MIA97LA222

During cruise flight, the pilot noted an engine vibration followed by a total loss of engine power. He made a forced landing, but was unable to reach suitable terrain. During the landing rollout, he observed a ditch and intentionally ground looped the airplane, but the airplane collided with a pole. Examination of the propeller revealed the propeller retention nut was hand tight. Galling was present on the propeller hub mounting flange and the rear cone. Disassembly of the engine assembly revealed the crankshaft was broken at the counterweight flange. Visual inspection of the shear surfaces on the crankcase showed tell tale benchmarks that appeared to have originated at the oil way plug threads and progressed through approximately one fifth of the crank circumference. Interviews with personnel involved with the installation revealed they had retorqued the propeller, but had not followed accepted industry practices in establishing proper hub to rear cone contact surfaces.

On July 28, 1997, about 1640 eastern daylight time, a Waco YMF5, N14128, registered to Waco Flying LTD, operated by Reilly Aviation, as a 14 CFR Part 91 sightseeing flight, experienced an engine vibration, followed by a total loss of engine power while in cruise flight. The pilot made a forced landing in the vicinity of Lake Buena Vista, Florida, and collided with a fixed object. Visual meteorological conditions prevailed and no flight plan was filed. The airplane sustained substantial damage. The airline transport pilot and two passengers reported no injuries. The flight originated from Kissimmee, Florida, about 30 minutes before the accident.

The pilot stated he was at 1,000 feet when the engine started to vibrate. He made a turn in the direction of a known short takeoff and landing airport (stol), and the engine quit. He maneuvered the airplane to the only available forced landing area (grassy area about 500 feet in length). On touchdown he observed a crossroad with a ditch. He intentionally ground looped the airplane to avoid a collision with the ditch, and the left horizontal stabilizer collided with a mono rail pillar.

According to the FAA, examination of the propeller assembly revealed the propeller retention nut was hand tight. Galling was present on the propeller hub mounting flange and rear cone. Disassembly of the Jacobs engine assembly revealed the crankshaft was broken at the counterweight flange. "Visual inspection of the shear surfaces on the crankcase showed tell tale benchmarks that appeared to have originated at the oil way plug threads and progressed through approximately one fifth of the crank circumference. The remainder of the shear surface exhibited a rough irregular surface with a dull gray coloration commonly associated with instantaneous shearing failures." The failed portion of the crankshaft was forwarded to the NTSB Materials Laboratory for examination.

Review of aircraft records revealed that Radial Engines, LTD, FAA Certificated Repair Station (CRS) No. RAQR194K overhauled and approved the engine for return to service on April 8, 1997, on work order 10208. The work order indicates that the internal engine components were magnafluxed by D & B Aircraft Engines, FAA CRS No. HX2R888K, and that a maintenance release for the non destructive inspection was issued on October 25, 1990.

"Surveillance of Radial Engines, LTD, has established that they are not equipped to perform Magnetic Particle Inspection. D & B aircraft engines is rated to perform maintenance on opposed aircraft engines manufactured by Lycoming, Continental, and Franklin only. D & B aircraft engines is not rated to perform Non Destructive Inspection, by the FAA."

Examination of the crankshaft by the NTSB Materials Laboratory revealed separation of the crankshaft was the result of fatigue cracking that initiated in the radius between the bearing surface and the forward face of the forward crankcheek. The fatigue cracking progressed through slightly less than half of the transverse cross section of the shaft portion of the crankshaft, turned and propagated through a portion of the one side of the crankcase before final separation. For additional information see, NTSB Metallurgist's Factual Report No. 98-30.

Interviews with personnel involved with the installation of the engine revealed they had retorqued the propeller, but they had not followed accepted industry practices in establishing proper hub to rear cone contact surfaces. A minimum of 70% contact for these surfaces is a required industry standard.

Discussions with industry personnel familiar with the overhaul and failure modes of small radial engines of this type, and Aircraft Propellers and Controls Training Manual Order No. EA-APC, established that there are basically three failure modes for crankshaft breakage at the crankpin, other than high cycle fatigue or material discontinuity. They are as follows:

* Improper indexing of the propeller to the crankshaft * A loosely installed propeller * An out of balance or out of track propeller

The crankshaft was released to Mr. Edward T. Reilly, President, Tom Reilly Vintage Aircraft on February 6, 1998.