FLORENCE, SC, USA
N907ME
MARK EWART VELOCITY STANDARD RG
Following departure from an en route fuel and rest stop, the pilot reported a propeller problem and returned to the departure airport. The propeller was disassembled to repair what the pilot reported was a failure of the electrical pitch change for the propeller blades. According to witnesses, who saw and talked with the pilot as he worked on the airplane, he (the pilot) was upset over the failure, and he was anxious about hurrying to reach his destination. The pilot, who also held a homebuilt repairman certificate, did not follow the manufacturer's torquing procedures, which were to be followed after replacement of the propeller. After about an hour of flight, following the repairs, one propeller blade separated from the airplane, and a forced landing ensued. The airplane subsequently hit trees, then collided with the ground, inverted. Postcrash examination of the propeller indicated that a propeller blade had loosened during the flight, and subsequently it failed due to overload forces.
HISTORY OF FLIGHT On November 26, 1997, about 1544 eastern standard time, a Mark Ewart Velocity Standard RG, homebuilt airplane, N907ME, collided with trees and the ground during a forced landing near Florence, South Carolina. The airplane was operated under the provisions of Title 14 CFR Part 91, and visual flight rules [VFR]. Visual meteorological conditions prevailed. A VFR flight plan was filed, but not activated, for the personal flight. The private pilot and three passengers were fatally injured, and the airplane was substantially damaged. The flight originated at Lantana, Florida, about 0850, on the same day. An enroute stop was made at 1100 at Savannah, Georgia. The flight's final departure from Savannah was at 1436. A VFR flight plan had been filed and activated for the flight between Lantana, Florida, and Savannah, Georgia. At 1147 the pilot contacted Macon, Georgia, Automated Flight Service Station (AFSS), to cancel the flight plan for the flight between Lantana and Savannah. At the same time, he obtained a preflight VFR weather briefing for a flight from Savannah to Wilmington, Delaware. The investigation determined that the following sequence of events occurred, at the approximate listed time: 1100-N907ME landed Savannah; local controller reported radio difficulty with N907ME 1147-Flight plan canceled and preflight weather briefing obtained from Macon AFSS 1305-Pilot of N907ME called Savannah Tower Operations Supervisor to request "no radio" departure instructions 1321-N907ME departed Savannah 1331-N907ME about 5 to 8 miles northeast of Savannah, informed departure control that there was a propeller vibration; aircraft vectored for a landing on runway 36. 1340-Aircraft landed Savannah 1345(approximately)-Pilot talked to Velocity kit manufacturer on telephone reporting that the propeller pitch was stuck in the takeoff position and that he could not change the propeller pitch with the toggle switch. Pilot stated he intended to remove the propeller, adjust the pitch to a neutral position and continue his flight with a fixed pitch. 1345(approximately)-Pilot borrowed tools from fixed base operator, removed and replaced propeller blades using a torque wrench 1400-Pilot of N907ME called Operations Supervisor at Savannah Tower and requested "no radio" departure instructions from Savannah to Wilmington, Delaware. Operations Supervisor asked pilot why he had to land. Pilot stated "the propeller had stripped" and that he had worked on it and it was ok; pilot's call transferred to clearance delivery for departure instructions 1425-N907ME taxis to runway 36 1436-Flight departs Savannah; subsequently cleared on course 1534-While in contact with Florence, South Carolina, Tower, N907ME reports a problem and does not respond to additional inquiries 1536-N907ME requested directions to nearest airport and was vectored toward Florence, heading 190 degrees; position reported as 9 miles north of the Florence Regional Airport 1538-N907ME responded to inquiries from Florence Tower that he was "deadstick and no engine" 1544-N907ME observed to crash in trees north of airport; radar contact lost 1.5 miles north of Florence Airport The airplane collided with trees and the ground about two miles north of the Florence Airport. Initial examination of the airplane revealed that one blade of the three bladed propeller was absent from the hub. It was not located at the accident site. PERSONNEL INFORMATION According to Federal Aviation Administration records, the pilot held private pilot certificate, number 266179073, with airplane single engine land rating. His certificate was obtained on April 29, 1994. The application for the certificate indicated that the pilot had 50.8 total flight hours on that date. Based on the flight hour meter in the airplane at the accident site, and the aircraft records, the pilot's flight hours in the accident airplane were estimated to be 368. A pilot log was not