History of Flight
On October 24, 2014, at approximately 0800 mountain standard time, an experimental amateur-built Winston W. Walker Thunder Mustang airplane, registration N244SW, was substantially damaged after colliding with terrain during a forced landing attempt following a loss of engine power about five nautical miles north of Globe, Arizona. The commercial pilot and owner, the sole occupant, sustained serious injuries. The flight, operated under 14 Code of Federal Regulations Part 91, departed La Cholla Airpark in Tucson, Arizona, around 0700. Visual meteorological conditions prevailed, and no flight plan was filed.
The pilot reported that the purpose of the flight was to practice instrument approaches. After completing the GPS 27 approach to San Carlos Apache Airport in Globe and executing the published missed approach procedure, he leveled off at 6,500 feet mean sea level. Shortly thereafter, he felt a power reduction as if the throttle was pulled back, and the rpm gauge indicated zero. Unable to maintain altitude to reach the nearest airport about 13 nautical miles to the southeast, he chose to land on what he described as "the least unfriendly ground." On short final, he flew under wires; the landing gear contacted a road and impacted a guard rail while the right wingtip struck a telephone pole. The airplane landed in brush-covered terrain, stopping immediately and coming to rest upright. The pilot noted the engine stopped with no advance indications or warnings.
According to a Federal Aviation Administration inspector, the airplane came to rest upright on an easterly heading in a dry riverbed covered with scrub brush. No postcrash fire occurred. The airplane was recovered to a secured storage facility for examination.
Personnel Information
The pilot held a commercial pilot certificate with single-engine land, multiengine land, and instrument airplane ratings. He reported a total flight time of 5,089 hours, including 353 hours in the accident airplane make and model. His most recent flight review was completed on October 7, 2014, and his most recent FAA third-class medical certificate was issued on August 4, 2014.
Aircraft Information
The airplane, serial number GITM025, was a kit-built scale model of the World War II North American P-51 fighter, completed in 2008. The fuselage was primarily composite material, with conventional flight controls actuated by pushrods for elevator and ailerons and cables for rudder.
It was equipped with a custom-designed 12-cylinder Falconer V-12 engine (serial number RFI 12018) producing 640 horsepower. The engine had accumulated 352.6 total hours, with 190.4 hours since its most recent overhaul and 14.6 hours since its most recent conditional inspection. Most engine accessories were mounted on or near the aft face of the engine, driven by three one-inch-wide serpentine belts. The engine had two MoTec M-48 engine control units and an Electronic Data Monitor MVP-50.
Postaccident examination of maintenance records revealed a December 16, 2009 entry at 149.3 engine hours: a rough flight with #11 cylinder cold due to a worn and loose exhaust valve pushrod, replaced with a factory new "long" pushrod. A subsequent January 9, 2010 entry at 153.4 hours noted replacement of a seal on the hydraulic pump with no leaks. The pilot reported that on January 14, 2010, during a local flight, the #12 cylinder went cold due to a rotated roller lifter causing a groove in the camshaft, leading to rocker arm and pushrod separation. The engine was removed and repaired at a marine shop, then reinstalled by July 30, 2010. The pilot stated he carefully followed the engine manufacturer's repair manual, including torquing three bolts holding the pulley and applying Loctite. All work was completed by October 19, 2010, and the airplane returned to service.
Meteorological Information
At 0751, the automated weather observation at Phoenix-Mesa Gateway Airport, about 42 nautical miles west-southwest of the accident site, reported wind calm, visibility 45 miles, sky clear, temperature 19°C, dew point 10°C, and an altimeter setting of 30.08 inches of mercury.
Wreckage and Impact Information
A postaccident examination of the engine on October 7, 2015, at Air Transport in Phoenix, Arizona, attended by NTSB, FAA, and engine manufacturer Ryan Falconer Industries, revealed that one of three bolts securing the main crank drive pulley to the damper had separated and was not recovered. The remaining two bolts were loose. Visual inspection showed a lack of Loctite, which the repair manual specified for each bolt prior to reassembly. The engine manufacturer representative stated the overhaul manual required Loctite application and torqueing each bolt to 35 foot-pounds.
Examination also found signatures on the main crank pulley's reluctor teeth indicating contact with the idler pulley, which would have caused the idler arm to separate, leading to two accessory belts becoming unsecure. This would dislodge both crank sensor plugs, interrupting ignition and causing complete engine stoppage; both sensor plugs were found dislodged. All 12 cylinders showed no indication of piston-valve collision, and all push rods were intact. Most spark plugs were wet and moderately corroded, consistent with contamination from first responders spraying the airplane and storage in a salvage yard for nearly a year. All 12 fuel injectors were clear of foreign material. The three accessory drive belts were not compromised.
Test and Research
The main crank drive pulley with two recovered bolts was sent to the NTSB Materials Laboratory. Visual examination found the pulley intact but with damaged attaching holes, labeled A, B, and C. Hole A was least damaged with slight ovalization and light thread marks. Holes B and C showed significant damage, including material removal from bolt head contact surfaces and large ovalization. The inner wall adjacent to holes B and C had marks consistent with bolt head contact, as if bolts were only partially threaded. The two bolts were intact with undamaged threads but had aluminum material trapped in thread roots; bolt head sides were polished, indicating relative motion. No Loctite was visually apparent, and Fourier Transform Infra-Red Spectrography showed material in threads not similar to Loctite 242. The pilot shaft on the engine side was worn unevenly and deformed, partially displaced from the pulley, consistent with tilting.
Three onboard electronic devices were recovered: an Electronics International MVP-50 engine data monitor and two MoTec M48 engine control units (ECUs). The EDM recorded parameters for the entire flight from 6:55:00 to 7:44:45, showing a sharp drop in oil pressure from 68 to 0 psi over three seconds at 7:43:06. The ECUs recorded data; ECU A ended at 7:43:00, and ECU B ended three seconds later.