The helicopter was engaged in the ship-to-shore transfer of a marine pilot at night, and was operating in accordance with the night visual flight rules (NVFR). The ship was reported to be steaming at 14.5 kts, steering 044 degrees M. Weather conditions at the time were reported to be fine, with visibility of 4-5 NM. The night was also reported to have been very dark, with some haze. The sea was almost calm, with a swell ranging between 0.25 m to 0.5 m. The wind was about 5 kts from the ENE and the temperature was about 27 degrees Celsius. The moon was waning, and its bearing relative to the accident site was 027 degrees M, at an elevation of 60 degrees above the horizon.
The deck landing area measured approximately 25 m fore and aft, and 20 m across the ship, which was 23 m wide. The landing surface consisted of steel cargo hatch covers. Sea containers were stacked to a height of 5.2 m, immediately forward of the landing area.
A potential obstacle to the operation was a crane immediately aft of the landing area. In its stowed position, the crane jib would normally be aligned along the fore/aft axis of the ship, above and parallel to the landing area surface. The ship's crew reported that the marine pilot requested the crane's jib be turned 90 degrees to the left side of the ship, and elevated to its upper limit. The pulley block of the crane's hook assembly weighed an estimated 1.1 tonnes and was painted with yellow and black stripes. A lifting hook hung below the block. The crane assembly and the deck landing area were floodlit. A light on the jib illuminated the pulley block. The height of the pulley block at the time of the occurrence could not be positively established. If the crane was hoisted as reported by the ship's crew, its height above the ship's deck was about 20 m. Had the crane been swung as reported by the pilot, the height of the pulley may have been lower than 20 m. The ship's crew reported that the crane operator had vacated the crane tower and was on the deck for the arrival and departure of the helicopter. The helicopter approached the ship from the right side and landed on the forward right hatch cover, facing towards the left side of the ship. The marine pilot boarded the helicopter through the right front door and occupied the right seat. During the subsequent takeoff, there was a collision between the helicopter and the pulley block, and the helicopter fell into the sea, where it floated inverted, supported by the buoyancy of its utility floats. Small pieces of rotor blade debris were found on the ship's deck. A fisherman heard the ship's master report the accident on the marine radio frequency and, after searching for about 25 minutes, located the wrecked helicopter. The passenger was fatally injured, and the pilot sustained minor injuries.
This was the first time the ship's master had accepted a helicopter marine pilot transfer from this ship. He was familiar with helicopter operations onto larger ships. He was hesitant to agree to a helicopter transfer until the ship's agent and the marine pilot convinced him that the size of the proposed landing site was adequate. This was also the first occasion that the helicopter pilot had conducted a marine pilot transfer with this ship. However, he had previously conducted two marine pilot transfers at night onto the sister ship, the most recent being on 11 February 1997.
The helicopter pilot held a commercial pilot licence, a valid class 1 medical certificate and a NVFR rating. He was endorsed to fly Hughes 500 helicopters (also known as Hughes 369HS) and was current at night flying. He held a helicopter float endorsement and had successfully undergone helicopter underwater escape training on 30 April 1995.
At the time of the accident the pilot had a total flying experience of 8,462 hours, of which 7,882 were in helicopters, including 1,408 hours in Hughes 500 helicopters. He had flown a total of 545 hours at night, and his total instrument flight time was 10 hours. A biennial flight review had been conducted on 8 March 1996, and his most recent company flight check was conducted on 14 January 1997. In the four years prior to the accident, the pilot had flown in excess of 800 marine pilot transfers at night. Most of those had been in Hughes 500 helicopters. The pilot described the helicopter landing sites on the sister ships as more than adequate for a Hughes 500 helicopter, which had a main rotor diameter of 8 m.
The pilot had been rostered for duty in accordance with an exemption against Civil Aviation Order (CAO) 48 - Duty Times which applied to company pilots engaged solely in marine pilot transfer operations. At the time of the accident, the helicopter pilot had been on call solely for marine pilot transfers for the previous two days, following two days off. After awaking at 0730 on 25 February 1997, he flew the first marine pilot transfer for the day between 1830 and 2030; the flight time was about 0.4 hours. He then slept, before departing Gladstone at about 0058 on the accident flight. The pilot reported that he had not engaged in any strenuous activities during the rostered duty period and had flown only 0.7 hours in the 24 hours before the accident.
