Hana, Maui, HI, USA
N87EW
EUROCOPTER AS350BA (FX2)
As part of the required normal 6-month competency check for the commercial pilot, a Federal Aviation Administration (FAA) inspector was performing an examination of the pilot's competency in responding to a total loss of engine power event while in cruise flight. To perform this examination, the FAA inspector announced the beginning of the simulated power loss procedure and moved the fuel flow control lever out of the flight detent (the full forward, full open position) and back just enough to keep the lever from springing back into the detent; this was done to ensure that the engine was not supplying power to the rotor system during the autorotation. However, when the fuel flow control lever was moved, the helicopter yawed right and the generator out warning light illuminated, indicating that the engine had flamed out. The pilot briefly attempted a restart, but the engine exceeded the temperature limit and he discontinued the start attempt. Because of their close proximity to the ground, there was inadequate time to attempt another restart of the engine. The pilot-in-command (PIC) identified and proceeded toward a forced landing site. Due to obstructing trees in the touchdown zone below the area where the simulated engine out was conducted, the PIC overflew the trees with up collective input, leading to a decay of the main rotor rpm. Thereafter, insufficient rotor rpm remained to cushion the touchdown. Also, because of down-sloping terrain, the distance between the helicopter and ground level increased seconds before landing, thereby increasing the helicopter's absolute altitude and contributing to a hard impact with the ground. The fuel control unit and the power turbine governor were removed from the engine and taken to a test facility where they were installed on calibrated test benches and tested in accordance with the manufacturer’s test procedures. The results of the fuel control unit test showed that the internal parts all worked properly; however, the fuel flow at every test point was below the specified minimum limit. The results of the power turbine governor test revealed that the unit was out of specified limits at each test point. The discrepancies noted would affect the fuel flow at the high end of the schedule and is indicative of an improper rigging procedure. Examination of the unit showed that the maximum stop setting had been adjusted in the field and that the travel was set at 80 degrees instead of the required 86 degrees. While some of the discrepancies found during the tests of the fuel control unit and the power turbine governor are not serious, those at the low end of the fuel schedule are of particular concern. The minimum fuel flow, idle, and the cut-off settings were found to be below the manufacturer’s specified minimum limits; when combined with hysteresis, or the lagging of a physical effect on a body behind its cause, the chances of insufficient fuel flow being delivered to the engine during any engine deceleration maneuver (i.e., moving the throttle out of the flight detent) increases dramatically. When the FAA inspector moved the fuel flow control lever as the pilot was manipulating the collective during the beginning of the autorotation, it is likely that the unloading of the engine sent a signal to the fuel control unit to rapidly decrease the fuel flow at the same time the fuel control lever was being brought out of the flight detent and moved aft, which helped induce the flameout. Review of the operator’s flight and maintenance records found a pilot write-up that noted that about 9 months prior to the accident the engine had flamed out when the throttle was manipulated during the start sequence. The operator’s maintenance department was unable to find a reason for the event and released the helicopter back to service. The company pilots were aware that this particular helicopter had a “touchy throttle.” The pilot noted that, when operating the throttle lever in this particular helicopter, “you have to be gentle and slow with it as you retard the lever…if you pull it back to far or fast, it will shut off the fuel.” At least four prior instances of flameouts as a result of minor throttle movements were uncovered in deposition testimony of company pilots. All occurred on the ground and three of the flameouts happened as pilots were bringing the fuel flow control lever back toward ground idle during the post-flight engine cool down period. These instances were not documented in the maintenance records, and no records of attempted remediation were found.