located. Additionally, the pilot was issued a Repairman certificate, on April 30, 1996, with a rating for Experimental Aircraft Builder Aircraft-Velocity RG s/n DMO-266. A third class medical certificate was issued to the pilot on December 21, 1995, with no limitations or waivers. Additional personnel information is contained in this report on page 3 under the heading First Pilot Information. AIRCRAFT INFORMATION N907ME, a Velocity RG Standard, serial number DMO-266, was an Amateur-Built airplane certificated in the Experimental Category. The airplane was built and operated by the registered owner. An airworthiness certificate and operating limitations were issued for the airplane on December 23, 1995. The operating limits required that the airplane receive a condition inspection annually, in accordance with Federal Aviation Regulation Part 43, Appendix D, and the airplane determined to be in a condition for safe operation. The condition inspection was to be recorded in the aircraft maintenance records. An entry in the aircraft log dated January 26, 1996, indicated that the airplane had completed the operating limitations flight test hours and was found controllable throughout the range of speed and maneuvers intended, with no hazardous operating characteristics. The entry was signed by the owner/pilot. It was noted that there was no entry in the aircraft log indicating that a condition inspection had been performed within the 12 months prior to the accident. The airplane was of composite construction, and designed with a canard and a swept main wing. Winglets were installed on the main wing tips with integral one-way rudders. The airplane was powered by a Lycoming IO-360-C1E6 engine, and equipped with an Ivoprop Corp. three bladed, electrically controllable propeller. The propeller blades were of composite material. According to the manufacturer's information on the propeller, the pilot controls the propeller blade pitch through a toggle switch mounted in the cockpit, which activates an electric motor. Rotary motion of the motor is converted to axial motion by a lead screw. The lead screw is connected to a spool that is linked to a cam. Axial movement of the cam moves a torsion rod changing the pitch of the propeller blade. The propeller blades are sandwiched between knurled plates and held in place by torquing the mounting bolts to 65 foot pounds. According to Ivoprop Corp. the propeller blades should be replaced in the same relative positions if removed. If the propeller blades are removed, the mounting bolts should be torqued, the engine and propeller operated through the full r.p.m range, and the mounting bolts re-torqued. The procedure should be repeated until the mounting bolt torque remains the same. Stainless steel tape should be applied across the gap between the adjacent blade butts of the installed propeller blades, in accordance with Ivoprop Corp. Service Bulletin #3 (attached), and the tape inspection schedule established in the service bulletin followed. Service Bulletin #3 states that the tape should be inspected for breaks following a ground run and operating the propeller through the full r.p.m. range, then following flight of 15 minutes, 30 minutes, one hour, two hours, 4 hours, then at each preflight inspection. If the tape breaks, it may be reapplied and the inspection schedule re-commenced. If the tape breaks again, the propeller should be removed from service and the manufacturer contacted. Service Bulletin #3 states that its purpose was to detect blade movement inside the hub due to "misinstallation" , and or harmonic resonance between the propeller and the power plant. The bulletin indicates that if the blades were loosely mounted or a harmonic resonance developed the blades could move inside the hub in the plane of propeller rotation that could result in breaking of the propeller blade(s). Additional airplane information is found on page 2 of this report under the heading Aircraft Information. METEOROLOGICAL INFORMATION Weather information is located on page 4 of this report under the heading Weather Information. WRECKAGE AND IMPACT INFORMATION The wreckage was located in a heavily wooded area with dense undergrowth, about two miles north of the Florence Regional Airport. Broken tree limbs and airplane debris were scattered on the ground for about 95 feet, along a magnetic heading of approximately 051 degrees. Initial damage observed was to a tree, about 45 feet above the ground, and about 64 feet southwest of the main wreckage. The left winglet was found about 21 feet northeast of the tree, separated from the wing. Additional small airplane debris was located along the 051 degree heading leading to the main wreckage. About 15 feet northwest of the main wreckage, the right wing with attached winglet was found. The main wreckage consisted of the fuselage, remainder of the left wing, the engine, and the propeller. The fuselage was found in an inverted position. A crater was found in the ground located about ten