On July 29, 1999, about 0915 Atlantic standard time, a Hughes OH-6A, N66377, registered to the U.S. Border Patrol, landed hard while making a forced landing, near Aquadilla, Puerto Rico, while on a Title 14 CFR Part 91 public-use flight. Visual meteorological conditions prevailed at the time and no flight plan was filed. The helicopter received substantial damage and the commercial-rated pilot was not injured. The flight originated from Aquadilla, Puerto Rico, the same day, about 0900. The pilot stated he was in cruise flight at 750 to 800 feet, 8 miles east of Aquadilla, when he heard a metal grinding sound in the back of the helicopter, followed by a short noise of air escaping under pressure. Immediately, he heard the engine-out audio warning in his helmet and saw the engine warning light on the instrument panel flash. He had a sudden loss of performance and altitude and applied collective. The helicopter slowed and continued to lose altitude. He entered autorotation and aimed for a road. As he approached the landing zone, he flared and raised the collective control. The helicopter touched down hard, to the north of the road, in a field. The main rotor contacted brush to the left and right of the helicopter. The helicopter swerved to the right, and he lowered the collective and twisted the throttle to the cut off position. The main rotor blades continued to rotate without resistance. He did not remember hearing the engine running before shutting it off. He shut down the electrical system, and after the main rotor blades stopped turning, he exited the helicopter. Postcrash examination of the helicopter at the crash site was conducted by representatives of the U.S. Border Patrol and Rolls Royce Allison Engines. The helicopter came to rest upright, on a 122-degree heading, with a slight nose down and right roll attitude. The main rotor had severed the tailboom just forward of the aft frame fitting. The landing gear skids were spread. The outboard section of one main rotor blade had separated and was located adjacent to the main wreckage. Examination of the engine showed no external damage. Hand rotation of the engine power turbine established engine to main rotor head continuity. Uncontaminated fuel was found in the engine fuel lines, fuel pump, and fuel control. The oil cooler blower assembly was damaged. Examination of the engine assembly and accessories was performed in Puerto Rico after removal of the engine from the helicopter and in Dallas, Texas, after shipment of the engine to an engine overhaul facility. An engine analytical teardown showed no anomalies with the engine assembly. Testing of the engine fuel system components showed they operated within manufacturer's service limits. (See Rolls Royce Allison Report an attachment to this report) Examination of the oil cooler blower assembly was conducted by Boeing Helicopters, Engineering Laboratory, Mesa, Arizona. One impeller blade was fractured due to overload, which opened a 2-inch gap in the impeller. No evidence of fatigue was found in the components. A black smudge on the upper panel consisted of transfer of magnesium material, indicating contact with a magnesium object. Examination of the oil cooler shroud and adapter showed the shroud was essentially undamaged. Nine of the twelve attachment rivets were sheared in a same direction. The absence of fretting damage at the rivet holes supports the conclusion that the impeller fractured by sudden overload. The shroud was essentially undamaged with some minor abrasion damage and some minor areas of debris build up. Based on damage to the adapter, it appears that the three rivet section of the blower assembly was fractured away from the blower, which subsequently caused the blower to stop spinning and shear from the adapter. (See Boeing Helicopters Report an attachment to this report) Examination of the transmission was conducted at NAC, Oakland, California. Disassembly of the transmission showed metal contamination in the low speed gear set. Conclusions of the disassembly was bearing failure, low speed gear set scoring due to incorrect pattern setup or lack of oil pressure due to a clogged nozzle. (See NAC Report an attachment to this report) Additional examination of the transmission was conducted at Aero Systems, Inc, Erie, Colorado, with a representative of Boeing Helicopter present. This examination concluded that the condition of the transmission with the exception of the Triplex bearing possibly making some metal, was normal for the recorded time in service since overhaul. There was nothing found that would cause seizing of the transmission. (See Aero Systems, Inc. Report an attachment to this report) Aircraft logbook records show the main transmission was installed on the helicopter after overhaul, in January 1999, at aircraft total time 2,181.2, 385.7 flight hours before the accident. (See logbook pages.)