HISTORY OF FLIGHT On December 16, 2009, about 1329, Hawaiian standard time, an Aerospatiale AS350BA(FX2), N87EW, operated by Sunshine Helicopters, Inc., Kahului, Maui, Hawaii, experienced a total loss of engine power during a simulated forced landing on the island of Maui about 1.3 miles southeast of the Hana (uncontrolled) airport. The helicopter impacted hard on uneven, downsloping, terrain and was substantially damaged. The commercial certificated pilot-in-command and the FAA inspector check pilot, who held an airline transport pilot certificate, were seriously injured. Visual meteorological conditions prevailed, and a company flight plan was filed. The instructional flight was performed under the provisions of 14 Code of Federal Regulations Part 91, and it originated from the Kahului Airport about 1257. The check pilot was the operator's assigned Federal Aviation Administration (FAA) principal operations inspector (POI). The purpose of the flight was for the POI to administer a 14 CFR Part 135.293 competency check ride to the pilot. Satisfactory completion of the check ride, and other requirements, would enable the pilot to continue operating Part 135 commercial air tours for his employer, Sunshine Helicopters, holder of an air carrier operating certificate. According to Sunshine's director of operations (DO), at the time of the flight the accident pilot was current in the operation of the helicopter. Several hours prior to the accident flight, the pilot had flown an air taxi flight in N87EW, and no maintenance squawks were noted. The helicopter operated normally, and it was dispatched for the pilot's use later in the day for his FAA check ride. The FAA coordinator reported to the National Transportation Safety Board investigator that performance of a simulated loss of engine power during this type of check ride was an authorized routine procedure used in evaluating the competency of airmen. The pilot and the FAA inspector were interviewed and provided written statements. The FAA inspector said that after he and the pilot completed the oral portion of the examination they then discussed what would be done on the flight. The check ride was to be a combined 14 CFR 135.299 and 135.293 check rides and a site air tour route review. The selection to fly to Hana reflected the need to integrate the tour check with the 135.299 and 135.293 check rides. During the briefing it was observed that there were Kona winds, fairly light but different from the usual northeast trade winds. A briefing was then conducted talking about the specific maneuvers to be performed during the check ride: 1) confined area pinnacle approaches, 2) site specific operations, 3) simulated engine power loss with a autorotation forced landing to 100 feet above ground level (agl) with a power recovery before touchdown, and 4) settling with power. The maneuvers would be done either going to or coming from Hana. The inspector said the helicopter flew fine throughout the entire flight. The inspector noted that they were about 3,000 feet mean sea level (msl) about 1 mile south of the Hana airport when he said "simulated forced landing" to the pilot. He said there is no defined flight idle position to put in the Fuel Flow Control Lever (which is basically the throttle; hereinafter referred to as the FCL) in to ascertain the power setting. He brought the throttle out of the full open flight run position detent and just aft enough back toward flight idle to keep it from springing back into the gate. The purpose of moving it out of the gate was to be certain the engine is not supplying power during the maneuver. He said that he and the pilot had briefed this prior to the flight and that they were to recover with power before getting too low. The pilot responded to the simulated loss of power by putting the collective down and he turned toward the Hana airport. The inspector questioned the pilot if he thought they were going to make it to the Hana airport. At that point he believed that the engine was no longer operating and he moved the FCL all the way forward back into the flight detent, but it appeared they were not going to get the engine back. The pilot did try to restart, but it was quickly evident that a restart was not an option due to the rapidly approaching ground. He said they were both looking for a suitable field, but there were not a lot to choose from. The only suitable area was the field the accident occurred in. He said the pilot did a good job trying to get to the field. The inspector said he remembers making a call to check the airspeed during the descent. . The pilot said it was a clear, VFR day with clouds at 2,500 to 3,000 feet. There were some scattered clouds in the area. The wind was light and variable in direction. The visibility was 5 to 10 miles. There is no weather reporting station at Hana. He said the flight began and they cruised about 1,500 feet along the shore line and went into the Hana Manu to Kano Falls, where he did site specific training and standoffs, then a confined area landing. He initiated the takeoff and they headed toward the coast line. Then, not long after that, the FAA Inspector initiated the simulated engine failure. The pilot said he became aware that the simulated forced landing had turned into a real forced landing when the helicopter yawed and then he saw the generator warning light illuminate. He said he and the FAA inspector talked briefly about making it to the Hana airport once they realized the engine had stopped running, but then he realized they did not have enough