FTW00LA042

During descent, the propeller separated from the experimental airplane. The pilot initiated a forced landing to a county road, and during the approach, the left main landing gear 'struck the bottom wire of a two wire service line,' which crossed over the road. The airplane 'stalled,' yawed 20 degrees to the left, and 'hit the middle of the road on all three [landing] gear.' It then headed off the left shoulder of the road where the right wing hit a road sign, shearing off the outboard 4 feet of the wing. The propeller system was not recovered. Examination of four of the six propeller attaching bolts revealed that they failed as a result of fatigue. The shank of one of the bolts, in the area located immediately adjacent to the fracture face had been subjected to 'extensive rubbing damage.' Two of the submitted bolt pieces contained washers. Examination showed that both washers were permanently deformed into dish shapes and were galled by rubbing. Rubbing and galling damage was also found on contact surfaces of the self locking nuts on all four bolts. The 3-bladed wooden experimental propeller had 47.4 hours since installation. The combination of low time since installation and rubbing damage on the bolts indicates the application of inadequate pre-stress (low torque) during propeller installation.

On December 12, 1999, at 1750 central standard time, a King Lancair 235/320 amateur-built experimental airplane, N717SK, owned and operated by the pilot, was substantially damaged during a forced landing following the separation of the propeller near Sterling City, Texas. The instrument rated private pilot, sole occupant of the airplane, was not injured. Visual meteorological conditions prevailed, and an IFR flight plan was filed for the 14 Code of Federal Regulations Part 91 personal flight. The IFR cross-country flight originated from Kerrville, Texas, at 1622, with Big Spring, Texas, as its destination.

The pilot reported that while at 10,000 feet msl and 70 miles from Big Spring, Texas, Fort Worth Center cleared him to descend at "pilots discretion to 4,600 feet." During the descent, at 6,500 feet, he heard a noise that sounded like a "magneto dropping out or a gear door coming open. There was no noticeable vibration in excess of the normal engine effects." He reduced power to 1,400 rpm and checked the magnetos and hydraulic system. All systems were "fine." The pilot further reported that one minute prior to hearing the noise, he detected a "smell like something was hot or burning. Nothing was apparent by visual observation. Thirty seconds after the noise began the sound ended abruptly. The only sound that you could hear then was the engine running extremely smooth." He looked at the tachometer and it was indicating 3,000 rpm. He reduced the power to 2,400 rpm and "realized that the propeller may have come off." He reported the problem to Fort Worth Center, and was advised of several nearby airports and ranch airstrips; however, he was unable to reach them.

The pilot stated that he turned to the northeast to look for a road to land on. He selected a county road, and during the approach, the aircraft's left main landing gear "stuck the bottom wire of a two wire service line" which crossed over the road. The airplane "stalled," yawed 20 degrees to the left, and "hit the middle of the road on all three [landing] gear." It then headed off the left shoulder of the road where the right wing hit a road sign, shearing off the outboard 4 feet of the wing. The airplane continued into an area between the road and a fence. As it crossed over a "road crossing" all landing gear were sheared off. The aircraft came to a stop upright 600 feet from where it had touched down on the road. The propeller assembly was not recovered.

The pilot reported that the Bruce Tiff/Collin Walker, 3-bladed wooden experimental propeller had accumulated 47.4 hours since installation.

Portions of four of the six propeller attaching bolts were removed from the engine and sent to the NTSB Materials Laboratory in Washington D.C., for examination. The examination revealed that all four bolts separated on relatively flat transverse planes. Examination of the fracture faces at low magnifications with the aid of a bench binocular microscope and then at higher magnifications with a scanning electron microscope (SEM) revealed ratchet marks and crack arrest positions. Both the fracture topography and fracture surface markings were typical of fatigue cracking. The shank of one of the bolts, in the area located immediately adjacent to the fracture face, had been subjected to "extensive rubbing damage." Two of the submitted bolt pieces contained washers. Examination showed that both washers were permanently deformed into dish shapes and were galled by rubbing. Rubbing and galling damage was also found on contact surfaces of the self locking nuts on all four bolts.