feet southwest of the main wreckage. The crater contained the nose of the fuselage, and the top of the cockpit. The canard was found about 23 feet northeast of the main wreckage. Both wings were broken off about the mid span point, and had been split open spanwise separating the upper and lower surfaces exposing the foam filler that shaped their design. Both main landing gear were found in the retracted position. Rescue personnel reported that no shoulder harnesses were found, and that all seat belts were cut from around the occupants. Local law enforcement personnel reported a strong odor of fuel at the impact site. Foliage around the airplane was discolored brown. Control continuity to the pitch control, the canard, could not be established because of its separation from the airplane. Continuity also could not be established to the ailerons on the main wings, or the winglet rudders, because of wing separation. The engine and attached propeller were installed on the rear of the fuselage and mounted as a pusher propeller. One propeller blade of the three bladed propeller was not located with the wreckage. It was later found by local residents north of the main wreckage site. Pieces of the butt of the absent propeller blade were still retained between the knurled plates of the propeller hub with the remainder of the blade broken off. One propeller blade exhibited essentially no damage, and the remaining blade was broken and bent aft about 90 degrees. The wooden bulkhead to which the engine mount was fixed was punctured and abraded, directly forward of the fuel servo. The engine mount ring was fractured adjacent to the top left mount vibration damper. All of the engine mounting ears were scored and abraded. One hold down bolt and nut for the alternator was missing, and the second hold down bolt was loose, and the engine side of the starter ring gear was circumferentially gouged. The alternator cooling fan also exhibited evidence of rubbing. The engine and propeller were moved to a hangar for additional examination. There was mechanical continuity of the engine drive train. The lower spark plugs were Bosch Platinum and exhibited moderate wear. The single magneto impulse coupling was heard to snap as the engine was rotated by hand. The right magneto was replaced with an electronic ignition. The fuel servo mixture was found in the mid range and the throttle valve was closed. The oil screen exhibited some plastic type debris, but no metal contamination. The safety wire on the propeller bolts was cut, the propeller removed, and the knurled plates of the propeller disassembled, revealing the lead screw that provides propeller blade pitch change. The lead screw was extended about one inch, which according to the propeller manufacturer was consistent with a low pitch, high r.p.m. position. There was no evidence of the stainless steel tape across the propeller blade butt gaps described in Service Bulletin #3, and the knurled plates exhibited carbon like transfer that formed a heavy circle around each retaining bolt hole. A double triangular dark imprint was also noted on the knurled propeller plate. MEDICAL AND PATHOLOGICAL INFORMATION A post mortem examination of the pilot was conducted by the Florence County Coroner's Office, City-County Complex, Drawer F, Florence, South Carolina 29501. A toxicology examination of the pilot was conducted by the Toxicology And Accident Research Laboratory, FAA. The report of the examination indicated that carbon monoxide and cyanide analyses were not performed because of a lack of a suitable specimen. The tests were negative for ethanol and other drugs. TEST AND RESEARCH The propeller was examined in the Safety Board's Materials Laboratory. The report of the examination is attached. The propeller blade that was absent at the site of the main wreckage had smearing damage produced by the points of the knurl on the hub slipping along the blade surface finish. The direction of smearing was consistent with the blade moving mainly outboard (due to centrifugal loads) coupled with a slight rotation component as if the tip was lagging from drag loads on the blade (moving opposite direction of propeller rotation). There was no evidence of pre-existing failure to the metal pitch change lever arm. "The characteristics of the blade break appeared typical of an overstress separation as if the blade moved outboard against the attachment sleeves." ADDITIONAL INFORMATION The wreckage was released to the owner's insurance representative Jim Clark, Crawford and Company 121 Executive Center Drive Suite 220 Congaree Building, Columbia, SC 29221.
improper maintenance procedures used by the pilot/repairman, which resulted in the loosening and eventual failure of the propeller. Factors relating to the accident were: the pilot's self-imposed pressure to hurry to the destination, and trees in the forced landing area.
Source: NTSB Aviation Accident Database
Aviation Accidents App
In-Depth Access to Aviation Accident Reports