altitude. He entered a normal autorotation and chose a landing spot to his right. He looked at the throttle and initiated a relight by pushing the (FCL lever) throttle forward and the T4 temperature gauge went to 855 degrees. Since that was over limits, he backed the FCL off to 700 degrees and then due to their altitude, was more focused on getting to the landing spot. He aborted the restart because the helicopter was getting low, about 1000 feet agl. There were not a lot of spots to choose from. There was a lot of grass and trees in the area. Since trees can penetrate the helicopter structure, he picked the grass area that they ultimately impacted. During the final part of the descent, about 150 feet agl, there was a low rotor rpm horn just before touchdown. The pilot said he did not see the FAA inspector reach over to the throttle quadrant, nor did he see him retard the FCL to begin the simulation. The FAA inspector said something like, "simulated engine failure is beginning". He said he was not surprised when the simulation began, and that he was expecting a simulated engine failure maneuver to be performed during the check ride. During this check ride, this was the first simulated engine failure that the FAA inspector had given him. He said that prior to the maneuver the FAA inspector was trying not to distract him and was very straight forward and was very open for questions. The pilot described movement of the FCL lever (throttle) in this particular helicopter as “you have to be gentle and slow with it as you retard the lever. If you pull it back to far or fast, it will shut off the fuel.” PERSONNEL INFORMATION Pilot-in-Command, Sunshine Helicopter The pilot, age 42, holds a commercial pilot certificate with a rotorcraft-helicopter and helicopter instrument ratings, the most recent issuance of which is dated June 8, 2004. In addition he held a flight instructor certificate with a rotorcraft helicopter rating that was issued on September 10, 2004. His most recent medical certificate, a second class, was issued on March 21, 2009, without limitations or waivers. According to the operators records, the pilot had accrued a total flight time of 4,458 hours, all in rotorcraft, with 662 in the AS350BA(FX). His most recent 14 CFR 135.293 and 135.299 checks were accomplished on January 15, 2009, in an AS350B2. In the 15 days prior to the accident, the pilot had 9 days scheduled time off from work and had flown 20 hours in 6 duty days in air tour operations. Passenger, FAA Principal Operations Inspector The left seat occupant, age 51, is a Federal Aviation Administration operations inspector in the Honolulu Flight Standards District Office, who is the assigned Principal Operations Inspector for the operator. He holds an Airline Transport Pilot certificate with an airplane multi engine land rating and type rating in the Shorts SD-3. His certificate is also endorsed for commercial pilot privileges in single engine land airplanes, rotorcraft helicopters, gliders, and instrument-helicopters. The most recent issuance of this certificate was dated April 15, 2009, with the addition of the Shorts SD-3 type rating. At that time, the inspector reported a total flight time of about 10,800 hours, with 6,345 accrued in rotorcraft helicopters. His most recent medical certificate, a second class, was issued on December 4, 2009. In an interview, the inspector was asked about his total experience in the AS350 BA, FX conversion with the Honeywell engine. He stated that he has given one other flight check, and further noted that he is not really aware of the difference between it and a BA model. He said he became aware it was an FX conversion during the oral. He said he did readily know the differences between the FX and BA models and could not recall if he ever received training in this specific FX model. When asked if he had ever performed a check flight in AS 350 FX conversion he replied that he had given other check rides in the FX model at one other operator. The inspector was asked to explain how a simulated forced landing was performed on those checks. He replied that he would first announce the simulated forced landing so the pilot can lower the collective and set up the helicopter for the autorotation, then he would bring the throttle out of the flight gate to ensure that the engine was not providing power to the rotor. The maneuver was terminated between 300 and 500 feet agl. During the procedure he looks at the rotor rpm, airspeed, EGT gages, plus monitoring the outside situation to ensure the pilot is going toward the selected landing area. The inspector said he has seen the company training manual and had received refresher training in a Robinson R44 and a Bell 206. With regard to the training manual, he said that during a check ride if a pilot would question the maneuver during the before the flight briefing, the manner of the item’s performance would be changed. AIRCRAFT INFORMATION The helicopter, serial number 2116, was manufactured by Aerospatiale in 1988 as an AS350B model with a Turbomeca Aerial 1B engine. The helicopter was sold to a customer in Japan and was registered there as JA9761. In March of 2000 it was imported into the United States from Japan and registered under its current FAA civil registration number. In November of 2002, the original Turbomeca Aerial 1B engine was removed and a Rolls-Royce Allison 250-C30M was installed in accordance with a Soloy Corporation Supplemental Type Certificate. On December 16, 2008, the Allison engine was replaced with a Honeywell LTS101-600A3-A