EUROCOPTER AS350-B2

Full Narrative

HISTORY OF FLIGHT

On December 19, 2010, about 1000 central standard time, a Eurocopter AS350-B2 helicopter, N549AM, sustained substantial damage when it landed hard after a loss of power during takeoff from a helipad at La Monte, Missouri. The pilot, flight nurse, and flight paramedic received serious injuries. The helicopter was operated by Air Methods Corporation under the provisions of 14 Code of Federal Regulations (CFR) Part 91 as a medical flight and was departing for an on-scene patient transfer at Fristo, Missouri. Visual meteorological conditions prevailed at the time of the accident. The flight was on a company visual flight rules flight plan.

The pilot reported that he arrived about 0645 for his shift, which started at 0700. He performed the required preflight planning activities, which included obtaining a flight release, checking the aircraft maintenance report, reviewing the logbook, checking weather, conducting a crew brief, and performing a preflight inspection of the helicopter. The pilot reported that he conducted a thorough preflight of the helicopter, which included checking the air filter for ice and snow. He reported that there was no visible moisture on the helicopter during the preflight.

Air Methods trip sheet documents indicated that the pilot was notified of an emergency medical flight (EMS) flight about 0711, and the helicopter departed about 0724. The helicopter arrived at the patient location in Lincoln, Missouri, at 0737, and departed with the patient on board about 0800. The helicopter arrived in Columbia, Missouri, about 0827. While in Columbia, the pilot fueled the helicopter with 60 gallons of Jet A fuel before the return flight to La Monte. The helicopter departed Columbia about 0906 and arrived at La Monte about 0934. After landing at La Monte, the flight crew went into the office to complete the required paperwork and flight debrief. The pilot reported that they did not encounter any rain or snow showers during the three flight legs, and it was not raining or snowing at La Monte when they landed.

The Air Methods trip sheet documents indicated that the pilot was notified of a second EMS mission, the accident flight, about 0948. The pilot reported that he did a weather check and accepted the flight. The flight crew boarded the helicopter and the pilot completed the start and takeoff procedures without incident. The helicopter was pointed directly west as the pilot began to lift straight up in a slow climb to about 100 to 125 feet. He applied left pedal to point the aircraft to the southeast. As he completed the pedal turn, he confirmed that he was clear vertically of a tree that was directly east of the helipad. As he began a slow transition to forward flight he heard a loud "bang." The pilot immediately reacted to the sound by slightly lowering the collective and turning to the right to get back to the helipad. He lowered the nose briefly and then began to flare the helicopter as he pulled up on the collective. The helicopter hit hard on the helipad, which spread the helicopter's landing skids so that the helicopter was lying on its belly. The fuel tank ruptured during the impact. The pilot noticed a flame about 2 inches in height about 3 feet from the helicopter, but the flame went out quickly and there was no postimpact fire. The pilot turned off the main electrical push button and began to yell "get out" to the flight crew. He crawled about 15 to 20 feet away from the helicopter. One of the first responders told the pilot that there was still electrical power to the helicopter. The pilot explained to a fireman how to turn off the "DC Bat" button that was located next to the main electrical button, which the fireman then turned off.

In her account of the accident, the flight nurse, who was sitting in the right rear seat, reported that the helicopter lifted off the pad normally to about 100 feet, then the engine went quiet. She did not hear any loud noise or bang. She heard the pilot yell, "Brace yourself" and she got into the crash position. She recalled that she was in extreme pain after the helicopter impacted the ground. She refused assistance in getting out of the helicopter until trained medical personnel arrived and placed her on a long board before exiting the helicopter. She reported that she immediately notified her husband of the accident on her cell phone then notified the EMS communication specialist that the accident had occurred. The time of the call to the communication specialist was about 1000.

The flight paramedic, who was sitting in the left rear seat, reported that everything seemed normal until the engine stopped making noise. The helicopter was above tree top level when the engine stopped. She braced for impact. The impact caused all the doors to open, and fuel was on the ground. She had seen sparks from the skids hitting the concrete, but she did not see any flames. Although she was able to get her feet on the ground, she could not walk and needed assistance from a first responder to get out of the helicopter.


