On September 17, 2007, about 1745 Pacific daylight time, an amateur-built experimental Root Sherpa, impacted runway 33 at the Scappoose Industrial Airpark, Scappoose, Oregon. The commercial pilot operated the borrowed airplane under the provisions of 14 Code of Federal Regulations Part 91 as a personal flight. The airplane sustained substantial damage. The pilot received minor injuries, and the pilot rated passenger was seriously injured. Visual meteorological conditions prevailed for the local flight, and no flight plan had been filed. According to the pilot it was a demonstration flight. He was interested in purchasing the airplane. He had flown the accident airplane the day prior to the accident with no problems, but with some concerns. He noted that the tail surface seemed "over-sensitive" and the airplane oscillated "very easily." The pilot reported that he was concerned about the low speed oscillations regarding the pitch attitude, as well as the need for "much" elevator down trim. In the pilot's submitted written report, he indicated that on base for runway 33, the airspeed was 60 miles per hour (mph) with 30 degrees of flaps. After turning on final, he input another 10 degrees of flaps, and reported that the indicated airspeed was between 55 to 60 mph, with some light to moderate turbulence present. The pilot stated that while on final, the sink rate began to increase for no apparent reason. He increased the throttle slightly to arrest the sink rate, which had a momentary effect. However, the sink rate began to increase again and did not stop when he applied power. At this point he was on short final, and he pulled the stick aft, which also had no effect on the sink rate or the pitch attitude of the airplane. Shortly thereafter, the airplane struck the runway in a nose down attitude. The pilot reported that the passenger had not been wearing his seatbelt. When the airplane hit the ground, the passenger came forward and struck the pilot's seat, pushing the pilot into the control stick and instrument panel. Federal Aviation Administration (FAA) inspectors responded to the accident site to examine the airplane. They noted the following damage to the flight control gimbal mechanism: internal corrosion with the presence of grease, oil, and dirt that was not as a result of the accident, paint peeling and corrosion under the paint in the area of the flight control gimbal mechanism separations. Their review of the maintenance history of the airplane revealed a previous accident history, and that the failure may have been caused during a previous event and had not been detected during repair. A review of the airplane logbooks revealed no entry for the installation of a new tail or wings. However, the FAA inspectors reported that a redesigned tail and wing sections were installed on the accident airplane prior to the accident. The purpose of the redesigned tail and wings were to lower the stall speed, improve slow flight handling characteristics, and to carry more fuel. TESTS AND RESEARCH The flight control stick gimbal mechanism and fuselage tube structure were shipped to the Safety Board's Materials Laboratory, Washington, D.C., for a metallurgical examination. According to the manufacturer's drawings for the airplane's elevator control system, when intact, the control stick is attached to the gimbal assembly, and a rod end is attached to the bottom of the gimbal assembly; these components are attached to the fuselage tube structure. A visual examination of the fuselage assembly by the FAA revealed that the control stick was not attached to the gimbal assembly, and the threaded portion of the rod end had fractured. The Safety Board's senior metallurgist also noted four tubes had fractured in the general area of the gimbal assembly that was part of the retaining bracket for the gimbal assembly. One of the tube's (identified as number 4 on the attached metallurgical report) showed a fatigue fracture that emanated from the inner surface at the root of the weld. The Safety Board's metallurgist reported that the fatigue had propagated through approximately 30 percent of the wall thickness. The fatigue crack was located near an area of the fuselage tube that contained a 0.7-inch-long region of incomplete weld penetration at the root of the weld.