HISTORY OF FLIGHT On July 31, 2004, approximately 2005 Pacific daylight time, a single-engine Stinson 108 airplane, N97599, sustained substantial damage after a loss of engine power and forced landing while maneuvering near Graham, Washington. The aircraft was registered to and operated by a private individual. The certificated commercial pilot was not injured, while the sole passenger received minor injuries. Visual meteorological conditions prevailed and a flight plan was not filed for the personal flight, which was conducted in accordance with 14 CFR Part 91. The local flight originated from the Shady Acres Airport (3B8), Spanaway, Washington, at 2000. According to the Pilot/Operator Aircraft Accident Report (NTSB form 6120.1/2), the pilot reported that shortly after departing 3B8, at an altitude of approximately 1,000 feet above ground level, the engine failed. The pilot stated that he immediately tried [applied] carburetor heat, however, didn't notice any increase in RPM. The pilot reported that he then tried restarting the engine, switching to the left and right [fuel] tanks, but with negative results. The pilot further reported that when he decided he wasn't going to get the engine started, he selected a field for a forced landing. During the landing attempt the aircraft struck a fence and nosed over, resulting in substantial damage to the airplane's empennage and both wing tips. PERSONNEL INFORMATION At the time of the accident, the pilot held a third-class medical certificate dated July 9, 2004. The medical certificate carried no limitations or waivers. The pilot also possessed a commercial pilot certificate with single-engine and multiengine land ratings, and a flight instructor certificate for single-engine airplanes. The pilot reported his total time in all aircraft was 3,100 hours, with 2,900 hours as pilot-in-command. He also reported he had accumulated 30 hours in the accident aircraft make and model. AIRCRAFT INFORMATION N97599, a Stinson 108, serial number 108-599, was manufactured in 1946 and had approximately 3,390 hours on the airframe at the time of the accident. The airplane's most recent annual inspection was completed on June 20, 2004. The aircraft was equipped with a Franklin 6A4150 engine rated at 165 horsepower. Maintenance records indicate that the engine had accumulated 295 hours since its last major overhaul (SMOH). The aircraft's most recent airworthiness certificate was dated August 13, 1956. WRECKAGE AND IMPACT INFORMATION The aircraft came to rest inverted facing south-southeast in an open pasture approximately 2 1/2 miles southeast of 3B8 and 1 mile west of Graham, Washington, at coordinates 47 degrees 02.391 minutes north latitude and 122 degrees 18.891 minutes west longitude. A Federal Aviation Administration aviation safety inspector, who traveled to the accident site, reported substantial damage to both wingtips and the empennage of the aircraft. TESTS AND RESEARCH On August 4, 2004, representatives from the NTSB and the FAA conducted an examination of the aircraft at the facilities of AvTech Services, Kent, Washington. The initial observation of the fuel selector handle revealed that the selector position was oriented at the 12-o'clock position, which is midway between the left tank and right tank positions. To asertain whether the engine would continue to run with the fuel selector in this position, two separate tests were conducted using a Stinson 108 aircraft with a Franklin engine similar to the engine on the accident aircraft. One ground run test and one in-flight test were conducted. Both tests revealed that with the fuel selector positioned midway between the left tank and right tank positions, the engine continued to run with no change in engine parameters or smoothness (see attached reports). A further examination of the fuel system and related components revealed that all fuel lines were clear, the fuel selector handle, although stiff, turned through its full range of moverment, and the fuel selector valve revealed no obstructions in any of its ports. It was observed that when the fuel selector handle was positioned to either the left, right, or off position, the fuel port openings were correct. It was noted during the examination that approximately 90 percent of the circular metal area, which retains the primer in the locked position was missing (see attached photographs). Based on information provided by the pilot, it could not be determined if the primer was in the locked or unlocked position at the time of the accident. In an attempt to determine if the engine would quit should the primer inadvertently come out of the locked position, two separate tests were conducted using two Stinson 108 aircraft with Franklin engines similar to the engine on the accident aircraft. The first test was conducted during a static ground run, while the second test was conducted in flight. During both tests the primer was unlocked and allowed to back out of its locked position. In both instances the engines were observed to run rough; however, neither engine quit completely (see attached reports). On August 9, 2004, under the supervision of the IIC, both magnetos were bench tested at the facilities of Galvin Flying Service, Seattle, Washington. Results of the examination revealed no anomalies which would have prevented normal operations. On August 12, 2004, representatives from Precision Aviation Products Corporation, under the supervision of the IIC, examined the carburetor at Precision's facility in Marysville, Washington. Flow bench testing and disassembly revealed no abnormalities with the carburetor (see attached report). ADDITIONAL INFORMATION The pilot purchased the airplane on July 13, 2004. On July 14, 2004, while on the first leg of a ferry flight from Columbus, Ohio, to Spanaway, Washington, the pilot experienced an engine failure and landed the airplane on a highway median outside of Indianapolis, Indiana. The airplane was not damaged during the forced landing. Under the supervision of an FAA aviation safety inspector assigned to the Indianapolis Flight Standards District Office, Indianapolis, Indiana, an examination of the airplane's fuel system revealed that the right fuel tank was near full, while the left fuel tank was empty. The pilot stated that during the flight he had switched the fuel selector between tanks in a normal manner, and about 15 minutes after switching back to the left tank the engine quit (See inspector's report attached). The inspector then requested a local FAA certificated airframe and powerplant mechanic examine the airplane and report his findings. After examining the airplane the mechanic submitted a written report to the IIC, dated August 5, 2004, in which he reported that after inspecting the fuel system he replaced the carburetor float needle (P/N 233-614) and installed a new carburetor float kit (P/N 666-916). Additionally, maintenance records revealed that on July 16, 2004, the mechanic repaired the aircraft's fuel selector by replacing the selector shaft cotter pin and externally lubricating the shaft to free up the selector valve. The mechanic also reported that the tank gage switching was checked and no problems were noted with the component, nor were there any problems noted with the fuel selector throughout its range, other than being stiff. In a telephone conversation with the IIC, the mechanic related that he found no problems with the airplane which would explain the engine failure the pilot experienced on July 14, 2004. On August 6, 2004, the aircraft was released to Mr. Jeffery R. Poschwatta, Managing Member, AvTech Services, LLC, Kent, Washington.