PERSONNEL INFORMATION

The pilot was a 53-year-old airline transport pilot with a helicopter rating and a helicopter instrument rating. He held a second-class medical certificate issued on March 3, 2010. He had about 7,928 total helicopter flight hours, which included about 296 hours in the AS350-B2 helicopters. In the 90 days preceding the accident, he had flown 34 hours, 9 hours of which were within the last 30 days. His most recent 14 CFR Part 135 airman competency/proficiency check was successfully accomplished on August 5, 2010.



AIRCRAFT INFORMATION

Manufactured in 2007, the helicopter was a Eurocopter AS350-B2, serial number 4339, with a three-blade main rotor, a conventional tail rotor, and skid type landing gear, and was powered by a single 625 maximum continuous shaft horsepower Turbomeca Arriel 1D1 free turbine engine. The helicopter was equipped with an Air Methods EMS Interior and configured for medical transport of a single patient on a litter and two medical flight crewmembers. The litter was located on the left side of the helicopter and extended from the left side of the cockpit into the left side of the cabin.

The helicopter's flight manual listed the maximum gross weight as 4,960 pounds. At the time of the accident, the helicopter was fueled to a 60 percent fuel load and was operating within the weight and balance limitations.

A review of the helicopter's maintenance records revealed that it had 1,288 total hours at the time of the accident. The most recent maintenance inspection, performed on December 17, 2010, was part of the Air Methods Approved Aircraft Inspection Program (AAIP), which included 50-hour, 100-hour, and 600-hour airframe and engine inspections. The inspection tasks cards did not contain any actions that required inspecting the main airframe fuel filter or the Le Bozec fuel filter assembly. The Air Methods task cards referred directly to Eurocopter's task cards with no additional information. The mechanics who performed the inspections reported that they did not depress the plunger on the filter bowl of the fuel filter assembly during the inspections. Maintenance records indicated that the fuel filter cartridge was replaced during the 1,000-hour airframe inspection at a total airframe time of 1,008 hours, about 280 hours before the accident flight.

The Air Methods mechanic who was assigned to the La Monte base reported that he started his employment with Air Methods on December 1, 2010. He received on-the-job training from December 1 to December 8. He performed the aircraft daily maintenance checks on the accident helicopter on December 9, 10, 13, and 14 by himself. On December 15, he performed the daily maintenance checks on the accident helicopter but with assistance from another mechanic. He reported that he had never been instructed to depress the plunger of the fuel filter assembly nor had he been instructed not to depress the plunger, since there was no guidance at the time that it was not to be depressed. He could not specifically recall if he depressed the plunger on the accident helicopter, but, if he had, it would have been on December 15 during the aircraft daily maintenance check. He stated that he would have turned on the fuel pumps while depressing the plunger.


Vehicle Engine Multifunction Display (VEMD)

The accident helicopter was equipped with a vehicle engine multifunction display (VEMD), which is a dual screen display that provides vehicle and engine information to the pilot during flight (known as flight mode). The VEMD also has a maintenance mode, which stores nonvolatile data for specified vehicle and engine exceedances and/or overlimits and VEMD-system-related hardware or communication failures. The VEMD also records and stores flight reports, which can be found in the maintenance pages. A flight report begins when aircraft electrical power is applied and engine generator speed (NG) reaches 60 percent. The flight report ends when the NG decreases below 50 percent with aircraft electrical power still applied. During a normal landing and shutdown, the VEMD will display the flight report for the most recent flight when NG drops below 10 percent and main rotor rpm (NR) decreases below 70 percent.

