20061217029689A

(-23) Inspector statement: I was contacted by my supervisor/faa communication center on December 17, 2006, and was assigned to an accident that had occurred in the vicinity of Hoschton, Georgia, where a scharnhorst RV-4 experimental homebuilt airplane had crashed while attempting a forced landing to an open field with terraces. The pilot and his son reported minor injuries. I traveled to the accident site on December 17, 2006. I arrived at the crash site and was met by local law enforcement personnel. Initial examination of the crash site revealed the airplane collided with a terrace on landing roll out on a heading of 225-degrees magnetic. The airplane nosed over inverted and came to a rest on a heading of 045-degrees magnetic. There was no smell of fuel and no evidence of leakage. Atlanta air salvage recovered the airplane on December 18, 2006, for further examination. The tachometer at the crash site read 277.9 hours. Review of isis information revealed the pilot held a commercial pilot certificate with ratings for airplane single engine land and instrument airplane issued on March 22, 1993. In addition the pilot held a repairman experimental aircraft certificate issued on April 2, 2001. The pilot holds a third class medical certificate with no limitations issued on may 11, 2005. The pilot indicated on his last application for the third class medical certificate that he had accumulated 975 total flight hours. The pilot informed the NTSB during an initial interview that he had conducted a preflight inspection of the airplane earlier in the day and no anomalies were noted. No flight plan was filed nor was a weather briefing obtained for the VFR flight from cumming, Georgia, to winder, Georgia, and back to cummings, Georgia. The pilot privacy data omitted stated he was located in the front seat and his son was located in the rear seat and both were wearing seatbelt and shoulder harnesses. They departed cummings at 1030 EST and arrived in winder at 1045. They visited with friends and departed back to cumming at 1515. The pilot privacy data omitted stated he departed from runway 5, climbed to 1,000 feet AGL, made a left turn and climbed to 2,500 feet and circled his friend's house before departing direct to cummings. A short time later the engine quit. He attempted an engine restart with negative results and made a forced landing to an open field. On touchdown the airplane collided with a terrace and nosed over inverted. Examination of the airplane revealed the engine mounts were bent and the firewall was damaged. The upper and lower engine cowling were damaged. The propeller remained attached to the propeller flange. The composite spinner was destroyed. Neither propeller blade was damaged. The cabin area was not damaged except for the canopy, which was broken out and the roll bar was bent down. The throttle and mixture control was at mid range. The fuel selector valve was in the both position. The ignition switch was in the off position. The airflow performance high performance fuel metering systems purge valve was in the open position. Continuity of the flight controls was confirmed from the forward control stick aft to all flight control surfaces. The aircraft registration and airworthiness certificate was in the airplane. The left and right main landing gear remained attached. The aircraft logbooks were not located at the accident site and were provided to the FAA by the registered owner. The last condition inspection was conducted on June 14, 2006 at 251.7 hours. The airplane has flown 26.2 hours since the annual inspection. The engine was assembled by an airframe and power plant mechanic using a garage repair order on may 5, 1999, and had 0 hours since major overhaul. The engine and airframe has 340 total hours. Review of airflow performance inc., invoice no. 99-C4740 revealed a fuel injection system was shipped to the registered owner on July 30, 1999. The registered owner noticed a fuel stain on the engine block below the flow divider on November 22, 2002. The fuel system was tested for leaks and a bad leak was present in the flow divider. The flow divider was removed and sent to airflow performance on November 27, 2002. Review of airflow performance, inc invoice no. Y2-C6753, dated December 6, 2002 states: " 1 purge valve/flow divider (AP-486) assembly. Flow test flow divider, repair and flow test. 1 replace diaphragm" the work no. 3249, job progress sheet, dated December 4, 2002 states: "1 replace diaphragm and test." The description was signed off by cw. "Relock wire" the description was signed off by cw. The assembly was returned to the registered owner and reinstalled on the engine by the registered owner on December 15, 2002. The engine logbook revealed the engine/aircraft was stored from March 1, 2004, to may 10, 2005. The complete fuel injection system was sent to the manufacturer for recalibration in January 2005. Airflow performance work order 3597 dated January 13, 2005 does not indicate the lock wire was reinstalled and signed off by the technician with the initials kd as required by the systems installation and service manual. The complete assembly was sent back to the registered owner on January 18, 2005. The registered owner reinstalled the assembly on the aircraft on may 12, 2005. Review of airflow performance high performance fuel metering systems installation and service manual states on page 28, paragraph 3-41 purge valve installation procedures..." Leak check all connections and plumbing before starting engine. Lock wire all hardwire. Make sure to lock wire the stop screw." Appendix 1 of the purge valve installation states on page I-1, paragraph 2, "if removal of the valve is necessary make sure to relock wire the purge valve stop screw. Failure to do so will result in sudden stoppage of the engine if the screw backs out." A warning in airflow performance installation and service manual states, "your unit has been lock wired and sealed. Do not remove the seal as this will result in changing the calibration." The registered owner stated, "during installation or removal of the flow divider/purge valve assembly, it is not necessary to remove the safety wires in the purge valve mounting bolts or the stop screw. At no time did the builder ever remove the safety wire from the purge valve mounting bolts or stop screw." Airflow performance stated in an e-mail to the NTSB that "all components that left their shop were assembled, tested and tested in limits as well as lock wired. The customer had to install these components and hook up the appropriate linkages and hoses on the engine. The parts were received from the customer assembled. The only part we took apart to repair was the flow divider to install a new diaphragm on December 4th, 2002. All other parts were just tested as received." Airflow performance did not mention the maintenance activity that was performed on January 13, 2005, work order no. 3597 in their e-mail. The right wing remained attached to the airframe and was not damaged. The aileron was attached to its attachment points and the flaps were not extended. The right fuel tank was not ruptured. The right fuel cap was removed and had a tight seal. Fifteen gallons of fuel was present in the right fuel tank. The empennage was damaged about two feet forward of the vertical fin. The top of the vertical fin and rudder was crushed downward about six inches. The rudder assembly remained attached at all hinge points and the rudder balance weight remained intact. The left and right horizontal stabilizers were not damaged. The left elevator remained attached to the horizontal stabilizer and was bent upward. The left aileron trim remained attached to the left elevator. The right horizontal stabilizer was not damaged. The right elevator was not damaged and remained attached to the horizontal stabilizer. The right elevator was bent upwards and remained attached to the right horizontal stabilizer. The tail wheel remained attached to the empennage. The ELT had activated and was turned off. The left wing remained attached to the airframe and pushed aft. The wing tip was bent downward. The aileron was attached to its attachment points and the flaps were not extended. The left fuel tank was not ruptured. The left fuel cap was removed and had a tight seal. Eleven gallons of fuel was present in the left fuel tank. The engine run was conducted at atlanta air recovery on December 21, 2006. The engine run was conducted with the engine still installed on the airplane. The aircraft engine controls, propeller system, battery source, and electric fuel pump were used. An alternate fuel source was supplied. Several attempts to start the engine were unsuccessful. Further investigation revealed the airflow performance fuel injection purge valve stop screw was loose, not safetied, and had migrated out approximately one-eighth of an inch. The purge valve was removed from the system by rerouting the fuel line from the purge valve directly to the flow divider. The engine was started, ran at idle power, was advanced to a high power setting, and shut down with the mixture control. The airplane was released to atlanta air recovery on December 21, 2006. The FAA maintained custody of the aircraft logbooks and airflow performance fuel injection purge valve.