engine, serial number LE46110C, under the provisions of Supplemental Type Certificate SR02295NY. The modifications changed the helicopter’s designation from a AS350B to a AS350BA (FX2). Review of the maintenance records disclosed that a 100-hour inspection was endorsed in the airframe logbook on December 2, 2009, at an airframe total time of 18,789 hours. The maintenance records disclose that the engine was a Honeywell LTS101-600-3A, which was manufactured in September of 1984. The most recent 100-hour inspection for the engine was endorsed on December 1, 2009, at a total time since new of 7,725 hours, with 988 hours since the last major overhaul. The major overhaul was noted as completed on December 26, 2007, by the Honeywell facility in Greer, South Carolina, at a total since new of 6,737 hours. At that time, an overhauled Fuel Control Unit and Power Turbine Governor were installed on the engine. After the overhaul, the engine was sold as part of the BA(FX2) conversion kit to Sunshine Helicopters. The maintenance records have no entries reflecting removal, replacement, adjustment, or maintenance actions for either the Fuel Control Unit or the Power Turbine Governor after installation on the engine at engine overhaul. Review of the component record card for the Fuel Control Unit identified it as serial number 84490022. The unit was overhauled at Precision Fuel Components on November 20, 2007 and installed on the engine at the time of the engine overhaul by Honeywell, Greer, South Carolina. The final acceptance test sheet prepared by Precision Fuel Components show the unit passed the test on November 16, 2007 and met all test specification points. Review of the component record card for the Power Turbine Governor identified it as serial number 31248. The unit was overhauled at Precision Fuel Components on November 16, 2007 and installed on the engine at the time of the engine overhaul by Honeywell, Greer, South Carolina. The final acceptance test sheet prepared by Precision Fuel Components show the unit passed the test on November 16, 2007 and met all test specification points. The company flight record sheets for the helicopter were reviewed. From the 100-hour inspection on December 2 until the date of the accident, no maintenance discrepancies or maintenance actions were recorded on the daily flight sheets. A more comprehensive review of the historical flight sheets found an entry on March 14, 2009, noting a pilot discrepancy write-up of an uncommanded flame-out of the engine. The pilot entry stated, “after lighting off at 42 percent Ng (gas generator speed), I advanced the fuel toward idle, at 60 percent Ng engine flamed out.” The listed maintenance corrective action states: “Inspect compressor inlet for obstruction, inspect pneumatic and fuel lines for loose fittings and chaffing okay, check throttle for proper position and movement okay. Engine started normal power checked normal. Possible flame out due to water accumulated at bottom of inlet housing after heavy rain. Aircraft okay for service.” The helicopter had been converted from its original manufactured AS350B type design. In part, the conversion involved installation of a Honeywell LTS101-600A-3A engine, modification of its electrical system and engine performance gauges, installation of a tail boom strake, and installation of modified tail rotor blades. The principal changes, commonly referred to as an "FX" conversion, altered the helicopter's operating parameters as indicated by supplements included in the helicopter's flight manual. A control quadrant is mounted on the floor between the two front seats. Three lever controls are located on the quadrant. The most left hand lever (looking forward from behind the quadrant) is the Rotor Brake Control. The middle lever is the Fuel Flow Control; on top of the handle the starter button control is located. The most right hand lever is the Emergency Fuel Shut-Off control. The Fuel Flow Control lever operates in a longitudinal track. Two gate detents are located at the extreme ends of the track, with the forward detent the “full open” or flight run position, and the most aft detent the “closed” or idle cut-off position. Movement of the lever between the detents meters fuel to the fuel control unit mounted on the engine and controls the speed of the engine between the 100 percent at the open position down to engine stopped at the closed position. This model helicopter does not have a minimum flight or ground idle detent for the fuel flow control lever on the power quadrant. The investigation found that the engine had inadvertently (uncommanded) shut down four prior times, including the March 14, 2009 event noted above. The three other events were disclosed in sworn testimony by company pilots. All four of the events occurred when the helicopter was on the ground, and three of them occurred when pilots were bringing the FCL back to slow the engine speed to ground idle during the post fl
An uncommanded engine shutdown due to an improperly calibrated fuel control unit (FCU) and power turbine governor (PTG). Also causal was the operator’s inadequate maintenance practices and procedures that failed to properly assess and correct the FCU and PTG irregularities/deficiencies. Contributing to the accident was the Federal Aviation Administration inspector’s selection of an area for the simulated engine failure that offered limited choices for a full-touchdown autorotation.
Source: NTSB Aviation Accident Database
Aviation Accidents App
In-Depth Access to Aviation Accident Reports