Fuel Filter Bypass Operation

The AS350-B2 is equipped with a Le Bozec fuel filter which contains a stainless steel filter element. The fuel filter is located forward of the firewall in the main gearbox compartment. The filter unit includes a bypass valve, a visual indicator of bypass, and a drain. An annunciator light labeled F. FILT illuminates on the Warning – Caution – Advisory Panel to indicate a clogged filter and an impending bypass. Contaminated fuel causes impurities to build on the filter cartridge. This gradually reduces the flow through the filter. Pressure then increases at the filter inlet and decreases at the filter outlet. When the pressure differential reaches 206 mb, a pressure switch illuminates the amber F. FILT annunciator to alert the pilot the bypass is likely to open. The bypass valve opens when the pressure reaches 350 mb +/- 50 mb. The engine still receives fuel, but it is contaminated fuel and unfiltered. The annunciator light remains illuminated as unfiltered fuel flows to the engine. On engine shutdown, the bypass open indicator (red pop-up button on the filter housing) is visible.

The emergency procedure when the F. FILT illuminates is:

Reduce engine power.

- If light goes out, continue flight at reduced power.

- If light remains on, land as soon as possible.

There was no indication that the F FILT annunciator light had illuminated during the accident flight, and the pilot reported that he did not see any light illuminated before the engine flameout.



FDC Aerofilter

The helicopter was equipped with an FDC Aerofilter (Supplemental Type Certificate [STC] SR00811SE). Located on top of the helicopter aft of the main rotor mast and transmission, the after-market unit filters the outside air before it enters the engine's intake plenum. During STC certification testing, it was demonstrated that 70 percent of the surface area of the filter needed to be blocked before engine limitations were reached—about double the blockage necessary to activate the differential pressure switch and the low pressure annunciator light in the cockpit. The system has a pilot-activated engine alternate air switch that opens the alternate air doors if the low pressure annunciator light illuminates. There was no indication that the light had illuminated during the accident flight, and the pilot reported that he did not see any light illuminated before the engine flameout.

The FDC low pressure annunciator light was inspected on March 31, 2010, during a 12-month calendar inspection. The next inspection was due in March 2011.

The Limitation section of the Rotorcraft Flight Manual Supplement for the AS350-B2 helicopter equipped with the FDC Aerofilter provided the following Takeoff limitation: "Takeoff with LOW INLET PRESSURE annunciator light illuminated…..PROHIBITED."



Engine Flameouts

The engine's ignition system is only in operation during the starting sequence. Once started, combustion is continuous and self-sustaining as long as the engine is supplied with the proper fuel-to-air ratio. If the rich limit of the fuel-to-air ratio is exceeded in the combustion chamber, the flame will extinguish. This condition is referred to as a rich flameout. It generally results from very fast engine acceleration, where an overly rich mixture causes the fuel temperature to drop below the combustion temperature. It also may be caused by insufficient airflow to support combustion, which may occur as a result of a blocked engine inlet or inlet barrier.

An interruption of the fuel supply can also cause an engine to flameout. Such an interruption may be due to prolonged unusual attitudes, a malfunctioning fuel control system, blocked fuel supply, air introduction into the fuel delivery system, turbulence, icing, or fuel exhaustion.

The Arriel 1D1 engine is not equipped with an auto-ignition system, nor is such a system required by regulation. If a flameout occurs, the engine will not automatically restart. The helicopter's aluminum intake plenum is not equipped with a de-icing or anti-icing capability.

Eurocopter identified the following general causes for a sudden loss of engine power during an un-commanded in-flight shutdown:

1. Loss of required conditions to ensure engine running: loss of fuel.

2. Loss of required conditions to ensure engine running: air supply.

3. Untimely helicopter stop or idle order due to a system malfunction and/or pilot action.

4. Engine power loss due to Arriel 1 engine failure.

5. Engine attachment degradation.



Air Introduction

The fuel boost pumps are equipped with check valves that incorporate bleed ports to allow pressure in the fuel lines to be bled off after engine shutdown. If the fuel system is opened, such as during filter servicing, air can be drawn into the system as gravity returns fuel to the tank through the check valve bleed port. Air can also enter the system if the Le Bozec fuel filter assembly is drained, or if there is an external leak which becomes an "air entry" with the boost pumps off (the examination of the accident helicopter revealed no external air leak). The AS350 B2 Rotorcraft Flight Manual and the maintenance manual make no reference to a daily draining procedure of the Le Bozec fuel filter assembly.