In recounting the accident, the pilot expressed the view that conditions on the night of the accident were such that no part of the flight would have been considered difficult for an experienced marine transfer helicopter pilot in a Hughes 500 helicopter. He reported that the flight to the ship was normal. After the passenger had boarded the helicopter, the pilot checked for obstructions in the intended direction of taxi and takeoff and noted the crane jib. He planned to depart to the north-east after clearing the left side of the ship and noted that two ships positioned to the north, as well as the lights of Gladstone to the west, would give him a good visual reference.
The pilot's position in the left front seat of the helicopter provided an excellent view forward across the deck, to the left and above, and above to the right. His view immediately to the right would have been slightly restricted by the passenger.
The pilot reported that he took-off and manoeuvred the helicopter into a low hover, then taxied across the hatch towards the left side of the ship. He yawed the helicopter slightly right in anticipation of weathercocking as the helicopter cleared the ship to the left side. The helicopter weathercocked as expected. The pilot said that he then stabilised the aircraft alongside the ship briefly, maintaining a constant altitude and keeping pace with the ship. This was in accordance with company procedures for departing from ships with obstructions. With the ship to his right and the lights of two other ships in the forward left quarter of his field of view, he reported that he established a zero-bank/zero-yaw attitude in preparation for transferring to flight by sole reference to the cockpit instruments. To be sure the helicopter would move away from the ship, he yawed 10 to 15 degrees left, then simultaneously increased power and moved the cyclic control forward to accelerate and climb away. Within a few seconds of initiating this sequence, he felt a jolt and the helicopter pitched nose-up and rolled to the right. Despite flight control inputs, he was unable to counteract the roll to the right. The helicopter then struck the water. The pilot believes that the ship's crane was swung into the helicopter's rotor arc as he took off from the ship.
The Hughes 369S is equipped with an articulated main rotor system, which permits the rotor blades to feather (change pitch angle), flap (move up and down vertically) and to lead and lag in the plane of rotation. The blades also have washout to equalise lift across the blade. When the blades are rotating, aerodynamic and centrifugal forces act on the rotor disc. These forces are finely balanced to keep the rotor disc stable and acting in the desired manner. If a critical component, such as a rotor blade, is damaged, the rotor disc is likely to become immediately unstable and its action unpredictable. During the investigation the pilot supplied a report, which proposed a mathematical model to verify his evidence. The report's author was not qualified as a helicopter aerodynamicist or as an accident investigator.
The helicopter was loaded within its approved centre-of-gravity and gross weight limits at the time of the accident.
The approved flight manual for the Hughes 369 states that controllability during hovering downwind, and both sideward and rearward flight, has been demonstrated to be adequate in winds up to 20 kts. The Hughes 369 also has a reputation for being fully controllable in much stronger crosswind and tailwind conditions.
Wreckage and survivability information
Examination of the wreckage did not reveal any fault that might have contributed to the accident. All flight control system damage was typical of main rotor and drive train sudden stoppage. The main rotor head assembly had suffered extensive damage, indicative of main rotor blade contact with a solid object while the rotors were being driven. The four main rotor blades and their grips were torn off the rotor head at the strap pack as a result of the blades impacting the pulley block, and the helicopter's subsequent contact with the sea. Both tail rotor blades, the tail rotor gearbox, and part of the tail rotor drive shaft had separated from the aircraft when the aft portion of the tail boom fractured during the impact sequence. Those items were lost at sea.
The engine-to-transmission drive shaft suffered an overload fracture typically caused by sudden stoppage forces. Smearing of the metal fracture surfaces on this drive shaft indicated that it had continued to rotate after the fracture occurred. An in-depth examination of the fuel system was not considered necessary, due to the physical evidence that the engine was performing at a high power setting when the main rotor strikes occurred.
The main damage to the fuselage occurred on the right side, where the fuselage skin exhibited extensive lateral/inward crushing deformation as a result of impact from one or more main rotor blades, as well as from water impact. Both front seat pans were crushed downward, consistent with the high g-loading experienced by both occupants when the helicopter impacted the sea. Both forward cabin doors separated from the aircraft. Most of the fibreglass engine intake fairing was missing after the accident.
Both flight attitude indicators fitted to the helicopter had recently been overhauled. Notwithstanding the extent of impact and salt-water damage, no fault was found with the instruments.
