Big Lake, MN, USA
N153B
BELL 47D1
The airline transport pilot was preparing to depart in the helicopter on his fifth aerial application flight of the day. While on the ground, he increased engine rpm from idle to 3,100 rpm. Before making a collective input, the pilot felt a "sudden pull" and heard a "very loud snap." The pilot then saw that a company truck, which was parked nearby, had just sustained damaged and that the helicopter's two main rotor blades were missing. He shut down the engine; exited the helicopter; and found the main rotor blades, which were constructed of wood, splintered and scattered on the ground from about the helicopter's 2 o'clock position counterclockwise to its 8 o'clock position. The helicopter's transmission case was fractured, and the rotor mast had separated from the transmission and came to rest about 120 ft behind the helicopter's 7 to 9 o'clock positions. The damage to the company truck was consistent with impact from one or both main rotor blades. The main rotor blades were installed on the helicopter about 6 years before the accident. Examination of the recovered sections of the wood from the main rotor blades found that the blades were constructed according to blade design specifications and that there was no evidence of fungal decay in the wood. One main rotor blade had a metal core assembly separation near its root. Examination of two adjoining sections of the assembly revealed a fracture through the bar at the outboard edge of a lug. Visual examinations of the fracture faces revealed a corroded region at the trailing edge of the bar with fracture marks and arrest lines that were consistent with fatigue progression from the trailing edge. Ratchet marks indicated multiple fatigue origins in the area. The fatigue origin area had propagated about 0.25 inch toward the leading edge of the blade. Given this information, the main rotor blades most likely separated from the helicopter due to the failure of the blade's metal core assembly as a result of undetected corrosion and fatigue. The metal core assembly's fatigue origin area within the wooden blade would not have been visible to a pilot or mechanic, but x-ray nondestructive techniques could likely have detected the fatigue crack within the wooden blade. However, the Bell 47D1 Federal Aviation Administration (FAA)-approved maintenance manual and related FAA service bulletins did not include procedures to inspect the condition of the main rotor blade metal core assembly. The wooden main rotor blades are no longer manufactured, and an estimated 200 sets of these blades were still in operation at the time of this investigation.
HISTORY OF FLIGHTOn July 6, 2017, about 1130 central daylight time, a Bell 47D1 helicopter, N153B, had main rotor blade separation and contact with a ground vehicle during the initiation of an aerial application flight near Big Lake, Minnesota. The airline transport pilot was uninjured. The helicopter sustained substantial main rotor blade damage. The helicopter was registered to the pilot and operated by Oleen Ag Air as a Title 14 Code of Federal Regulations Part 137 aerial application flight. Day visual meteorological conditions prevailed in the area about the time of the accident, and the flight was not operated on a flight plan. The local flight was originating at the time of the accident. According to a statement from the pilot, he had completed four aerial application flights and landed the helicopter on the ground to be refilled for his fifth flight. He, in part, stated, "At approximately 11:30 am, I was parked on the ground. After being refilled with pesticide, I increased RPM from idle to 3,100 RPM. I noted a smooth engine and synchronized needles with the main rotor and the tail rotor. Before pulling any collective, I felt a sudden pull and heard what sounded like a very loud snap. Then, I looked to my right and saw that the water tank on my load truck had been punctured, and that the driver's side cab had also been dented." The pilot observed that the main rotor blades were missing and the engine was still running. He shut the engine off and exited the helicopter. The rotor blades splintered and the splintered debris was collected and relocated before Federal Aviation Administration (FAA) inspectors could examine the accident site. PERSONNEL INFORMATIONThe 49-year-old pilot held an FAA airline transport pilot certificate with a multi-engine airplane land rating. He also held instrument commercial pilot privileges in single engine land airplanes and in rotorcraft helicopters. He held an FAA second-class medical certificate issued on July 1, 2016, with no limitations. He reported accumulating 16,300 hours of total flight time and 20 hours of flight time in the same make and model as the accident airplane. AIRCRAFT INFORMATIONN153B was a Bell 47D1 helicopter, with serial number 7. The helicopter was manufactured in 1947 and was issued a normal category airworthiness certificate. Its helicopter's type certificate data sheet indicated it seated three and had a maximum gross weight of 2,200 pounds. The helicopter was configured for agriculture and pest control and was issued a restricted category airworthiness certificate. The last helicopter annual inspection was conducted on May 15, 2017. At the time of that inspection, the helicopter had a total time of 11,902.5 hours. The helicopter was equipped with a 200-horsepower Lycoming VO-435-A1F engine with serial No. L-1047-31C. The engine drove two wooden main rotor