METEOROLOGICAL INFORMATION

About 0953, the surface weather observation at Sedalia Regional Airport, Sedalia, Missouri, located about 10 nautical miles east of the accident site, was wind from 170 degrees at 5 knots; 7 miles visibility; overcast skies at 3,200 feet; temperature 3 degrees Celsius (C); dew point -3 Celsius; and altimeter 30.08 inches of mercury.

The pilot reported that snow showers were in the forecast, but the actual weather was better than the forecast. During the preflight inspection, no moisture was visible on the helicopter, and no snow or ice was present on the FDC Aerofilter. He reported that there were no snow showers or other weather issues during the flights to and from Columbia, Missouri. He reported that there were no rain or snow showers at La Monte before the accident flight.



HELIPAD INFORMATION

The Air Methods helipad is located at 200 West Front Street, La Monte, Missouri. Orange balls mark the wires near the 30 feet by 30 feet concrete helipad, but there are no red lights to mark the light poles, wires, and trees near the touchdown and lift-off (TLOF) area. There are no lights bordering the TLOF. A 3,000 gallon above ground fuel tank is located about 35 feet from the TLOF. A windsock is present, but it is not lighted. There is no approach and takeoff path to the helipad, which results in the helicopter needing to enter a high out-of-ground-effect hover for takeoffs and landings. There is no safety area extending beyond the final approach and takeoff area.



FLIGHT RECORDERS

The helicopter was not equipped with a cockpit voice recorder or flight data recorder.

WRECKAGE AND IMPACT INFORMATION

The accident helicopter came to rest, upright, on a westerly heading. An on-site examination of the cockpit found the fuel flow control lever (FFCL) in the flight detent. The full range of travel and continuity were confirmed from the FFCL to the fuel control. The fuel shutoff lever was found forward and lock-wired in the "open" position. The hydraulic cut-off switch was in the forward, guarded position. The fuel pump switches were found in the "on" position.

Flight control continuity was confirmed from the cyclic, collective, and anti-torque pedals to their respective control input mechanisms; however, full movement was not possible for the cyclic and collective due to impact damage of control tubes around the mixing unit. The main transmission rotated freely with a corresponding rotation of the tail rotor. The engine to main transmission drive shaft was intact and remained connected at both ends via flex couplings. The main rotor system appeared intact, and all hardware appeared to be properly attached. The main rotor blades did not exhibit any obvious impact damage. The tailboom was intact and was not struck by the main rotor blades. The lower vertical fin on the tailboom was crushed and bent to the right. The tail rotor drive system was found intact. The tail rotor blades exhibited evidence of impact at the tips and flapping (consistent with low rpm) damage at the hub. Both skids separated from the cross tubes above the attachment points, which was consistent with high vertical velocity impact with a slight forward airspeed and/or nose-down attitude. The fuel tank was compromised due to impact. Fuel was present in the lines and filter. The fuel filter appeared clean, and the bypass button was not extended.

The examination of the engine revealed that the engine rotated freely and did not exhibit any visible damage. The fuel control unit (FCU) fuel flow pointer was at 52 degrees (flight) and the anticipator pointer was at 30 degrees. All fuel, oil, and air lines and connections were tight. The gas generator (axial and centrifugal compressors and the first- and second-stage turbines) rotated by hand. Power turbine and power shaft continuity and rotation were confirmed. The freewheeling assembly rotated smoothly by hand in both the driving and freewheeling direction. No foreign object damage or distortion was noted on the axial compressor blades. The power turbine blades were intact. Gear train continuity was confirmed. The magnetic plugs and chip detectors were clean.



SURVIVAL ASPECTS

The National Transportation Safety Board's (NTSB) "Survival Factors Specialist's Factual Report of Investigation" can be found in the docket material associated with this investigation.



