NTSB Factual Report

HISTORY OF FLIGHT

On April 18, 2003, at 0930 Pacific daylight time, a Eurocopter/France (Aerospatiale) AS355N, N355J, experienced a loss of control during landing at the Redding Municipal Airport (RDD), Redding, California. Sierra Pacific Industries operated the helicopter under the provisions of 14 CFR Part 91. The helicopter sustained substantial damage after the tail rotor gearbox struck a brush pile. The two commercial rotorcraft rated pilots were not injured. Visual meteorological conditions prevailed for the local area familiarization flight, and no flight plan had been filed. The flight was departing at the time of the accident.

In an interview with the National Transporation Safety Board investigator-in-charge (IIC), the chief pilot, who was the flying pilot (FP), noted a loss of engine power and tail rotor thrust while in a hover. He made a precautionary landing, and landed harder than normal. He lowered the collective, and the helicopter spun around hitting the brush pile. The tail rotor and tail rotor gearbox separated from the tail boom, and the right landing skid collapsed.

In the FP's written statement he reported that after liftoff he brought the helicopter to a hover with the intent of repositioning it to an adjacent clear and level grassy area prior to calling for a takeoff clearance. While hover taxiing in a northerly direction over a graded gravel access road, he noted a discrepancy in the power torque indicator needles. The number 1 needle fluctuated between 0 percent and 10 percent lower than the number 2 needle (which had a normal indication), with an average position of 25 percent lower than the number 2 needle indication. All of the other engine, systems, and rotor indications were normal.

The FP elected to land the helicopter for further investigation, even though he felt it was a torque indication problem and not a power problem. He attempted to land on the gravel road; however, as the helicopter came in to land the rotor wash kicked up dust and debris. He decided to relocate the helicopter another 50 yards north of his current location to a less dusty, level, and unobstructed area. He stopped the taxi, hovered between 5 and 10 feet, and began a slow counterclockwise (left) turn. His intention was to keep a brush pile, located about 36 yards to the west of his current location, in his sight while turning to face the direction of "anticipated vehicular or pedestrian traffic."

The FP reported that the initial 45 degrees of turn were normal, with anticipated levels of control movement required. The helicopter then began an unanticipated change in attitude and moved in "contradiction to control inputs." The FP noted that the noise levels in the cabin changed noticeably; "abnormally quiet with a subtle increase in turbine or main rotor tone." The FP stated that the helicopter began an abrupt nose-low attitude change while rolling slightly left with the yaw movement increasing in rate in the counterclockwise direction. The helicopter started to descend while moving left (sideways) and forward towards the brush pile. He immediately applied "pedal, cyclic and collective forces," which eventually became full aft cyclic and full right pedal input in an attempt to arrest the helicopter's movement.

The FP stated that the helicopter's response to the flight control inputs were not what he expected. The helicopter continued to pitch down, the yaw rate continued to increase to the left, and the helicopter continued to move in the direction of the brush pile, all while descending. At that point he reduced cyclic and collective forces and applied full right pedal, and then moved the cyclic slightly to the left to redirect the helicopter to a point adjacent to the brush pile. The yaw rate increased "dramatically" and he knew that a loss of helicopter control was imminent. He lowered the collective, neutralized the cyclic, while maintaining full right pedal application and landed within a few feet of the brush pile.

The FP indicated that full down collective had been applied upon touchdown, and he then heard a series of loud crashing and breaking noises as the main and tail rotors contacted the brush pile. The helicopter immediately rotated to the left, "and in opposition to the direction of main rotor movement." The tail boom was severed and the right main skid was displaced. The FP applied the main rotor brake to stop the main rotor blade rotation.

In an addendum to the FP's original written submission, he reported that the event "produced little apparent vibration;" however, both pilots noted a sudden change in noise level in the helicopter. He stated that the sound change "could be associated with a main rotor blade falling out of track with the remaining blades."

In the non-flying pilot's (NFP) written statement, he indicated that the engines were started normally and the FP completed the checklists after startup; all checks were normal. They lifted off and came to a low hover (about 6 feet), and the then FP proceeded to reposition the helicopter to a vacant field near the hangar. During the hover taxi the FP noted a split in the torque needles, and decided to land in the field. The NFP cleared the area to the right of the helicopter, as the FP had started a left pedal turn. While the NFP's head was turned to the right he noted a sudden change in noise levels. He looked forward and the helicopter pitched down and yawed to the left. The FP then directed the helicopter to a landing area adjacent to a "brush/burn" pile. After touchdown the helicopter continued in a counterclockwise rotation striking the brush pile and damaging the tail rotor, tail boom, main rotor blades and the right skid. The helicopter was shut down and secured, and the main rotor brake was pulled, "stopping the main rotor blades."

