At approximately 1500 hours the pilot conducted a flight with an external load from Lake Mackenzie to Lake Nameless Hut. He then landed to embark three passengers for transfer to another hut. Two of the passengers occupied the remaining two front seats and the third passenger occupied the cabin right rear seat. The pilot reported that, while on the ground, the fuel low level advisory light had momentarily illuminated, but that he attributed that illumination to the distribution of fuel in the tank due to the slope of the ground. At that time, he reported also noting 100 lbs (86.9 lbs useable) of fuel indicated on the fuel quantity indicator. At 1515 hours, the helicopter departed Lake Nameless Hut for Tom Whitely's Hut, which was located approximately 5 km to the north-east. A passenger reported that, during that flight, a caution advisory light had illuminated. The investigation could not confirm the identity of that light. Having overflown the hut landing area, the pilot initiated a left descending turn to the south prior to commencing an approach to land.
The pilot reported that at 1524 hours, as the helicopter descended through about 200 ft above ground level (AGL), and at a speed of 70 kts, the main rotor speed decreased and the engine auto reignition advisory light illuminated. Assessing that the engine had lost power, the pilot reported that he initiated an autorotation to land. He stated that "...the aircraft landed normally, although heavily". He reported that, after the initial ground contact, the aircraft was "...suddenly rotated through 180 degrees". That rotation was reported by the pilot to be as a result of entanglement with an unseen "...little wire or whatever hooked the aircraft".
The helicopter was destroyed by impact forces. There was no fire. The pilot and three passengers sustained serious injuries.
The helicopter impacted the ground heavily on the rear of the right landing skid, collapsing it and separating the left landing skid. The fuselage impact ground scar measured about 2 m in length. The main rotor blades struck the ground and severed the tail boom. The helicopter came to rest about 7 m and bearing 200 degrees magnetic from the initial impact point, facing the direction from which it had approached, and lying on a fence line. There were no ground impact scars between the fuselage impact ground scar and the helicopter's final position. The right side rear fuselage floor area sustained severe impact damage and the right fuel cell bladder was ruptured.
The forward section of the cockpit was destroyed during the impact sequence. The two front seat passengers were ejected from the helicopter, in the direction of flight. On-site inspection found the pilot's and passengers' seat belts and attachment points intact and that the pilot's shoulder harness was separated at the harness-to-inertia reel strap buckle. There was no evidence that the passengers' seat belt buckles had received damage due to impact forces. The pilot and front seat passengers' seat structure was deformed and wrinkled. Information from the helicopter manufacturer indicated that a vertical impact force loading of the airframe in excess of 10 g would have been required to deform the seat structure in that manner.
The investigation determined that there was minimal rotation of the tail rotor driveshaft at ground impact. That was confirmed by the lack of any impact or rotary damage to the tail rotor blades. The engine output driveshaft was separated at the driveshaft lobes and displayed little or no rotation at the time of separation.
The external load long-line was found attached to the cargo hook. There was no evidence that the long-line had snagged on the ground, other obstacles or the helicopter prior to impact.
An old wire and timber post fence was located in the vicinity of the accident site. The fence was about 1 m high and aligned about 050/230 degrees magnetic. The fence was laterally displaced about 3.5 m from the initial impact point. The fence posts and wire exhibited no evidence of having been contacted prior to, or during the helicopter's initial ground impact.
Testing of components
Analysis of the helicopter fuel system determined that a common fuel-sending unit activated the fuel quantity indicator and fuel low level advisory light. The fuel-sending unit, fuel quantity indicator and fuel low level warning system were removed from the helicopter and tested. Testing indicated that those components were serviceable in accordance with the manufacturer's maintenance manual. The fuel low level advisory light illuminated at 35 lbs fuel indicated on the fuel quantity indicator, in accordance with the manufacturer's maintenance manual. Testing, disassembly and inspection of the engine fuel pump, fuel control unit, fuel nozzle, bleed valve and power turbine governor, revealed no anomalies.
The Bureau of Meteorology Area Forecast, valid at the time of the accident, indicated Visual Meteorological Conditions with moderate southerly winds. The pilot and passengers reported bright, sunny conditions and a light and variable southerly wind.
The pilot in command held an Air Transport Pilot (Helicopter) Licence, a Command Multi-Engine Instrument Rating and was endorsed on the Hughes 369E helicopter type. At the time of the occurrence, the pilot had accumulated a total of 3,565 flying hours, including 74.0 hours on type. He had flown 34 hours in the previous 90 days, of which 5 hours was on type. He was reported to be fit and well rested prior to the flight.