TESTS AND RESEARCH

Fuel Tests

Fuel samples, oil samples, and the fuel filter element from the accident helicopter's fuel filter assembly were sent to Wright-Patterson Air Force Base in Ohio for testing. The tests indicated that the fuel appeared to be typical of jet fuel, and the oil appeared to be typical of MIL-PRF-23699 oil. No anomalies were noted. The fuel filter element examination found no water in the filter or other anomalies. The particulates found in the fuel filter element included insect parts, wood fibers, a piece of a paper filter, magnesium aluminum allow pieces, paint chips, flourosilicone gasket pieces, black and clear plastic pieces, chlorocarbon tape, and glass fibers.


Fuel Pumps

The helicopter was equipped with two fuel pumps. The fuel pumps were removed from the helicopter and tested at American Eurocopter with NTSB oversight. No anomalies were noted on either pump.


Vehicle Engine Multifunction Display

The VEMD was removed from the helicopter for further investigation at American Eurocopter in Grand Prairie, Texas, with NTSB oversight. Power was applied to the accident helicopter VEMD on a test bench. The VEMD's maintenance mode flight report pages indicated that the accident flight was VEMD flight number 2547, which lasted about 3 minutes. There were no failures or overlimits recorded for the accident flight.

American Eurocopter reported that depending on the circumstances of an accident, some system and/or hardware failures may be recorded on the accident flight report as a result of damage sustained during the impact, for example, outside air temperature (OAT) sensor failure. There were no OAT impact-related failures recorded. Therefore, if electrical power was still applied up to the time of the accident, it was likely that NG decreased below 50 percent (triggering the end of the flight report) and that an engine flameout had occurred before impact.

Turbomeca Engine Inspection and Test Cell Engine Run

The engine was shipped to Turbomeca USA in Grand Prairie, Texas, for inspection, with NTSB oversight. The engine was installed on the test bed, and tests were performed with the engine's original equipment. The engine was run for 51 minutes. The engine is normally run at an external boost pump pressure of 90 kiloPascals (kPa). These tests were conducted with a boost pump pressure of 60 kPa, which corresponds to the boost pump pressure of the two boost pumps. During the tests, boost pump pressure was decreased to 0 kPa and the engine continued to operate. All tests, except the extinction test, were conducted in the "position mode" (fuel regulation controlled by the FCU). The extinction test was performed in the "speed mode" (fuel flow selected by the test cell operator). This test requires the fuel flow to be reduced from 175 liters per hour (l/h) to ground idle within 1 second. The engine did not flame out during the rapid deceleration. This test was performed twice. The engine operated within the original equipment manufacturer's specifications for all tests performed. The engine was shipped to Turbomeca France for further inspection.

Turbomeca France Engine Inspection

The engine was disassembled at Turbomeca France to remove the fuel system components and the axial compressor for further investigation, with oversight of the Bureau d'Enquetes et d'Analyses (BEA). Tests were performed on the following units: FCU, starting drain valve, overspeed drain valve, pressurizing valve, bleed valve, and start electrovalve. A three-dimensional check of the axial compressor was conducted. The various examinations did not reveal any findings that would explain the engine flameout. Subsequent to this examination, the fuel injection manifold was flow-checked at the Turbomeca USA facility. There were no discrepancies.

Airframe Fuel Circuit Test

The accident helicopter's fuel circuit was tested at an aircraft storage facility near Wrightsville, Missouri, on April 26 and April 27, 2011, with NTSB oversight. The purpose of the examination was twofold: 1) to verify the operation of the airframe fuel circuit, i.e., fuel pressure, fuel flow, and the absence of any fuel leaks, and 2) to determine the effect when the Le Bozec fuel filter assembly drain valve is operated or when the fuel filter cartridge is replaced during scheduled maintenance. Of particular concern was whether air can be introduced into the fuel lines through the fuel filter assembly and if the air can be purged during the engine start sequence and/or by normal engine operation. For test purposes, the fuel lines were changed with transparent plastic lines of the same internal diameter to allow for an unobstructed view of the fuel flow through the system.