Damage to the helicopter included the separation of the vertical tail, tail rotor, and tail rotor drive gearbox assemblies. The red arm of the STARFLEX rotor head separated at a 45-degree angle from the leading edge outboard to the trailing edge inboard. The right landing skid had buckled.

PERSONNEL INFORMATION

According to the flying pilot's written report he held a commercial pilot certificate with a rotorcraft-helicopter rating issued on December 8, 1999. The pilot reported that he had accrued a total rotorcraft flight time of 200 hours, with 150 hours in the accident make and model.

AIRCRAFT INFORMATION

According to a sales order packing list, on February 20, 2002, the core STARFLEX (part number (350A31-1907-3; serial number M4741) became the property of American Eurocopter (AEC) as a special warranty exchange between AEC and Sierra Pacific. A maintenance logbook entry recorded on March 4, 2002, indicated that the main rotor head assembly had been disassembled and returned AEC to for evaluation. A new STARFLEX (part number 350A31-1916-00; serial number M3008) was installed.

Per part number and serial number, the new STARFLEX was subject to Airworthiness Directive (AD) 2002-03-52 issued on March 30, 2002, which required a visual inspection of the star arm end for a gap in the adhesive bead between the bushing and each arm end. The purpose of the AD was to address reports of bonding failures between the metal bushing and the STARFLEX arm end that led to severe vibrations that had resulted in several emergency landings. If the bonding discrepancy was not detected, the FAA ascertained that severe lateral vibrations could result, with a subsequent loss of helicopter control. To remain in compliance with the AD, the time inspection interval was required before each start and thereafter at intervals not to exceed 4 hours time-in-service.

A review of the aircraft's daily flight logs for the past year showed that the AD had been complied with, with no defects found. On the day of the accident the AD inspection had been complied with and signed off by the company's airframe and power plant mechanic, with no defects noted.

The STARFLEX assembly, integral to the three bladed rotor system, was configured with three equally spaced arms in a star like fashion, constructed of semi rigid composite material, and was integrated into the rotor head assembly. The purpose of the STARFLEX was to react to the torsional forces associated with pilot control inputs and blade pitch.

The pilot reported a torque split between the number 1 and number 2 engines. According to AEC's flight manual, when the needles split on the torque indicator, the pilot should utilize the equalizing trim to match the power loads between the two engines. In the event of a torquemeter failure, the pilot should not exceed 65 percent on the other torque meter.

TESTS AND RESEARCH

Eurocopter France submitted a report regarding testing of the accident STARFLEX. The full report is attached to the factual report. Eurocopter France stated in their report that they analyzed the damaged STARFLEX rotor head arm, and that the failure mode appeared to have initiated with an incipient delaminations of the arm occurring in plies showing an area of dark contamination. This condition may have stemmed from ply edge contamination and resin voids allowing progressive delaminations to occur. The actual failure of the arm is a static type failure in flapping/drag mode, predominately in the drag mode. Eurocopter France concluded that the fracture of the STARFLEX arm in flight did not seem plausible because this type of failure would occur at maximum sustained loads, which do not occur in hover flight. The delaminations would result in a 10 percent drop in rigidity in the flapping axis, and the STARFLEX arm would easily absorb the drag loads. Additionally, delaminations of this type would have been visible and detectable during daily inspections.

According to the AS355N FAA approved flight manual, in Chapter 3 - Emergency Procedures, section 9 titled Tail Rotor Malfunction, a loss of tail in power-on flight will result in the helicopter yawing left.

Both engines were test run at Turbomeca, Grand Prairie, Texas, under the auspices of a Safety Board Investigator on September 3, 2003. No discrepancies were noted with the test run.

The Safety Board investigator-in-charge conducted a search of the FAA Service Difficulty Report (SDR) database for Aerospatiale AS355N helicopter STARFLEX issues. The SDR search revealed no AS355N STARFLEX related issues.

NTSB Probable Cause

Failure of the pilot to maintain adequate obstruction clearance during landing, which resulted in the tail rotor contacting a brush pile. The accident sequence was precipitated by the initiation of a precautionary landing by the pilot due to a discrepancy with the engines' torque meter gauge.