VACAVILLE, CA, USA
N36VR
COOPER VANS RV-6
The pilot made a forced landing into rough terrain following a bang and sudden loss of engine power. The engine had accumulated about 70 hours since overhaul. The sump contained a connecting rod, a wrist pin, bearings, many pieces of metal debris, and several rod cap bolts and nuts. None of the rod cap bolts in the sump were intact. They were all bent, fractured, and separated approximately at their midpoint. The pieces were covered with oil. The connecting rods for cylinders number two and four were leaning on the camshaft. The camshaft was not discolored. The crankshaft did not show evidence of discoloration or mechanical damage. A report of metallurgical testing of various engine pieces and their fracture surfaces strongly suggested that the nuts were not tightened on the connecting rod bolts according to specifications. A bolt that corresponded to the head of the bolt from cylinder number two exhibited failure characteristics of low cycle-high stress reverse bending fatigue. Reverse bending fatigue is indicative of an under-torqued bolt. The inspection revealed additional evidence that some of the nuts were not tightened to specification. Some bolts had flattened and polished bolt threads, as well as wear marks on the bolt's outer diameters. Some nut threads had only one end flattened. No fractured bolt had a nut on it except the fatigued bolt.
On February 26, 1999, about 1330 hours Pacific standard time, an experimental Cooper Vans RV-6, N36VR, sustained substantial damage when it collided with terrain during a forced landing following a loss of engine power approximately 15 miles north-northwest of Vacaville, California. The airline transport pilot was not injured; his passenger sustained minor injuries. The pilot owned, operated, and constructed the airplane, and conducted the local area personal flight under the provisions of 14 CFR Part 91. The flight originated at Vacaville about 1310. Visual meteorological conditions prevailed and no flight plan had been filed. The pilot stated he was selling the airplane. He departed his home base of Hanford, California, about 1030 that morning and arrived in Vacaville approximately 1 hour 15 minutes later. He flew the 1/2-hour demonstration flight from Vacaville and encountered no problems. He wanted to take a friend, who had helped him build the airplane, up for a local flight. He departed Vacaville at 1310, and had been flying for 20 to 25 minutes when he detected a faint metallic clicking sound coming from the engine. The sound grew louder and he detected a noticeable loss of thrust. This was followed by loud metallic sounds, and white smoke was observed in the cockpit. The smoke smelled like oil and was cleared by opening the fresh air vents. The propeller kept spinning a short while; then the engine seized. He tried to turn it with the starter, but it would not turn. He was over rough terrain so he glided to a spot on the shoreline of Lake Berryessa. Impact with the uneven terrain collapsed the nose wheel. The front of the airplane sustained impact damage and the firewall was buckled. The left wing was wrinkled and the empennage buckled under the left horizontal stabilizer. Several engine pieces were found along the wreckage path. The engine was a Textron Lycoming O-360, serial number L-8649-36A. The engine was assembled by a specialty shop, which catered to the homebuilt airplane market. A review of logbook entries revealed that maintenance releases were obtained from Federal Aviation Administration (FAA) Certified Repair Stations for the engine case, crankshaft, and camshaft. The shop installed new pistons, rings, exhaust valves, rod bolts and nuts, oil pump gears, piston pins, bearings, and other miscellaneous keepers and bolts. The assembler signed the engine off as ready to run in. The owner installed the engine in the airplane and accumulated a total time of approximately 70 hours prior to the mishap. The Safety Board investigator inspected the engine on April 22, 1999, at a hangar in Hanford, California. The owner provided one cam follower and one valve lifter that he recovered in the debris field about 50 feet behind the airplane. The investigator removed the cowling. He observed the No. 2 rod cap, a rod bolt piece with a nut on it, and two pieces of bearing material lying on the bottom right side of the lower cowling. The nut only engaged two threads of the bolt and the investigator could easily turn the nut with his fingers. The top of the case had two large sections missing and the bottom of the case had a section missing at cylinder No. 3. The connecting rods for cylinders 2 and 4 were leaning on the camshaft. The connecting rods and their bearings were not discolored. The camshaft was not discolored. The crankshaft did not exhibit any discoloration or mechanical damage. The investigator removed cylinder Nos. 1, 2, and 4. Cylinder No. 1's valves were seated, the rings rotated freely, and the piston was not mechanically damaged or discolored. This was the only cylinder that had its connecting rod attached to the crankshaft. However, the nuts on the connecting rod were finger tight. The investigator was unable to remove cylinder number three. The piston remained in the top portion of the cylinder barrel. The piston pin boss fractured in an irregular pattern and the piston pin was missing. All of the spark plugs were oily. Their electrodes were circular, the gaps were similar, and none exhibited any mechanical damage. The oil sump had a hole in the front side. The