The results of the test confirmed that the fuel circuit of the accident helicopter operated correctly. The tests also revealed that air was drawn into the fuel circuit between the fuel filter assembly and the fuel pumps when the fuel filter assembly drain valve was pressed without either of the fuel pumps operating. Air remained in the fuel lines even after the following procedures were performed: 1) turning the fuel pumps on for 30 seconds with a fuel flow of 50 liters per hour (l/h), which simulated the before engine starting purging procedure; 2) increasing the fuel flow to 175 l/h, which simulated the engine at takeoff power; 3) increasing the fuel flow to 275 l/h, which simulated the engine at max NG; and 4) increasing the fuel flow to 480 l/h, which simulated maximum fuel pump fuel flow.

Eurocopter Fuel Bench Test and Fuel Circuit Test

In May 2011, Eurocopter France conducted fuel bench tests, with BEA oversight, to determine the effect of air that is introduced into the fuel system via the fuel filter assembly. By using transparent flexible lines and a transparent fuel filter bowl, the air in the system could be observed during the tests. The test procedure, which simulated purging the fuel system after the fuel filter was replaced, or when the fuel filter assembly drain valve was pushed without the boost pumps "ON," revealed that some air remained trapped in the upper cavity of the fuel filter assembly. The volume of the fuel filter cavity measured 33 milliliters.

The additional tests performed on dedicated test benches, which simulated one sequence of engine starting and the takeoff (fuel flow rates at flight idle, maximum takeoff power, and maximum engine generator speed NG), after a fuel filter cartridge replacement, revealed that the volume of air that remained trapped upstream in the cavity was about 1 to 2 cubic centimeters after 7 to 8 minutes of run time.

Eurocopter France conducted similar tests of the fuel circuit using an actual AS350 B2. The tests revealed no significant difference in the fuel filter's behavior between the test bench configuration and the tests performed on the aircraft.

Turbomeca Fuel Control Unit Bench Tests

Turbomeca France, with BEA oversight, performed tests using a fuel control bench test rig. Transparent flexible lines and a transparent fuel filter bowl were used so that the air in the system could be observed during the tests. The test procedures used the fuel flow rates for the 30-second purging procedure for engine start (30 l/h), the 80 l/h fuel flow rate for idle, and 230 l/h fuel flow rate for maximum takeoff power. The tests revealed that air injection at the FCU inlet can disturb the FCU, but no fuel flow interruption occurred at the FCU outlet under nominal conditions. Under nominal settings, there appeared to be no risk of an in-flight shutdown. However, questions remained concerning the volume of air trapped in the upper cavity of the fuel filter bowl and whether there were conditions that could release the trapped air through the fuel bypass during engine operation, which prompted further tests.

Turbomeca USA Test of FDC Aerofilter Delta Pressure Test

The accident FDC Aerofilter was tested on an Air Methods AS350-B2 located in La Monte, Missouri, on December 20, 2011, by Turbomeca USA, with NTSB oversight. The purpose of the test was to establish a curve of the differential pressure between the ambient air and the engine intake with the accident helicopter's FDC inlet barrier filter installed. Five conditions were tested: 1) no filter installed; 2) new filter installed (no oil); 3) exemplar clean oiled filter; 4) the accident filter (dry); and 5) the accident filter(wet). The cabin was loaded with ballast and the power was increased from idle power to 90 percent NG. The test revealed that, at 90 percent NG, the pressure differential measured in kPas (6.9 kPa = 1 psi) was: 1) -0.901 kPa for no filter installed; 2) -1.188 kPa for new filter installed (no oil); 3) -1.267 kPa for the exemplar clean oiled filter; 4) 1.414 kPa for the accident filter (dry); and 5) -1.582 kPa for the accident filter (wet). The pressure differential at 90 percent NG between the best condition (no filter) to the worst condition (wet accident filter) was 0.681 kPa, or 0.1 psi.

