Primary finding
Probable cause
A partial loss of engine power due to the reverse bending fatigue failure of the propeller hub bolts, which resulted from the bolts not fitting tightly into the propeller hub holes. Also causal was the pilot's decision to use improper bolts for this type of installation/operation.
Investigator assessment
Analysis narrative
The commercial pilot, who was also the owner/builder of the experimental amateur-built airplane, reported that, after departure and while climbing through 3,500 ft mean sea level, he heard the engine sound abruptly change and observed a loss of rpm and higher-than-normal exhaust gas temperature readings. The pilot turned back toward the airport and attempted to troubleshoot the problem; however, the airplane continued to produce less power than expected. The pilot established a normal traffic pattern for the runway. After turning to final, the airplane started descending and had insufficient power to reach the runway. The airplane subsequently impacted terrain short of the runway. Postaccident examination of the airplane revealed that all six propeller bolts that held the propeller hub to the drive hub were fractured. Further examination of the bolts revealed that the fracture surfaces all exhibited similar features consistent with reverse bending fatigue, likely from not fitting tightly into the propeller hub holes. The pilot reported that he had converted the engine and that, given its unique conversion, the bolts that were typically used to connect the propeller hub to the drive hub were too short; therefore, he chose to use the bolts that were on the hub at the time of the accident. Given the condition of the bolts and the pilot's statement, it is likely the bolts did not fit tightly in the propeller hub holes, which led to their failure due to reverse bending fatigue.
Source record
Factual narrative
On March 31, 2017 about 1655 Pacific daylight time, an OTT 601XL-B airplane, N610TT, the pilot executed a precautionary landing about one mile southeast of the Columbia Gorge Regional/The Dalles Municipal Airport (DLS), The Dalles, Oregon after the engine experienced a partial loss of engine power. The commercial pilot and one passenger sustained minor injuries, and the airplane was substantially damaged throughout. The airplane was registered to and operated by the pilot as a 14 Code of Federal Regulations Part 91 personal, local flight. Visual meteorological conditions prevailed at the time of the accident and no flight plan was filed. The pilot reported that the purpose of the flight was to verify the proper fuel mixture setting for the electronic mixture system. The airplane departed the airport to the northeast and while climbing through 3,500 feet, the pilot heard the engine sound abruptly change. Concurrent with the change, he observed a loss of RPM and high exhaust gas temperature readings. The pilot returned towards the airport and attempted to troubleshoot the problem, however, the airplane was producing less power than expected. The pilot established a normal traffic pattern for runway 31. After turning final the airplane was low, and despite the pilot adding power, the airplane impacted terrain short of the runway surface. A postaccident examination revealed no indications of catastrophic malfunction. The engine was rotated and compression was established on all cylinders. The spark plugs were removed and the cylinders were boroscoped; all cylinders exhibited normal operating signatures. The external timing marks at the rear of the engine were not consistent with the markings on the multi-toothed plate on the propeller shaft. Further examination revealed the bolts holding the propeller hub to the drive hub were all fractured. The bolts were removed and sent to the National Transportation Safety Board (NTSB) materials laboratory for further examination. The NTSB materials laboratory reported that the bolts were fractured in the 13th-15th thread root from the base of the bolt. The general features of the fracture surfaces were consistent with each other. The fracture surfaces exhibited opposite-facing flat thumbnail-shaped regions with a middle rougher region. The thumbnail regions exhibited crack arrest marks, which were orientated with propagation inward from the surface of the thread roots. The thumbnail regions also exhibited ratchet marks near the thread root surfaces, consistent with multiple crack initiation sites. These features were all consistent with fatigue in reverse bending. The pilot reported that the engine was a Corvair conversion engine built by him and a friend. They built this engine to be slightly different than other conversions, which allows it to be more powerful. The bolts that are normally used to connect the propeller hub to the drive hub were too short; therefore, he elected to use the accident bolts. He further reported that he believes he torqued the bolts properly, however, he noted that the holes on the propeller hub were not very tight.