The carrying capacity of the crane was 25 tonnes, with a maximum outreach of 28 m. Marine surveyors subsequently advised that the design of the cables and the pulley block counteracted any tendency for the block to turn and twist the cables. Consequently, the block face that was struck by the rotor blades, was probably facing out to sea at the time of the accident.
The vertical face of the pulley block struck by the helicopter was approximately 1.4 m high by 1 m wide. Contact between the main rotor blades of the helicopter and the pulley block resulted in several distinct impact marks on the face of the block. Four of the marks displayed features that were consistent with contact by the main rotor blade leading edge abrasion strips and threaded tip weights. The sequence of the blade strikes could not be established. However, the presence of the tip weight impact marks on the block face indicated that the main rotor blades had not contacted any solid object before hitting the block.
Multiple scratch marks were found on the opposite face of the block to the main rotor blade strike marks. Those marks were considered to have been a result of contact with the tail rotor, the tail boom or the stabilisers.
No evidence was found of rotor strike marks on the hook, the swivel, the chain, or the cables above the pulley block, nor were any marks found on the upper or lower edges of the block, or on the narrow vertical edges. However, the narrow vertical edge of the block nearest the trailing end of the main rotor strike marks showed evidence of white paint and fibreglass consistent with the engine intake fairing contacting the block. Wreckage evidence indicated that the main rotor blades probably dislodged the intake fairing. Other fibreglass items attached to the airframe were relatively undamaged and showed no evidence of contacting the block.
The helicopter manufacturer reported that, "Once the first main rotor blade struck the pulley block, all blades would have been affected by the tremendous forces generated. The sudden stoppage forces imparted and damage done to the main rotor system would have resulted in severe main rotor imbalance and caused the blades to go divergent in the lead/lag axis and possibly in the flapping and feathering axis as well; in other words the blades would no longer 'fly' as you would expect normal rotor blades to. The drive train and fuselage would also have been affected by these same forces. Engineering and/or mathematical modelling of the accident scenario then becomes a wild guess, as the performance and/or actions of the fuselage, main rotor system (to include main rotor blades) and drive train are no longer predictable."
The helicopter was fitted with utility floats. A life raft was stowed in the rear passenger compartment. Both the helicopter pilot and the marine pilot wore life vests. Both occupants also wore full harness seat restraints and remained strapped in their seats during the accident sequence. Examination of the wreckage indicated that a main rotor blade had penetrated the cabin area on the right side of the aircraft, fatally injuring the passenger. The pilot was able to escape unaided from the helicopter after the accident.
Police spoke to the ship's captain by telephone two hours after the accident. The captain reported that the helicopter was almost out of the confines of the ship when it started turning left and the main rotors then struck the crane hook which was hanging in the air.
The ship's crew subsequently reported that the crane operator had turned the jib to the left side of the ship and raised both the jib and the hook before vacating the crane tower and standing on the deck for the landing and takeoff of the helicopter. They said that the helicopter initially rose into a hover about 1 m above deck level, where it paused briefly before accelerating across the deck, climbing and turning left at the same time. They reported seeing the helicopter then collide with the pulley block and begin to rotate, before the tail rotor also struck the block. The helicopter then fell into the sea. Shortly thereafter, crewmembers saw the helicopter floating inverted about 15 m from the left side of the ship.
The company operations manual (page D8.10) stated:
"During each takeoff, when established in the hover over the deck, a check should be made of power available, centre of gravity and temperatures and pressures before moving clear of the landing area.
At night a climb to 500 ft is to be completed before any substantial turns are made. All turns are to be made at the standard rate. Steep turns are not to be carried out".
CAO part 95, section 95.7.3: Exemption of Certain Helicopters from Compliance with Provisions of Sub-regulation 174B (2) of the Civil Aviation Regulations provides for special requirements for helicopters engaged in charter operations at night for the purpose of marine pilot transfers to/from ships. No evidence was found that the operator or the pilot had not complied with the requirements of CAO 95.7.3.