On the afternoon prior to the occurrence, the pilot completed a 0.5 hour proficiency check flight with the company Chief Pilot, in accordance with the company Operations Manual and CAO 20.11 appendix 4. It was reported that the check flight did not include external load or autorotation sequences. The pilot reported that he had significant prior external load experience, conducted in several helicopter types. He was unsure when he last practised an autorotation in the Hughes 369E. He reported, however, that he had completed autorotation and other emergency training in an Augusta 119 Koala helicopter about one week prior to the occurrence, and in a Bell 205 helicopter about one month prior to the occurrence.
The maintenance release was current and there were no outstanding maintenance requirements. A routine 100-hourly engine inspection was carried out on 25 May 2002. Post-accident technical examination of the engine and wreckage indicated that the helicopter was capable of normal operation prior to the occurrence.
The gross weight of the helicopter at the time of impact was estimated to be within the authorised maximum operating and Height Velocity Diagram weight limits. The longitudinal and lateral centres of gravity were estimated to be within published flight manual limits. Helicopter performance was estimated to be sufficient for both in and out-of-ground effect flight.
The company Operations Manual stated a flight planning fuel consumption rate of 100 L (176 lbs) per hour for the Hughes 369 type. Charter helicopter fuel planning was required to include the provision of 20 minutes fixed and 15 per cent variable reserve. However, a reduction to a 10-minute fixed reserve was authorised for helicopter external load operations. That amounted to 42.4 lbs (29.3 lbs useable) indicated on the fuel quantity indicator at the company planning fuel consumption rate.
The pilot reported that the company Chief Pilot suggested a planning fuel consumption rate of 200 lbs per hour and that 100 lbs (86.9 lbs useable) indicated on the helicopter fuel quantity indicator equated to about 15 to 20 minutes flying time. He stated that, throughout the day's operations, he maintained a fuel log indicating an average fuel consumption of approximately 200 lbs per hour. The ground search and rescue party reported that, on arrival at the accident site, they collected paper and other loose items in the immediate vicinity of the wreckage. Those papers and items were secured in a large bag left at the accident site. The pilot's log was not recovered from that bag of items.
The pilot reported that a total of 280 L of fuel was added to the helicopter during the day using the operator's drum fuel stock and hand rotary fuel pump located at Lake Mackenzie. That amount of fuel was based on the pilot's understanding that approximately 280 turns of the rotary pump were made during the day's refuels and that pump output was 1 L per turn. He reported that he visually checked the fuel quality after each refuel. Post-accident examination of the remaining company drum fuel stock confirmed that it was JetA1 and did not reveal any contamination. Post-accident testing of the hand rotary pump used to refuel the helicopter determined an actual pump output of 0.7 L per turn.
The helicopter flight manual stated that the fuel low level advisory light illuminated when approximately 35 lbs of fuel (21.9 lbs useable) remained in the fuel tank. The manual further stated that illumination of the fuel low level advisory light required the pilot to 'land as soon as possible', which was defined as:
Execute a power-on approach and landing to the nearest safe landing area that does not further jeopardise the aircraft or occupants.
A warning was included in the flight manual that, with the fuel low level advisory light illuminated:
Sideslips may cause fuel starvation and result in unexpected power loss or engine failure.
The flight manual also contained a Height Velocity Diagram that represented combinations of altitude and airspeed from which "a successful autorotation landing would be difficult to perform". Those figures were calculated at mean sea level, over a smooth hard surface and on a standard day (15 degrees C temperature, 1013.2 mb atmospheric pressure). The manual mandated adjustment to the helicopter gross weight limits, as a function of density altitude, in order for the Height Velocity Diagram to remain applicable. The pilot reported that he entered autorotation from a descending left turn at approximately 200 ft AGL. While the speed of the helicopter as the pilot rolled out of the turn could not be accurately determined, the pilot reported that he established 65 kts in the autorotation descent.
The Civil Aviation Safety Authority (CASA) had conducted regular surveillance audits of the company since issuing the company with an Air Operator's Certificate. The last on-site audit was conducted on 6 July 2001 and a remote audit was conducted on 30 January 2002. Those audits did not indicate any safety deficiencies.