blades with blade part No. 47-110-120-30. The main rotor blades are constructed of laminated wood and covered with lacquered fiberglass cloth. For protection, a stainless-steel cap is installed on the leading edge. Fixed trim tabs located near the outboard end of the trailing edge of each blade may be adjusted (by bending) to track blades. Each blade is secured to its grips by a special bolt passing through the root of the blade and metal plates which are attached to the top and bottom of each blade root. These metal plates extend over the trailing edge of the blade to provide attachment lugs for the blade drag brace. The main rotor blade had a finished length of 200 +/- 0.125 inches and a maximum chord plane length of 13.75 inches. The blade design called out the use of pine, birch, spruce, and balsa wood species for its construction. The blade design incorporated an internal steel core assembly that had a specified bar length of 199.62 inches and width of 1.62 inches. A lug was attached to the inboard portion of the core assembly, which had a maximum length of 9.62 inches and a combine bar and lug width of 4.75 inches. According to the operator, the main rotor blade serial Nos. were 9545 and 9546. The logbook endorsement indicated that the blades were installed on the helicopter on June 14, 2011 and at that time, they had accumulated 108.5 hours since overhaul. METEOROLOGICAL INFORMATIONAt 1155, the weather at the Buffalo Municipal Airport, near Buffalo, Minnesota, was Wind 170° at 6 kts; visibility 10 statute miles; sky condition clear; temperature 28° C; dew point 18° C; altimeter 29.94 inches of mercury. AIRPORT INFORMATIONN153B was a Bell 47D1 helicopter, with serial number 7. The helicopter was manufactured in 1947 and was issued a normal category airworthiness certificate. Its helicopter's type certificate data sheet indicated it seated three and had a maximum gross weight of 2,200 pounds. The helicopter was configured for agriculture and pest control and was issued a restricted category airworthiness certificate. The last helicopter annual inspection was conducted on May 15, 2017. At the time of that inspection, the helicopter had a total time of 11,902.5 hours. The helicopter was equipped with a 200-horsepower Lycoming VO-435-A1F engine with serial No. L-1047-31C. The engine drove two wooden main rotor blades with blade part No. 47-110-120-30. The main rotor blades are constructed of laminated wood and covered with lacquered fiberglass cloth. For protection, a stainless-steel cap is installed on the leading edge. Fixed trim tabs located near the outboard end of the trailing edge of each blade may be adjusted (by bending) to track blades. Each blade is secured to its grips by a special bolt passing through the root of the blade and metal plates which are attached to the top and bottom of each blade root. These metal plates extend over the trailing edge of the blade to provide attachment lugs for the blade drag brace. The main rotor blade had a finished length of 200 +/- 0.125 inches and a maximum chord plane length of 13.75 inches. The blade design called out the use of pine, birch, spruce, and balsa wood species for its construction. The blade design incorporated an internal steel core assembly that had a specified bar length of 199.62 inches and width of 1.62 inches. A lug was attached to the inboard portion of the core assembly, which had a maximum length of 9.62 inches and a combine bar and lug width of 4.75 inches. According to the operator, the main rotor blade serial Nos. were 9545 and 9546. The logbook endorsement indicated that the blades were installed on the helicopter on June 14, 2011 and at that time, they had accumulated 108.5 hours since overhaul. WRECKAGE AND IMPACT INFORMATIONAn FAA inspector examined the wreckage, and forwarded images of the wreckage to the National Transportation Safety Board (NTSB) investigator in charge. Additionally, the pilot supplied a sketch of the accident site. The examination, images, and sketch revealed that wood split from the main rotor blades was scattered randomly on the property. The separated wood splinters were scattered roughly from the helicopter's 2 o'clock position counter-clockwise to its 8 o'clock position. The transmission case was fractured, and its rotor mast was not present. The mast came to rest about 120 ft behind the helicopter's 7 to 9 o'clock position. The ground near the helicopter did not exhibit any witness marks consistent with the helicopter impacting terrain or being moved. ADDITIONAL INFORMATIONThe wooden main rotor blades are no longer manufactured. A wooden main rotor blade repair station representative was asked how many main rotor blades were in operation. He estimated that there are about 200 sets of Bell 47 main rotor blades remaining in current operation. The fatigue origin area was within the wooden blade and was not visible. The helicopter type certificate holder forwarded excerpts from the FAA approved helicopter maintenance manual. The excerpts were reviewed, and they included direction to inspect wooden blades for condition of finish, loose areas or blisters, damage that punctures or tears fabric, trim tabs and leading-edge cracks, dents, distortion, and punctures, and wood checks in the butt area. Additionally, the index of service bulletins were reviewed. Neither the maintenance manual nor the service bulletins contained direction to examine the main rotor blade core assembly for its condition. The helicopter type certificate