WPR23LA278

The flight was planned to transport a vegetation removal crew to power distribution lines in mountainous, high-elevation terrain. As the helicopter approached the drop-off point, the pilot observed through the engine instruments that the No. 2 engine was operating at reduced power and that the main rotor speed was decaying. He maneuvered away from power lines and made adjustments to the engine trim settings; with power seemingly back to normal, he began the approach again. However, as the helicopter approached the landing zone the main rotor speed again began to decay such that the pilot knew it would be insufficient for landing. He turned the helicopter away from terrain, and the helicopter began to descend into the surrounding trees, where it landed on its belly and rolled over. The engines were still operating after the accident, and the pilot shut them down and secured the electrical system. The compressor section of the No. 2 engine had been replaced almost two weeks before the accident. Both the pilot and maintenance personnel stated that following the replacement, they encountered intermittent problems synchronizing the engines. While troubleshooting they performed a series of adjustments to the engine trim system followed by test flights, after which they were eventually able to match the engines. Following the compressor change, the engine was making a loud “howling” sound that neither the pilot nor maintenance crew had heard on this helicopter before. Despite this, they continued to fly the helicopter on other missions leading up to the accident. Postaccident examination did not reveal any anomalies with the engine trim system or No. 1 engine. Examination of the No. 2 engine revealed that the compressor bleed valve had a hole in the rolling section of its diaphragm, which was causing the valve to intermittently fail. Such a hole would have resulted in the intermittent engine performance observed leading up to the accident flight, and the reduction in power observed before impact. Examination also revealed that the compressor section incurred a balance and clearance anomaly during its overhaul. These discrepancies would have resulted in a minimal reduction in engine efficiency and would likely have explained the howling sound, but not the abrupt loss of power. The bleed valve was manufactured under an FAA Parts Manufacture Approval (PMA) process. Although the diaphragm used the same materials specified by the manufacturer, it failed about 77 hours short of its service life. The pilot adjusted the engine trim setting during the emergency in an attempt to increase power to the failing engine. However, the trim system was differential in design, such that using it in this manner would have caused a reduction in power to the running engine. Regardless, the loss of power occurred at an altitude and flight phase that would have made a safe recovery challenging considering the close proximity to the terrain.

The helicopter was manufactured in 1981. It was powered by two Rolls-Royce/Allison 250 C20F turboshaft engines. Both were mounted at the top of the fuselage to the rear of the main transmission gearbox. The engines were controlled mechanically through a cable-driven collective anticipator system. The system allowed for engine power balance adjustment using an electro-mechanical “engine trim” assembly, controlled by the pilot with a rocker switch on the collective lever grip. The system was comprised of a series of mixing bellcranks mounted to the main transmission deck and it allowed the pilot to adjust engine governor balance through an electric motor controlled by the rocker switch. The system was differential in design, such that if the pilot commanded an increase in power on one engine, it would reduce engine power on the other. The helicopter’s flight manual stated that under normal flight conditions, the trim system is used to synchronize the engines during takeoff, climb, and hover, and that it is also used to increase power to each engine during power check operations. There are no specific instructions to use the trim system during a loss of engine power event; however, the emergency procedures section stated: “The procedures outlined in this section deal with the common types of emergencies; however, the actions taken in each actual emergency must relate to the complete situation.” The emergency procedures section, along with the height-velocity diagram, indicated that the helicopter can be flown safely after a single-engine failure during takeoff at altitudes of up to 7,000 ft and a weight at or below 4,740 lbs. The helicopter’s maximum permissible weight (ambient conditions dependent) was 5,291 lbs, and according to the operator its weight at the time of the accident was 4,963 lbs. Maintenance The compressor section of engine No. 2 had been replaced 12.1 flight hours before the accident,. Both the pilot and maintenance personnel stated that in the days following the replacement they encountered difficulty synchronizing the engines. The pilot described the problems as intermittent, and while troubleshooting they performed a series of trim adjustments and test flights and were eventually able to match the engines to within the manufacturer’s power specifications. During troubleshooting, the pilot noted that the bleed valve was approaching its overhaul time, but because the other engine was operating correctly, they decided to leave it in place. Both the pilot and maintenance personnel reported that following the compressor change and up to the accident flight, the engine was making a loud “howling” sound. The sound could not be heard within the cabin due to ambient noise but was very pronounced from outside during both ground and flight operations. They reported that it did not appear to be affecting engine performance. Although they had not heard the engine make this sound before, the mechanic put it down to a function of the neweroverhauled