El Centro, CA, USA
N947A
North American Navion A
The engine lost partial power on takeoff and the pilot force-landed the airplane in a field. The airplane was involved in a similar accident 2 years prior. After being rebuilt, the pilot (who was also the aviation maintenance technician who rebuilt it) was relocating the airplane to a nearby airport. Post accident examination and testing of the fuel selector showed that the detent ring was bent and the air bypassed internally between the ports. With the detent ring bent, positive engagement of the fuel selector's position was difficult to ascertain. In addition, the shaft was worn. The type certificate holder for the airplane had released two Service Bulletins to replace the aging valves. In addition, the Federal Aviation Administration issued a Special Airworthiness Information Bulletin recommending aircraft owners test the valves for internal leakage and condition, and if necessary, replace them.
HISTORY OF FLIGHT On August 28, 2006, at 1006 Pacific daylight time, a North American Navion A, N947A (previously registered as N91729) experienced a loss of engine power after takeoff and made a forced landing in rough terrain near El Centro, California. The airplane was privately owned by the passenger and operated by the pilot under the provisions of 14 Code of Federal Regulations Part 91. The commercial pilot and pilot-rated passenger sustained minor injuries; the airplane sustained substantial damage. The flight originated from Douthitt Strip Airport (23CN), El Centro, at 1000, and was destined for Imperial County Airport, Imperial, California (KIPL). No flight plan had been filed for the local personal flight, and visual meteorological conditions prevailed. The National Transportation Safety Board investigator interviewed the pilot. The pilot reported that he was an aviation maintenance technician (AMT) with inspector's authorization. The airplane was one that he had been rebuilding over a period of 2 years, after it was involved in another loss of power during takeoff event on April 8, 2004 (NTSB Accident Number LAX04LA191). His maintenance facility was based at the departure airstrip and he had flown the airplane to and from the airstrip on 15-20 occasions. Since the completion of the rebuild, it had accrued approximately 17 hours. The purpose of this flight was to fly the owner of the airplane to Imperial where the owner would fuel the airplane and depart solo to Sacramento. The run-up was uneventful and all systems checked within limits. The pilot extended the flaps 15 degrees for takeoff from the 1,700-foot airstrip and applied power. The pilot stated that the first 500 feet of the runway is pavement and the remainder is a dirt surface. The airplane seemed to be operating normally when it reached the dirt surface portion of the airstrip. The revolutions per minute (rpm) were 2,650 rpm but he could not recall the manifold pressure. Rotation occurred about 1,200 feet down the runway and the airplane climbed to 30 feet. Once the airplane was airborne, the pilot felt the engine's available power decrease. He pitched the airplane for 80 miles per hour and began troubleshooting by switching fuel tanks, applying the fuel pump (he recalled positive fuel pressure), and leaning the mixture. However, the airplane would not climb or accelerate. As it continued to settle, the pilot set up for a forced landing and noted that the landing gear was in transition when the airplane touched down. The airplane came to rest in a field about 1/4-mile from the airstrip; the firewall was damaged during the forced landing. The night prior to the accident, the pilot and passenger had topped-off the airplane with fuel, and flew from Sacramento to Douthitt Airport, leaving an estimated 25 gallons in the airplane. The tip tanks were indicating empty and the pilot estimated approximately 2 gallons of fuel remained in each tip tank. Upon departure the following morning, the pilot selected the main fuel tank. TESTS AND RESEARCH On September 26, 2006, two inspectors from the San Diego Flight Standards District Office, and a representative from Sierra Hotel Aero, Inc., examined the fuel system and engine. A representative from JL Osborne, Inc. (the STC holder of the accident fuel valve), assisted in the examination and field testing of the fuel selector. The fuel selector was visually examined and no fuel staining was evident. When the fuel selector lever was manually operated by hand, it was noted that the detents were excessively worn at the tip tank positions. The detents were recognized at the "Off" and "Left Main" and "Right Main" positions, but worn. Fuel flowed freely when selected from each source. When the fuel selector was tested using shop air blowing into each port, the air bypassed internally between the ports. Air leakage through the stem of the selector was also noted. The fuel selector was retained for further testing. The gascolator was full of fuel and contained a minimal amount of sediment. The Teledyne Continental engine IO-470-D (serial number 79891-2-D) was examined. The induction system was free from obstruction and all of the scat hosing was intact and open. When the fuel line was disconnected from the flow divider, fuel was immediately apparent. The spark plugs were removed and the coloration and gapping were consistent with normal operation when compared with a Champion Aviation Check-A-Plug chart. Thumb compression was obtained on all cylinders and the ignition leads sparked upon manual rotation of the propeller. The throttle, propeller, and mixture controls linkages were continuous from the cockpit to their respective attach points at the engine. On January 9, 2007, the engine was operationally test run at Teledyne Continental Motors, Inc, with a representative of the Federal Aviation Administration present. According to the Teledyne Continental representative, the operation of the engine was normal and did not reveal any abnormalities that would have prevented normal operation and production of rated horsepower. On January 12, 2007, a Federal Aviation Administration Airworthiness inspector functionally tested the fuel selector using a vacuum pressure test. During the test about 19.5 inches of mercury were lost which exceeded parameters as specified by the type certificate holder for the airplane (maximum specified loss is 1 inch). The fuel selector was examined at JL Osborne facilities on January 31, 2007, by the NTSB investigator and a representative from JL Osborne. Compressed air was bled into the selector and a soap solution applied at each of the outlet ports. All outlets showed leakage, with a larger amount of leakage at the main tank position. The selector detents felt loose and did not click when positioned to each outlet. The fuel selector was disassembled. The fuel selector shaft is secured to the slug using a dowel pin. The slug then seats in the selector housing. The o-ring was soft and pliable and all washers and the spring were intact. The detent ring was bent. The dowel pin was not centered in the shaft and movement was obtained from side to side. The dowel pin was removed. When investigators reinserted the dowel pin in the shaft they rotated it and it locked into place. No grease was evident on the slug or base of the shaft. The components were greased and the test was performed again with the fuel selector components as installed during the initial test. A small amount of leakage was evident at the left tip tank position and the base of the shaft. The detent ring was replaced with a serviceable ring and the fuel selector was tested again. No internal leakage was noted. The selector was positioned to each outlet and when turned by hand, provided resistance and clicked into place at each respective outlet. According to the Sierra Hotel Aero representative, greasing of the valve in the field is not a recommended procedure. Neither JL Osborne nor Sierra Hotel Aero, Inc., have published procedures regarding field maintenance and/or overhaul of the fuel selector valve. A fuel sample was tested at Aviation Laboratories, Inc. There was no evidence of contamination, and the test results were listed as normal for 100 low lead (LL). On August 23, 2005, Sierra Hotel Aero, Inc., the type certificate holder for the airplane, published Navion Service Bulletin (SB) 101A (superceded Navion SB 101). The SB called for the mandatory removal and replacement of fuel selector valves. On November 29, 2005, the Federal Aviation Administration published a Special Airworthiness Information Bulletin (CE-06-11) advising owners of the worn out fuel selector valves and recommending functional testing and replacement, if necessary. ADDITIONAL INFORMATION All retained components were returned to Ryan's Aircraft, El Centro, California. No parts or pieces were retained by the NTSB.
The partial loss of engine power during takeoff, as a result of air being sucked into the fuel system due to a worn fuel selector valve stem and a bent detent ring.
Source: NTSB Aviation Accident Database
Aviation Accidents App
In-Depth Access to Aviation Accident Reports