The Civil Aviation Safety Authority approved company Operations Manual directed that "...all operating personnel associated directly with..." the company were to observe the "...instructions, procedures and information contained in..." the manual. The Manual also directed that "...all company personnel associated with piloting and flight line management..." must sign the signature sheet in the master copy of the Operations Manual "...as evidence of having read, understood and agreed to apply the procedures and data contained in it". The occurrence pilot was employed by the operator on a "standard day" contract, and was therefore required to comply with the provisions of the Operations Manual, but was not required to sign the master copy of the Manual.
The coordinator for the Western Tiers operation reported that there was no formal contract in place with the operator for the day's operations and no formal audit of prospective helicopter support organisations by the charter client. It was reported that the operator was contracted for the day based on extensive previous experience operating with the charter client and the statewide experience of its pilots. The occurrence pilot had not previously flown in the Western Tiers area of Tasmania.
The pilot reported that a flight operations brief was conducted with personnel present at Lake Mackenzie prior to commencement of the day's operations. That brief included operating around the Hughes 369E helicopter and the operation of the aircraft doors and safety belts. The pilot also reported that, prior to takeoff for the occurrence flight, he had asked the passengers to confirm their seat belts were secure. Passengers reported that they were not wearing headsets during that flight.
Flight notification details for the flight were not submitted to Airservices Australia, nor was there any requirement to do so. There was no formal flight-following process undertaken by the operation. The pilot reported that radio communications with Air Traffic Services (ATS) had not been possible. The pilot reported making a Mayday broadcast on the forestry service channel following the reported engine power loss. That broadcast was not reported as having been received by any station.
Prior to being noted overdue, the pilot had departed Lake Mackenzie for Lake Nameless with an external load and was to return to Lake Mackenzie. At about 1600 hours, the helicopter was reported overdue to the operator by the coordinator of the Western Tiers operation. The operator then alerted Melbourne ATS of the overdue helicopter. At 1652 hours, ATS alerted Australian Search and Rescue (AusSAR). The pilot reported manually activating the Emergency Locator Transmitter (ELT) shortly after 1700 hours. AusSAR directed an aircraft to the area to conduct a beacon search at 1715 hours. That aircraft flight crew made the initial detection of the ELT signal on 121.5 MHz at 1720 hours. The ELT signal was first detected by the COSPAS/SARSAT satellite constellation at 1756 hours.
At about 1900 hours, a rescue helicopter from Hobart located the wreckage and survivors. A number of attempts were made to land at the accident site. Low cloud and fog prevented the landing and the rescue helicopter departed for Launceston airport to refuel. The survivors reported that departure of the helicopter resulted in a marked decrease in their morale.
The ground search and rescue party arrived at the accident site at 2338 hours. The four survivors required treatment for varying degrees of hypothermia and spinal and other injuries. They were transported from the site by rescue helicopter and arrived at Launceston General Hospital by 0516 hours on 29 May 2002.
The impact damage to the right fuel cell bladder and extended periods of ground running during the day's operations prevented the investigation from determining an accurate fuel consumption. The investigation determined that a landing on sloping ground should have affected both the fuel quantity indicator and fuel low level advisory light equally. An indicated fuel quantity of 100 lbs (86.9 lbs useable) and coincident illumination of the fuel low level advisory light, both reported by the pilot, could not be explained.
Technical examination of the helicopter's fuel indicating system established that illumination of the fuel low level advisory light coincided with 35 lbs (21.9 lbs useable) indicated on the fuel quantity indicator. At the company flight planning fuel consumption rate of 176 lbs per hour, 21.9 lbs of useable fuel would likely have equated to a flight time of approximately 7 minutes. In that case, a reported departure from Lake Nameless Hut at 1515 hours would have likely resulted in engine fuel starvation at about 1522 hours.
Technical examination of the helicopter and engine revealed no anomalies. Therefore, the helicopter was considered capable of normal flight prior to the occurrence. The amount of fuel onboard the helicopter, less than that expected by the pilot, likely resulted in unporting of the fuel cell fuel supply hose, and engine fuel starvation during the turn to land at Tom Whitely's Hut. Due to impact rupture damage of the right main fuel cell, an accurate fuel quantity remaining could not be measured.
The pilot reported that the autorotation landing was normal and that contact with a wire rotated the aircraft through 180 degrees. As there was no evidence of helicopter contact with the fence prior to, or during the initial ground impact, the investigation concluded that the fence did not contribute to the accident sequence.
