WPR21FA114

The total loss of engine power due to the failure of the No. 6 exhaust valve for reasons that could not be determined based on available evidence.

Shortly after takeoff, the pilot reported to air traffic control that the airplane was experiencing engine problems and he was going to make an off-airport landing. The airplane contacted a semi-truck and a concrete barrier during the landing in a shipping harbor. Postaccident examination of the engine revealed that the No. 6 cylinder exhaust valve head was separated from the stem. A fatigue crack was found in the exhaust valve upper spring seat of the No. 6 cylinder assembly that likely led to the release of the exhaust valve. The crack formed at the base of the cup that housed the valve stem keys, causing a portion of the flange that supported the keys to separate. This could have allowed the exhaust valve to drop into the cylinder, impacting the piston, the cylinder head, and the exposed end of the valve guide before being sheared through the stem near the stem-to-fillet transition. The findings suggest that the No. 6 exhaust valve may have been floating, which would have removed the compressive load on the keys and allowed them to move and rub against the support flange and sidewall of the spring seat cup. The compression test of the exhaust valve springs indicated that they did not pass, suggesting that a floating valve is a possibility. However, the No. 5 exhaust valve springs also did not pass, and similar wear features around the key cup were not observed in that instance. Because the keys and valve stem were not found (likely destroyed during the continual operation with the failed valve), it could not be determine if there was some issue with either component that could have contributed to the failure.