Fuel Filter Assembly Examination

The fuel filter assembly was examined with BEA oversight at the Le Bozec manufacturing facility located in France. The tests indicated that the fuel filter performed according to the manufacturer's specifications. The fuel bypass "pop-up" activated at the proper pressure (384 millibars). It was also observed that the pop-up activated in an "immediate" fashion and that just before the bypass was activated, there was no visual movement of the pop-up indicator.

Turbomeca Engine Test Cell Results

On April 11 and April 12, 2012,Turbomeca France, with BEA oversight, performed tests on an engine test cell using an exemplar Arriel 1D1 engine. The test objectives were to activate the fuel bypass opening located atop the fuel filter assembly at various fuel flows, allowing the volume of trapped air to go directly to the engine and then analyze the disturbances to the FCU and engine parameters, monitor the engine stability (that is, whether it would flameout), and determine the associated effects on the engine.

The test cell run was conducted under the following conditions: Jet A fuel at 20 degrees C; 592 feet above mean sea level, stable electrical supply, and ideal air intake (that is, no air filter and no air flow disturbances due to rotor wash); no use of compressor outlet pressure (P2) bleed air; and no electrical load on the engine. The test used a fuel filter assembly with different 8-micron cartridges, a transparent cartridge bowl, and transparent piping. The cartridges included new, clean, clogged, and artificially clogged cartridges. A metering jet was fitted inside the cartridges to simulate and adjust different clogging conditions. A flow meter device was placed at the FCU outlet (metered fuel flow). The test mixed different system configurations at different acceleration rates with different fuel filter cartridges.

The engine did not flameout during any of the 16 tests performed. The worst-case scenario observed during the tests occurred when the pre-clogged cartridge was used. During that test, once the filter differential pressure reached 41 kpa, the filter bypass fluctuated between open and closed for 0.5 seconds before remaining in full bypass. About 0.5 seconds after the initial filter bypass, the fuel pressure at the injection manifold (P_inj) dropped and nearly matched the compressor discharge pressure (P3 air). There was a fuel flow interruption lasting 5 milliseconds, a transient disturbance of the engine parameters, and a transient slower rate of power increase for about 5 seconds; however, the engine did not flame out.

Test of Intake Plenum Temperatures after Shutdown

On September 27, 2013, a test was conducted in Wasilla, Alaska, with NTSB oversight, using an Air Methods AS350-B2, N147LM, equipped with a FDC Aerofilter. The purpose of the test was to measure the temperature of the filter and intake plenum in 5-minute increments for a total of 30 minutes after engine shutdown. A digital temperature laser probe was used to measure the following temperatures: 1) the top side of the filter; 2) the bottom side of the filter; 3) the top of the plenum chamber; 4) the engine compartment near the intake plenum; 5) engine inter turbine temperature (ITT); and 6) outside air temperature (OAT). The test results indicated the following temperatures:

1. Pre-start: 1) top side of filter – 3.7 C; 2) bottom side of filter – 4.7 C; 3) top of plenum chamber – 5.0 C; engine compartment – 4.0 C; 5) ITT – 7 C; and 6) OAT – 7.5 C.

2. After shutdown: 1) top side of filter – 6.6 C; 2) bottom side of filter – 21 C; 3) top of plenum chamber – 27 C; engine compartment – 75 C; 5) ITT – 210 C; and 6) OAT – 7.4 C.

3. 5-minute interval: 1) top side of filter – 13 C; 2) bottom side of filter – 24 C; 3) top of plenum chamber – 25 C; engine compartment – 66 C; 5) ITT – 175 C; and 6) OAT – 7.4 C.

4. 15-minute interval: 1) top side of filter – 15 C; 2) bottom side of filter – 19 C; 3) top of plenum chamber – 21 C; engine compartment – 51 C; 5) ITT – 130 C; and 6) OAT – 8.0 C.