Ground impact marks indicated a relatively steep approach with low forward ground speed. Examination of the Height Velocity Diagram indicated that, at the pilot reported height of 200 ft above ground level, and airspeed in autorotation of 65 kts, a successful autorotation landing should have been possible. Impact damage indicated that the autorotation landing was unsuccessful. It was therefore likely that the pilot's estimate of height and airspeed at the time the rotor speed decreased was less than actual. In that case, the helicopter may have been at a height and airspeed from which a successful autorotation landing would be difficult to perform.
The pilot reported that he had minimal recent experience on the Hughes 369E helicopter type and had practised autorotation landings in an Augusta 119 Koala helicopter type during the previous week. While it cannot be discounted, the investigation could find no evidence to indicate that lack of type-specific recency, or contradicting cross-type pilot handling, contributed to the unsuccessful autorotation landing.
While the pilot reported asking the passengers to confirm the security of their seat belts prior to take off for the occurrence flight, the passengers reported that headsets were not worn during that flight. The ambient cockpit and other noise as the passengers boarded the engine-running helicopter may have prevented them from hearing any direction from the pilot. The front seat passengers were ejected forward of the helicopter during the impact sequence. The front seat passengers' seat belts, shoulder harnesses and attachment points exhibited no evidence of damage, or having been forced by impact forces. Therefore, it was unlikely that the front seat passengers were wearing seat belts at the time of impact.
In effect, the selection process employed to contract the operator for the day's operation included an informal risk assessment. Risk assessments represent a valuable safety tool. They can range from an informal experiential and environmental audit, similar to that conducted by the coordinator of the Western Tiers operation, to an in-depth analysis of all hazards likely to affect the operation of an aviation system. That analysis includes consideration of the likelihood of an identified hazard to an operation, and the possible consequence to the aviation system resulting from that hazard occurring.
A more formal and inclusive risk assessment, conducted by all participants in the Western Tiers operation, could have enhanced the overall safety of that operation. Some of the risks to the operation, and possible risk treatments that might have been considered by the interacting participants in the operation were:
Pilot experience. The pilot's reported unfamiliarity with the area of operations and lack of recency in the Hughes 369E helicopter type could have been mitigated by a more extensive orientation and check flight and briefing procedure. That process could also have included appraising the coordinator of the operation of the pilot's background and lack of local experience.
Fuel reserve. The 10-minute fixed reserve authorised for external load operations in the company Operations Manual likely maximised flexibility and payload during such operations. However, the operation in the Western Tiers involved the movement of external loads and carriage of passengers in an at times inhospitable area, by a pilot unfamiliar with that area. In that case, modifiers to the company 10-minute fixed reserve may have been pertinent, and the company charter minimum fuel requirements been more relevant to the operation.
Flight following. The operation was conducted in an at times inhospitable and remote area of north-western Tasmania. There was scope for a more formal flight following procedure to decrease rescue agency response time and optimise the safety of the operation overall. Available flight following options included formal employment of a monitored flight and details schedule by the participants in the Western Tiers operation, regular radio contact between the pilot and Air Traffic Services, or the nomination of a SARTIME by the pilot.
The departure of the rescue helicopter from the accident site, without landing, was reported by the survivors to have adversely affected their morale, and confidence in their subsequent rescue. They were not aware that the ground rescue party was enroute to their location. A means of communication from the rescue helicopter to personnel on the ground may have prevented that decline in survivor morale and confidence.
The helicopter engine lost power at a critical stage of flight.
The pilot was unable to conduct a successful autorotation landing.
Australian Search and Rescue is considering the promotion of a means of communication between rescue helicopters and personnel on the ground.
The ATSB will monitor and publish any subsequent action on the ATSB website.
|Date:||28 May 2002||Investigation status:||Completed|
|Time:||1524 hours EST|
|State:||Tasmania||Occurrence type:||Fuel starvation|
|Release date:||26 February 2003||Occurrence class:||Operational|
|Report status:||Final||Occurrence category:||Accident|
|Highest injury level:||Serious|
|Aircraft manufacturer||Hughes Helicopters|
|Type of operation||Charter|
|Damage to aircraft||Destroyed|
|Departure point||Lake Nameless, TAS|
|Departure time||1515 hours EST|
|Destination||Tom Whiteley's Hut, TAS|
|Role||Class of licence||Hours on type||Hours total|