The Lycoming O-540-E4B5, was the original engine installed on the airplane at manufacture and was overhauled in October 2010, 495 operating hours before the accident. The airplane’s last annual inspectionwas in August 2020, about 7 hours before the accident; all compressions were noted to be within operating limits. The engine was rebuilt in 2018, about 35 operational hours before the accident. The invoices from the rebuild showed the mechanic purchased 4 (2 sets) exhaust valve keepers; 12 thrust buttons (fitted at the ends of the rocker shafts); and a seal and gasket overhaul set. Despite numerous attempts, the Safety Board investigator was unable to interview the mechanic that performed that maintenance so it is unknown the reason for purchasing the keepers and which cylinders they were installed on. The accident site was located in a container shipping yards about 6.5 nautical miles southeast of Torrance. The wreckage was found distributed over about 200 ft on a median magnetic bearing of about 79°. Figure 4: Accident Site The first identified piece of debris was the outboard left wing that had come to rest in the truck cab. The hood of the truck, about 6 ft above ground level, had paint transfers, consistent with the tip of the left wing contacting cab in a left-wing low attitude. Adjacent to the truck was a 120 ft tall light post, which is where security cameras were located. Figure 5: First Identified Point of Impact There was an approximate 3-ft high concrete barrier 40 ft east of the truck; a 15-ft section was laying on its side with numerous paint transfers visible. About 40 ft from the barrier was the mid portion of the left wing with a majority of the aileron attached. The inboard wing was about 80 ft east of the barrier, which included the flap and left main landing gear. The fuselage had come to rest upright and was impact damaged the majority of which was near the pilots’ seats. The left forward cabin area was fragmented, and the floor skin was deformed aft. Control continuity was established by manipulating a cockpit control and observing the respective control surface move in response. The electric fuel pump remained mounted within the cabin floor and the external examination revealed no evidence of mechanical anomalies. The inlet line remained installed on the pump; however, the copper outlet line was pulled from the “B” nut. About one ounce of liquid consistent in appearance and odor with that of avgas remained in the pump. In the fuel was also a substantial amount of sediment and a submersing water-detecting paste indicated that water was present in the sample. Removal of the fuel screen revealed that it was brittle and covered in similar sediment and rust. Figure 6: Electric Fuel Pump The copper fuel line between the electric fuel pump and the firewall was fragmented and entangled in the deformed fuselage skin folds. The line was fractured from the firewall fitting. The hoses installed on the firewall forward in the engine compartment had bright orange torque stripes that were consistent with a recent installation. The engine driven fuel pump was removed and disassembled. The diaphragms were pliable, and no anomalies were noted. The fuel line was removed at the carburetor and it contained liquid that was clean and consistent in appearance and odor with that of avgas. The water detecting paste did not reveal evidence of water. Engine An external examination of the engine revealed that there were no holes in the crankcase and there was about 9 quarts of oil, as measured with the dipstick. Continuity from the throttle and mixture control in the cockpit to the arms on the carburetor was established. The controls were located in the full forward position and after moving the friction arm (relieving the tension), manipulation of the controls moved the respective arms from stop to stop. The oil sump drain had penetrated the cabin air scat tube. Oil was draining in the scat tubing to the carburetor airbox. Disassembly of the carburetor airbox revealed that the filter had sustained burn damage consistent with a carburetor fire. The lower cowling had corresponding soot marks at the vent. The carburetor was dissembled, revealing intact plastic floats and liquid resembling Avgas in the bowl. All of the upper spark plugs contained an oily film and were dark. The Nos. 1, 3, and 4 cylinder spark plugs did not have a gap between the electrodes. The No. 3 cylinder spark plugs were not damaged. Removal of the bottom spark plugs disclosed that the Nos. 3, 4, and 6 cylinders were damaged. The magneto timing was tested and found to be within the Lycoming recommendations A spark was visible on each lead when the propeller was rotated. Removal of the rocker box valve covers revealed that the exhaust rotator cap on the No. 2 cylinder was laying in the cover. The exhaust valve springs on the No. 6 cylinder were displaced and protruded out at an angle with the rocker arm contacting the upper valve spring seat (retainer) and the No. 6 exhaust pushrod was bent. Removal of the No. 2 cylinder revealed that there were numerous pieces of metal in the engine crankcase. Removal of the No. 4 cylinder revealed that the piston face had numerous gouges consistent with metal consumption in the barrel. There were pieces of worn metal found in the intake and exhaust pipes. Removal of the No. 6 cylinder revealed that the upper portion of the piston was trapped inside the barrel and dislodged from the wrist pin. Further examination of the No. 6 piston revealed it was fractured into multiple pieces, and impact marks were observed on the crown and the interior surface of the cylinder head. The exhaust valve guide had been extruded outward, and the cam follower on the No. 6 exhaust tappet body had been worn. The No. 6 exhaust valve was fractured through the stem, and the valve seat face exhibited a shiny metallic appearance. The No. 6 exhaust valve spring seats exhibited wear and fracture features. The upper spring seat flange at the base of the No. 6 exhaust valve spring seat had fractured along an approximate 225° arc segment, and the compression-spring-facing side of the seat exhibited striated features, consistent with fatigue crack progression. Figure 7: Cylinder No. 6 Figure 8: No. 6 Piston Head and Spring Seats Separation of the crankcase halves revealed numerous lifter (tappet) heads were broken and large pieces were found in the crankcase. The camshaft lobes and heels appeared to have normal wear with no spalling noted. The No. 6 and No. 5 cylinders and their respective components were taken to an engine overhaul facility to evaluate the wear on each component. The intake and exhaust guides on both cylinders over the new tolerance, but within the service limit. The intake valves on the No. 5 and No. 6 had 0.0007 and 0.0009 stem wear, respectively and were within the service limit. The exhaust valve on the No. 5 cylinder had 0.0011 stem wear and was within the service limit. The No. 5 And No. 6 inner and outer exhaust springs were compression tested and both were out of limits. A magnaflux inspection of the intake valve spring seats revealed cracks in both the upper and lower seats for the No.6 cylinder. The lower exhaust valve spring seat had cracks in the No. 5 cylinder. The No. 6 and No. 5 cylinders and their respective components were then sent to the Safety Board Materials Laboratory for analysis. A fatigue crack had formed in the exhaust valve upper spring seat of the No. 6 cylinder assembly, which is consistent with it eventually leading to the release of the exhaust valve. The crack formed at the base of the cup that housed the valve stem keys. Under typical operation, the exhaust valve is either on the valve seat or off the valve seat. When the valve is on the valve seat, the compression in the valve springs push against the underside of the spring seat, and the keys are