compressor blades. They continued to fly multiple missions with the engine making the sound. On July 19, 2023, about 0830 mountain daylight time, an Aerospatiale, AS-355E, N102UM, was substantially damaged when it was involved in an accident near Farmington, Utah. The pilot and four passengers were not injured. The helicopter was operated as a Title 14 Code of Federal Regulations Part 135 on-demand air taxi flight. The flight was planned to transport a vegetation removal crew to power distribution lines in the Francis Peak area, within the Wasatch Mountain Range. The pilot initially departed from the helicopters base in Ogden-Hinckley Airport (OGD) Ogden, Utah, about 0800 and flew directly to the staging area in the foothills of the peak. Upon landing, he briefed the crew, and the flight departed. During the climb-out to the work area, the pilot observed that the main rotor speed was beginning to decay. As he maneuvered for landing, he noticed that the engine speed and temperature readings were not appropriately matched between the engines, with the No. 2 engine showing an 8-10% reduction in torque. He performed a series of engine adjustments using the trim system, maneuvered away from the power lines near the landing area, and increased the helicopter speed; the main rotor speed returned to normal. He began the landing approach again; however, as the helicopter came to within about 20 to 30 ft of the landing zone (elevation 8,600 ft), the main rotor speed again began to decay such that the pilot knew it would be insufficient for landing. He turned the helicopter away from the landing zone to maintain clearance from the mountain. The helicopter then began to descend, and the main rotor blades started to cut through the surrounding trees. The helicopter landed on its belly and rolled over. The pilot and crew egressed through the left door. The engines were still operating after the accident, and the pilot shut them down and secured the electrical system. Postaccident examination of the engine trim system indicated that it was functioning appropriately. Both engines were removed and examined at a Rolls-Royce-approved overhaul facility under the oversight of the NTSB. Engine No.1 was intact and essentially undamaged. The first stage compressor blades exhibited leading edge nicks consistent with foreign object ingestion, but the remaining stages throughout the engine were undamaged and could be rotated freely by hand. There was no evidence of catastrophic or uncontained failure. There was no pneumatic or fluid leak, and the fuel control unit, fuel pump, fuel nozzle, power turbine governor and compressor bleed valve were all removed and tested and were within serviceable limits. Engine No. 2 was similarly intact and displayed similar foreign object damage to the first stage compressor blades, with no evidence of catastrophic failure. The core engine sections were disassembled, and the fuel control unit, fuel pump, fuel nozzle and powerturbine governor were all tested, and were within serviceable limits. Examination of the compressor section revealed that, although there was no evidence of catastrophic failure to any of the blades or stators, a prominent groove of the inner liner was present in the area of the Nos. 5 and 6 blade tips in one of the case halves. The other case half had similar rub signatures but to about ½ the depth. The neighboring stator vanes on one case half exhibited blade tip smearing opposite the direction of rotation, with corresponding rub marks on the adjacent area of the blade hub. During testing of the compressor bleed valve, the unit appeared to intermittently hang up when transitioning from the open to closed position. It did not meet the specifications for the interstage pressure reduction required to close the valve; however the diaphragm leakage test yielded normal results. The unit was dissembled, and an approximately 1/16th inch hole was observed in the rolling section of the diaphragm (see figure 1). Figure 1- Diaphragm The diaphragm was an FAA PMA component (6874725HT), manufactured by HYE-Tech LLC. It had a service life of 1,500 hours and, according to maintenance records, it had 76.6 hours remaining at the time of the accident. The bleed valve assembly was sent to the NTSB Materials Laboratory for examination. Fourier-transform infrared (FTIR) spectroscopy analysis indicated that the diaphragm material composition was consistent with those specified by Rolls-Royce for their similar OEM component. Scanning electron microscope examination of the diaphragm revealed the ends of many fibers were elongated. Several fibers in the area of separation (hole) were reduced to nearly a knife edge, and many of the fibers exhibited evidence of severe mechanical damage consistent with the fracture faces contacting each other. The non-damaged fracture faces of the fibers showed evidence of either a shear fracture, or cup and cone features typical of overstress separation. The elastomer portion showed no evidence of fatigue cracking. The compressor assembly was examined at the facilities of Rolls-Royce, with the results reviewed by an NTSB powerplant specialist. Examination confirmed that there was rub on the stage 4 through 6 axial compressor blade tips and their associated case tracks, consistent with rotor imbalance, and that when measured, the imbalance was 5.1 times the allowable limit on the forward end, and 8.4 times the limit at the aft end. The observed metal to metal contact is consistent with the imbalance causing the rotor to orbit on its centerline. Further examination of the compressor section assembly revealed circumferential rub on the compressor impeller, with uneven wear to the abradable areas of the compressor shroud. The build clearance between the impeller and shroud did not fall within tolerance, resulting in a clearance 55% larger than maximum specification. Such an increase in clearance would have resulted in reduced compressorefficiency.