface of the hub averaged 0.315 inch, within the specified range. Visual and tactile comparison between the chamfer surface and a commercial exemplar indicated that the chamfer surface satisfied the specification for surface roughness. AIRFRAME The left wing remained attached to the fuselage; the right wing separated during the collision sequence. Wire cables connected the cockpit controls to the fuel selector valves in the wings. The cockpit control was in the off position for the left engine, and in the crossfeed position for the right engine. The retriever cut the left cable near the valve end; the valve was in the "off" position. The right cable pulled out; the valve was in the "on" position. The manufacturer's representative determined that the landing gear was in the up position. ADDITIONAL INFORMATION Neither of the pilots or the operator submitted a Pilot/Operator Aircraft Accident Report (NTSB Form 6120.1/2). The IIC released the wreckage to the recovery agent.
a fatigue crack in the propeller hub due to the inadequate design location of the grease fitting and the chamfering process.
Source: NTSB Aviation Accident Database
Aviation Accidents App
In-Depth Access to Aviation Accident Reports
