HISTORY OF FLIGHT On August 30, 1999, approximately 1902 Pacific daylight time, a homebuilt Rotordyne 200 gyrocopter, N85Z, co-registered to/being flown by a commercial pilot, was destroyed when it collided with terrain following a loss of control in flight while maneuvering near the Crest Airpark, Kent, Washington. The pilot was fatally injured and a post-crash fire destroyed a portion of the aircraft. Visual meteorological conditions prevailed, and no flight plan had been filed. The maintenance check flight, which was reportedly for the purpose of balancing the rotor blades, was to have been operated under 14CFR91, and originated from the Crest Airpark shortly before the accident. A number of aeronautically experienced witnesses saw the accident sequence and their statements provided the following consistent observations: The gyrocopter departed runway 15 climbing to an altitude of approximately 300 feet above ground level (AGL) and maneuvered towards the west. The engine was developing power and the rotor blades were observed turning. The gyrocopter was then observed to execute a series of "pitch oscillations" as described by several witnesses, "oscillating" by another witness, and was reported by another witness to have "violently oscillated." Two other witnesses reported the maneuver as a "jump to the left fairley [sic] violently then back to the right" and it "rocked to the left then to the right." Most of the witnesses reported hearing a "popping" sound. All reported that the gyrocopter then pitched over and descended steeply to the ground (refer to attached statements). PERSONNEL INFORMATION The pilot possessed a Federal Aviation Administration (FAA) commercial pilot certificate with airplane single-engine land and sea ratings, as well as a multi-engine and airplane instrument rating. His last FAA medical (third class) was issued July 12, 1999, at which time the pilot reported a total of 2,000 hours of flight experience. The pilot was an engineer employed by the Boeing Company. The pilot sent an email to an associate on the morning of August 30, 1999, containing the following statement: "I did a little real work yesterday and then decided to play with the gyro instead. I reset the angle of attack of the blades and then flew it a couple of times. It does fly much better but now I have to rebalance it!" AIRCRAFT INFORMATION The aircraft was a Rotordyne 200, experimentally certificated gyrocopter (serial number 1). The two co-owners had made modifications to the basic design and the aircraft was issued a special airworthiness certificate on January 2, 1998. The third, and last, entry in the airframe log book showed an inspection being signed off at a total time in service of 56.4 hours. METEOROLOGICAL INFORMATION Visual meteorological conditions existed throughout the area at the time of the accident, and witnesses reported "choppy" or "gusty" winds from the west to south approximately 15 miles per hour. WRECKAGE AND IMPACT INFORMATION The aircraft crashed in a residential yard at 17648 SE 297th Place, Kent, Washington, and slightly north and west of the approach end to runway 33 at Crest Airpark. The approximate elevation of the accident site was 465 feet above mean sea level (refer to CHART I). On-site examination of the wreckage was conducted on August 30/31, 1999, by an inspector from the FAA's Renton, Washington, Flight Standards District Office after which the wreckage was verbally released to the surviving owner. The inspector reported that he verified control continuity for the gyrocopter's controls. Compression was established on all four cylinders of the Lycoming O-290-D2 engine, as well as rotation from the propeller throughout the engine, including the accessory gear section. The inspector also reported that he found all eight of the main rotor blade pitch adjustment bolts to be "loose", and "the bolts would rotate with the weight of a wrench", and that the bolts "could be rotated by hand, when pressure was relieved" (refer to ATTACHMENT FAA I). MEDICAL AND PATHOLOGICAL INFORMATION Post-mortem examination of the pilot was conducted by Paul D. Gosink, M.D., at the facilities of the King County Medical Examiner's Office, Seattle, Washington, on August 31, 1999, (case number KCME 99-0959). Toxicological evaluation of samples from the pilot was conducted by the FAA's Toxicology Accident and Research Laboratory, Oklahoma City, Oklahoma. All findings were negative (refer to attached TOX report). ADDITIONAL INFORMATION The main rotor head assembly was subsequently examined. The assembly was observed to consist of a center bar attached to the rotor drive shaft. Each rotor blade was connected to the center bar via a short attach bar (refer to photograph 1 and IMAGE I). The attach bars were connected to each end of the center bar by means of a block assembly. This assembly consisted of a metal block bolted to the upper and lower outboard edge of the center bar end, and a similar metal block bolted to the upper and lower inboard edge of the respective blade attach bar. Four bolts oriented perpendicular to the blade's lateral and longitudinal axes retained these four blocks. The blocked ends of the blade attach bar and outboard ends of the center bar were held in place with four additional bolts oriented parallel to the blade's lateral axis (refer to SCHEMATIC I). Additionally, the upper outboard shim plate of each blade attach bar had been marked with the following: "Pitch Bolts Torque 190 in. lbs" (refer to photograph 2). One of the four retaining block bolts from one rotor blade attach arm was observed to be sheared. The center third of the bolt was found within it's appropriate block. The head side of the bolt was later located and displayed a distinct lateral "smear lip" characteristic of lateral shearing separation (refer to photograph 3). Each outboard end of the center bar and inboard end blade attach bars was observed to have small (approximately one-quarter inch) holes drilled coincident with the centerline of the bar/blade's lateral axis. A metal pin was inserted into this hole about which the blade attach bar (and rotor blade) could then be rotated at varying pitch angles (refer to photograph 4). The four holes drilled in each of the retaining block bolts were observed to be slightly oversize when compared to the outside diameter of the bolts. This oversize condition allowed some free play of each bolt, thus enabling the mechanic to adjust the blade pitch over a range of several degrees (refer to photograph 5). Should these bolts not be torqued tight enough to retain blade pitch angle set during ground maintenance, the blade's pitch could change due to centrifugal forces to a flatter pitch.