5. 30-minute interval: 1) top side of filter – 19 C; 2) bottom side of filter – 22 C; 3) top of plenum chamber – 22 C; engine compartment – 41 C; 5) ITT – 100 C; and 6) OAT – 8.3 C.

The reports documenting the examinations and tests performed during the course of the investigation can be found in the docket material associated with this accident.


ADDITIONAL INFORMATION

Eurocopter Information Notice No. 2351-I-28

On July 26, 2011, Eurocopter issued Information Notice No. 2351-I-28 concerning the fuel filter. The information letter stated the following:

The investigations and tests performed following two uncommanded in-flight engine shutdowns (whose reason has not yet been identified) have revealed the possibility of air being introduced in the fuel system by activating the drain located at the bottom of the filter unit assembly…

Furthermore, activation of the drain, without at least one of the two booster pumps operating, has no effect with respect to the objective (no drainage effect and therefore no evacuation of possibly present water), but may on the contrary lead to air being introduced in the system, because the fuel column is not leak-tight (according to the definition: check valve of the booster pumps is not leak-tight).

Although no cause-effect relation has yet been identified between this potential introduction of air in the fuel system and the events encountered, EUROCOPTER reminds you of the following:

– The drainage operation on the fuel filter unit is not required in the documentation, except for the operation before opening the filter bowl…

– If for any reason, a draining operation is performed, it must be performed in any case with at least one of the two booster pumps operating.

In parallel, EUROCOPTER and TURBOMECA continue the investigations, in order to determine the causes of these uncommanded in-flight engine shutdowns, by analyzing the possible influence of the presence of air, and considering that the aircraft concerned were also equipped with additional filters (STC: Supplement Type Certificate) or filter cartridge PMA (Part Manufacturer Approval).

EUROCOPTER therefore asks you to comply with the instructions specified in the maintenance documentation and reminds you that section 4 of the Flight Manual requires operation of the booster pump during 30 seconds before engine starting.


Previous Engine Power Losses

On December 12, 2010, concurrent with this investigation, the Transportation Safety Board of Canada (TSB) began an investigation of an AS350-B2 (Canadian registration C-FORS, serial number 4001) that had an engine power loss event near Pickle Lake, Ontario, Canada (TSB Accident No. A10C0214). The information surrounding the two similar occurrences was shared between the NTSB and TSB to facilitate common testing. In that occurrence, the Turbomeca Arriel 1D1 engine was run with no anomalies noted. The fuel systems of the two accident helicopters were identical with the exception of the Pickle Lake helicopter having an additional Eurocopter Canada Limited STC anti-icing fuel filter and the standard Le Bozec airframe fuel filter equipped with an FAA/PMA Puroflow/WFC filter cartridge. The outside air temperature in the Pickle Lake accident was –29 degrees C (-19 degrees Fahrenheit).

Eurocopter Information Notice No. 2030-I-00

On January 28, 2009, Eurocopter issued Information Notice No. 2030-I-00 concerning engine flameouts or damage occurring shortly after takeoff. The notice stated the following:

The experience acquired on our helicopters has revealed some cases of engine flameout or damage occurring shortly after takeoff. When the helicopters had previously been subject to cold weather in snowy or rainy conditions, and parked outside.

A turbine engine has a good rainwater or falling-snow absorption capacity in continuous operation. On the other hand, the engine is sensitive to a "sudden quantity" [emphasis in original] of water, snow or ice, because this quantity (even limited) corresponds to a very high instantaneous concentration exceeding its absorption capacities. This can occur because snow and ice can build up in the engine air intakes and plenums when the aircraft is on the ground without the engines operating or when the engine is at a low power setting for an extended period.

The factors associated with engine flameouts covered by Information Notice No. 2030-I-00 are well known and documented. Neither the La Monte accident helicopter, N549AM, nor the Pickle Lake accident helicopter, C-FORS, exhibited the kind of engine damage that was associated with Information Notice No. 2030-I-00. A search of aviation accident databases indicated that there were no similar related events when compared with the La Monte and Pickle Lake accident details.