LAX98LA033

The pilot reported that during the preflight inspection, he examined the propeller while pulling the engine through each cylinder and found no discrepancies. The aircraft was about 7 minutes into a 15-minute flight when he saw a 'spark or flash' at the 11 o'clock position on the nose, followed quickly by the complete separation of the propeller. He turned the aircraft back toward the airport but had insufficient altitude and landed in a hilly area of a canyon. A postcrash fire began and consumed the aircraft aft of the firewall. The aircraft was examined on site and following recovery of the wreckage. An unsuccessful search was made for the wooden propeller. The metal propeller hub remained attached to the engine crankshaft with fragments of the wood center section intertwined in the bolts. The wood remains were insufficient to determine either the model or serial number of the propeller, or the condition of the boltholes. Both the engine crankshaft splines and propeller hub splines were unremarkable. The remaining pieces of wood from the hub were submitted for laboratory evaluation. The examining laboratories reported that there was insufficient evidence available from the small fragments submitted to identify an initiating event in the propeller or the failure cause of the wood hub.

On November 8, 1997, at 1212 hours mountain standard time, a Waco YMF-5, N369DP, sustained an in-flight separation of the propeller during cruise flight and forced landed on a hillside near the mouth of Oak Creek Canyon in Sedona, Arizona. The aircraft was owned by Red Rock Biplane Tours and was being operated on a 14 CFR Part 91 local area for-hire sightseeing flight. The aircraft was destroyed during the impact sequence and postcrash fire. The commercial pilot and one passenger sustained minor injuries; however, the remaining passenger sustained serious injuries. Visual meteorological conditions prevailed.

In written and verbal statements, the pilot reported that after greeting the passengers who were to take this tour, he inspected the aircraft. During the inspection, he examined the propeller while pulling the engine through each cylinder and found no discrepancies. The aircraft was about 7 minutes into a 15-minute flight when he saw a "spark or flash" at the 11 o'clock position on the nose, followed quickly by the complete separation of the propeller. He turned the aircraft back toward the airport but had insufficient altitude and landed in a hilly area of a canyon. A postcrash fire began and consumed the aircraft aft of the firewall.

The aircraft was examined on site and following recovery of the wreckage. An unsuccessful search was made for the wooden propeller.

The metal propeller hub remained attached to the engine crankshaft with fragments of the wood center section intertwined in the bolts. The wood remains were insufficient to determine either the model or serial number of the propeller, or the condition of the boltholes. Both the engine crankshaft splines and propeller hub splines were unremarkable.

The remaining pieces of wood from the hub were submitted to the Sensenich Wood Propeller Company, Borden Chemical Company (manufacturer of the glue), and to the U.S. Department of Agriculture's Forest Products Laboratory for evaluation. Reports of the examinations are attached. The glue bond quality was found to be consistent with the manufacturer's instructions. The remaining fragments were not large enough to conduct a meaningful shear strength evaluation.

According to the Forest Products Laboratory report, evidence of hub overheating was found in the fragments submitted. No failures were identified in the glue bond lines. The report notes that excessive heat induces a loss of strength in the wood fibers. The varnished face of the hub was not flat, and was not making full contact with the metal hub retainer plate. In conclusion, the report stated that there was insufficient evidence available from the small fragments submitted to connect the noted defects with an initiating event in the propeller or the failure of the wood hub.

The aircraft maintenance records were examined for entries pertaining to the propeller. The last logbook entry concerning installation of a propeller was dated June 22, 1994, and documented the installation of Sensenich wood propeller model W96JA-4-68, serial number AB7725. The next entry dated August 1, 1996, noted the removal and replacement of eight propeller hub bolts and nuts. An invoice from General Aviation, Inc., was supplied by the operator documenting the purchase of propeller W96JA-4-68, serial number AC8560, on October 30, 1996. The operator stated that this propeller was installed on the accident aircraft on November 8, 1996; however, no logbook entry to this effect was found.