holder, in a letter to the NTSB investigator in charge, recommended "looking closer at the transmission for signs of overload common with a rotor strike" as opposed to a liberated rotor blade. The type certificate holder reported that "a good witness feature for this type of case could be the metal ballast box located just inboard of the blade tip." The type certificate holder stated that an assessment of where the main rotor blade debris actually came to rest would be a great indicator of the blade trajectory. However, the debris had been moved before any investigators had arrived. The type certificate holder also reported that the red blade's ballast box was fully intact, showing no signs of deformation. However, the yellow blade's ballast box was not found in the recovered helicopter wreckage, inspected in the state in which it has been purchased and delivered from the insurance company. The NTSB aviation accident database and associated dockets were reviewed and there were no transmission or ballast box images available in the investigation dockets where a Bell 47 had a main rotor blade strike, which precludes the type certificate holder's recommended comparison. OTHER DAMAGEA servicing truck was parked nearby. The truck, to include the driver side of the cab and water tank, was damaged in one impact plane consistent with the size and shape of a main rotor blade. The height of the impact plane was consistent with the height of the rotor blade when the helicopter at or near ground level. TESTS AND RESEARCHAn engineer with the FAA Airplane Certification Office examined the recovered wreckage. The separation surfaces that he examined showed evidence of failure in overload. The engineer did not see anything that pointed to fatigue on any of the parts. One of the main rotor blades shattered and he described it like a baseball bat when hit by a ball in the wrong spot. The other main rotor blade showed similar but much less damage. The mast appeared to have ripped out of the transmission towards the left (aft looking forward). The engineer reviewed accident pictures and noted that the fuel truck was parked on the right side of the helicopter, and the pilot does not have much for visibility over his right shoulder as the pilot's view blocked by a holding tank. Another picture showed that the fueling hose was laid out on the ground. The recovered sections of the wood from the main rotor blades were shipped to United States Department of Agriculture, Forest Service, Forest Products Laboratory, near Madison, Wisconsin, for detailed examination. The examination revealed that the blades were manufactured from plywood and defect-free lumber, in accordance with the rotor blade design. The plywood was made from birch veneers and laminated with an exterior grade adhesive. The wood used in the remainder of the laminations was determined to be Sitka spruce. No evidence of fungal decay activity was observed in the plywood or wood laminations. Visual and microscopic examination of sections from the blades confirmed that no decay activity was present. One main rotor blade had a metal core assembly separation near its root. Two adjoining sections of that separated main rotor blade core assembly were removed and sent to the NTSB Materials Laboratory where a Senior Metallurgist conducted a detailed examination and produced a Materials Laboratory Factual Report No. 17-092. The core assembly, part No. 47-110-122-1, was a welded steel structure consisting of a tapered lug welded to the trailing edge of a longer bar. Engineering drawing specified a grade of plain carbon steel with an ultimate tensile strength between 55,000 and 100,000 psi. The metal core assembly was fractured through the bar at the outboard edge of the lug. Visual examinations of the fracture faces revealed a corroded region at the trailing edge of the bar with fracture markings and arrest lines indicative of fatigue progression from the trailing edge. Ratchet marking also indicated multiple origins in the area. The fatigue had propagated toward the leading edge about 0.25 inch. The initial portion of the fatigue was heavily corroded and the fracture surface obscured while the later areas were not. The remaining fracture was consistent with brittle overstress propagation with small shear lips around the periphery of the bar. Rockwell hardness (HRB) measurements on a cut cross core assembly section averaged 93.4 HRB, which was consistent with a carbon steel having an ultimate tensile strength of approximately 98,000 psi. Cross sectional examination revealed a ferrite and pearlite microstructure consistent with the specified material. The upper surface of the metal core assembly displayed significant surface corrosion with almost all the yellow green primer missing or discolored. The lower surface had less attack but had corrosion removing and undercutting the primer in many places particularly in the fracture area. The fatigue origin area had some surface corrosion, but no large pits were noted and remnants of the primer remained. Large pits and material removal were seen in the adjacent areas of the upper surface of the bar. The laboratory report is appended to the docket associated with this investigation.
The failure of the metal core assembly within a wooden main rotor blade due to undetected corrosion and fatigue, which caused the wooden blade to separate from the rotor mast.
Source: NTSB Aviation Accident Database
Aviation Accidents App
In-Depth Access to Aviation Accident Reports
