MIA02LA107

Just after takeoff, at an estimated altitude 200 feet, the airplane's engine lost partial power. The pilot checked the throttle for full power, and began a turn back to the airport under partial power. At a point about 90 degrees to the departure runway, he said the "engine power was lost completely," and the airplane "lost significant altitude." He elected to perform a water landing, rather than risk hitting homes on the shore. The airplane impacted in about 7 feet of water. An engine teardown revealed a hole in the upper case halves between cylinders 3 and 4. The number 3 and 4 connecting rods were broken from the wrist pin to the crankshaft. Laboratory examination revealed the fracture surface of the threaded half of one of the connecting rod bolts was very flat, contained arrest marks, and had a thumbnail pattern indicative of fatigue propagation. The fatigue region consumed approximately 80 percent of the cross-sectional area.

On June 6, 2002, about 0715 eastern daylight time, a Cessna P206, N2594X, registered to an individual, impacted with water during a forced landing shortly after takeoff from the Dare County Regional Airport, Manteo, North Carolina. Visual meteorological conditions prevailed at the time. A visual flight rules (VFR) flight plan was filed. The personal flight was conducted under 14 CFR Part 91. The airplane was substantially damaged. The private-rated pilot and one passenger reported serious injuries. The flight was originating at the time, and was en route to Treasure Cay, Abaco, Bahamas (MYAT), with a stop over for fuel at Brunswick, Georgia.

According to the pilot, just after takeoff, and just as he passed the departure end of runway 23, the engine "began stumbling, and seemed to lose partial power." He estimated the altitude to be "maybe 200 feet." He checked the throttle for full power, and began a turn back to the airport under partial power. He turned to the left, in an attempt to make runway 35 or any cleared portion of the field, under power. At a point about 90 degrees to the departure runway, he said the "engine power was lost completely," and the airplane "lost significant altitude." He attempted to establish a glide, but then elected to turn the airplane back toward the water (Croatan Sound), rather than risk hitting homes on the shore. He brought the airplane down in a "wings level water landing." The airplane impacted in about 7 feet of water.

The airport manager reported that the pilot refueled the airplane before he departed, and the result of the fuel tests on the in-ground tank were negative for contamination. However, the airport manager did see the wreckage, and he reported that he had observed a hole in the engine case.

The engine from N2594X was torn down and examined under the supervision of NTSB investigator Eric Alleyne of the Southern Region Office, Atlanta, Georgia, at the facilities of Atlanta Air Salvage, Griffin, Georgia, on June 13, 2002.

The power plant examination revealed a hole was observed in the upper case halves between cylinders 3 and 4. The number 3 and 4 connecting rods were broken from the wrist pin to the crankshaft. The counter weight was also broken off by the number 4 connecting rod. One of the lifter housings broke from the case with the lifter still in it. Both rod end caps, rod bearings were damaged; pieces of rod bolts, piston pieces, ring pieces, casing pieces and miscellaneous metal fragments were located in the oil pan. One of the connecting rods had exited the case through the top of the engine. The right forward engine mount was broke and the oil pan had a hole in it.

Examination of the fuel system revealed that the fuel tanks were not compromised. The left fuel cap was vented and the gasket was pliable. The right fuel cap was not observed. The fuel selector handle was in the left tank position and the linkage to the valve was intact. The valve position was not verified. No fuel was observed.

The following parts were sent to the NTSB Materials Laboratory, Washington, D.C. to be examined further; Two connecting rod bolt sections, two counterweight fragments, an oil filter, two lifters, the number two connecting rod assembly (rod, cap, bearing, and bolts), a fractured connecting rod, the number three and four rod caps, counterweight snap rings (two items), rod bearing sections (four pieces), and the number 3 main bearing.

According to the NTSB Materials Laboratory Factual Report, the examination revealed that both halves of a fractured connecting rod bolt were found with minimal deformation or post-fracture damage. Visual examination of the fractured bolt revealed that it was fractured approximately 0.85 inch from the head in the raised shoulder in the middle of the bolt shank, corresponding to the connecting rod split line location. The fracture plane was located within a non-uniform 360-degree band of fretting/wear approximately 0.05 inch wide. The nut was still on the bolt with a cotter pin in place. Approximately two full threads were visible extending beyond one side of the nut at the end of the bolt, and three threads were visible on the opposite side of the nut. Examination of the fracture surface of the threaded half of the bolt revealed that it was very flat, contained arrest marks, and had a thumbnail pattern indicative of fatigue propagation. The fatigue region consumed approximately 80 percent of the cross-sectional area. Higher magnification examination confirmed that the fracture initiated in a band of fretting/wear at the surface with multiple origins. Secondary cracking was observed adjacent to the fracture surface. The fracture surface revealed the presence of iron, chromium, manganese, nickel, molybdenum, and silicon consistent with the requirements of 4337/4340/8740 steel (AMS 6412, 6415, or 6322, respectively) in addition to small amounts of aluminum. The macro-hardness taken on the head of the bolt was HRC 36 (Hardness on the Rockwell C Scale) average, slightly below the requirement of HRC 38-42.

The debris captured in the oil filter was consistent with material from the connecting rod bearings. No evidence of foreign material was observed.

Examination of the connecting rod bearings revealed that all of the anti-rotation tabs were flattened. The exterior surfaces of the bearings were not worn, indicating that the bearings did not rotate against the rod end and rod cap. Also, there was no evidence to suggest that there was a lack of lubrication or overheating in the system. No determination could be made as to the identity of the damaged connecting rod. It could not be determined if the fatigued bolt was from the number 3 or 4 rod. Examination of all the split-line faces did not reveal the presence of significant wear or fretting.