investigator removed the sump and found that it contained a connecting rod, wrist pin, bearings, many pieces of metal debris, and several rod cap bolts and nuts. None of the rod cap bolts in the sump were intact. They all were bent, fractured, and separated approximately at their midpoint. The pieces were covered with oil. The number one intake tube inside the sump had a hole that was 4 inches long around half of its diameter. Cylinder Nos. 1, 2, and 4, were retained by the investigator. He also retained all connecting rods, rod caps, connecting rod nuts and bolts, and bearings that he found. He delivered these pieces to Seal Laboratories in El Segundo, California, for metallurgical examination and analysis. The following is a summary of their findings. The lab arbitrarily assigned numbers to the pieces. They assigned H-1 to stand for the head of bolt 1, B-1 to stand for the body of bolt 1, and N-1 to stand for nut 1. Among the various nuts retrieved, they could positively match only three bolt bodies to the corresponding head or nut. These included H-1, B-1; B-2, N-2; and H-3, B-3. They observed flattened and polished threads on B-1, B-2, B-3, B-4, and B-5. They stated that these markings are an indication that the bolt was loose within its fixture. The report stated the fracture surfaces of B-1 and B-3 indicated overload, and the fracture surfaces of B-4 and B-5 resembled overload. The fracture surface of B-2 showed fatigue followed by overload. They inspected the fracture surfaces for the bolts under a scanning electron microscope (SEM). Bolts B-1, B-4, and B-5 indicated the presence of ductile overload fracture. One section of B-4 also exhibited evidence of some shearing during fracture. Bolt B-3 had three areas that exhibited ductile overload; one area also exhibited some shearing, and the other some shearing or rubbing. They examined three areas on the fracture surface of bolt B-2. One area indicated ductile overload fracture. The second area exhibited some characteristics of fatigue including multiple crack origins, indicative of high stress low cycle fatigue and fatigue striations. The third area revealed evidence of fatigue striations. The fatigue area contained some secondary cracks, which indicated high stress. This fracture surface resembled low cycle-high stress reverse bending fatigue followed by ductile overload. They made no observations for the fracture surface on H-10 due to the severity of the damage. Head H-10 remained in the rod cap for cylinder number three. The fracture surface on H-12 indicated overload, while the fracture surface for H-11 indicated fatigue followed by overload. They inspected the fracture surfaces for the heads under the SEM. They examined four areas on H-1. Three areas indicated ductile overload fracture. The fourth area exhibited a ductile overload fracture containing some shearing and mechanical damage by rubbing. Head H-3 indicated ductile overload fractures. They examined three areas on H-11. Two areas contained features characteristic of fatigue including multiple crack origins. They could not distinguish the mode of fracture for the third surface. The remaining surface was contaminated; however, it revealed a ductile overload fracture. This fracture surface resembled reverse bending fatigue followed by ductile overload fracture. They thought this head was the mating fracture surface to B-2. Head H-12 displayed an area with features of a ductile overload fracture; severe rubbing prevented determination of the fracture mode on another area. Nut N-2 sustained slight damage; its threads appeared to be mostly undamaged. Nut N-6 sheared and sustained severe damage. The threads sustained damage, but appeared flattened. Severe damage prevented inspection of the fracture surface. Nut N-7 sustained damage and was out of round. It had flattened threads on one end and secondary damage throughout. The report noted that flattened threads on only one end of the nut indicated a loose fit on the bolt. Nut N-8 sustained slight outside surface damage, and the threads appeared to be mostly undamaged. The outside surface of N-9 contained slight secondary damage. They observed severe flattening on one end of the threads; the remaining thread exhibited some secondary damage. They observed nut N-6 under the SEM. They examined three areas. They determined all three areas fractured due to ductile overload, and that one area had some secondary damage. They performed a Rockwell "C" hardness test for all of the bolts. The values were consistent. The values ranged from 36 to 40. The lab summarized its findings and conclusions and opined that the evidence strongly suggested that the nuts were not tightened on the connecting rod bolts. A thread section from a bolt that corresponded to the head of the bolt from cylinder number two exhibited failure characteristics of low cycle-high stress reverse bending fatigue. Reverse bending fatigue is indicative of an under-torqued bolt. The inspection revealed additional evidence that some of the nuts were not tightened to specifications. Some bolts had flattened and polished bolt threads, as well as wear marks on the bolt's outer diameters. Some nut threads were flattened on only one end. No fractured bolt had a nut on it except the fatigued bolt.
The failure of the engine assembler to apply the proper torque on the connecting rods bolts, which led to fatigue cracking and ultimately fracture of the rod bolts.
Source: NTSB Aviation Accident Database
Aviation Accidents App
In-Depth Access to Aviation Accident Reports