Although there were main rotor blade strike marks on the seaward face of the pulley block, there was no strike damage on the leading or trailing edges of the block. This evidence appeared consistent with the main rotor hub being forward of a line perpendicular to the seaward block face, and the main rotor blade leading edges impacting the block face at an angle of less than 90 degrees. The ship-facing block face had no evidence of main rotor blade strike damage, however it had sustained impact damage from the tail rotor assembly. Information provided by the pilot and the ship's witnesses indicated that the main rotor disc struck the block before the helicopter began rotating. The evidence was consistent with the main rotor blades hitting the seaward block face and, as a result of the impact, the helicopter rotated clockwise and the tail rotor assembly then struck the ship-facing block face.
Once one of the main rotor blades was damaged, it was probable that the main rotor system became unstable and its subsequent motion unpredictable due to the variables involved. A complete understanding of the relative movement between the helicopter and the pulley block requires accurate data on the motion of each object with respect to a known frame of reference. Both the ship and the helicopter were moving independently of each other. Although the speed and heading of the ship were reported, as was the sea swell, none of the information was calibrated or recorded to sufficiently fine tolerances. While the extent of the rolling and/or pitching motion of the ship was probably negligible, any movement would affect the variables in any calculation. As a consequence, the movements of the pulley block, as a result of the ship's rolling or pitching, or the influence of wind, could not be precisely determined.
These considerations, associated with the expected unstable behaviour of damaged main rotor blades, precluded an accurate assessment of the relative motion between the rotor disc and the pulley block. As a result, the strike marks on the pulley block provided insufficient physical information to reconcile the significant differences between the accounts of the ship's crew and that of the pilot. Determination of the attitude and position of the helicopter at the moment of collision was therefore not possible and any attempt to do so would be, at best, speculative. The investigation considered that, although the mathematical model proposed by the pilot and his advisors was possible, there was insufficient physical evidence to preclude other scenarios.
The factors contributing to the accident could not be determined with certainty.
The helicopter operator amended the company operations manual Section D8 - Marine Pilot Transfer to more clearly document the procedures already carried out by company pilots flying marine pilot transfers. The amended text reads as follows:
"After landing, while waiting for the marine pilot or after the marine pilot has disembarked, the pilot shall recheck any obstructions, confirm the departure route...etc.
When in the hover, check the centre of gravity, and hover power prior to flying the planned departure. Ships with obstructions require a transition between the obstructions via an over-water final approach and takeoff area before initiating an altitude over airspeed takeoff profile. Night departures then require an instrument takeoff and climb to 500 ft before a turn is commenced. During the hover or transition the helicopter may be weather-cocked as necessary".
The Bureau of Air Safety Investigation suggested that the marine pilot's employer subject its organisation to an independent audit by an aviation consultant. The employer, in consultation with the helicopter operator, has since incorporated the following safety improvements:
- The marine pilot organisation has retained the services of an aviation consultant to audit all aspects of helicopter transfers of their marine pilots.
- Marine pilot transfers are now conducted by this company in a McDonnell Douglas 500E helicopter, which has been audited by the aviation consultant. (Any replacement helicopters must also be audited before use for marine pilot transfers.)
- A left and right cockpit door jettison system has been installed in the helicopter.
- Consideration was given to relocating the life raft to the front of the helicopter for better access but, according to the helicopter operator, this has proven to be impractical.
- A 360-degree rotatable 400,000 candlepower searchlight has been mounted on the underside of the fuselage.
- The second attitude indicator is now powered by its own independent battery.
- Underwater emergency exit lighting has been installed.
- A 406 M Hz emergency locator beacon has been fitted to the helicopter and the helicopter pilot must carry a SABRE Type 6 voice capable survival beacon.
- Minimum experience requirements for helicopter pilots have been increased for marine pilot transfer operations.
- All company helicopter pilots engaged in marine pilot transfers must be endorsed on fixed/utility floats, as well as being subjected to an annual proficiency check involving autorotative touchdowns onto water by day.
- Periodic pilot check-and-training flights will include very high frequency (VHF) omni directional radio range and non directional beacon approaches under simulated instrument meteorological conditions.
- Marine pilot helicopter underwater escape training has been enhanced.
- All persons on board the helicopter now wear Civil Aviation Safety Authority approved dual-chamber life jackets, each with a survival beacon attached.
- Radio communications have been enhanced by improved helicopter to ship radio procedures: VHF or frequency modulated (FM) side-tone has been added to all communication stations in the aircraft.
- Survival equipment within the four-man life raft has been improved and a register of survival equipment is now kept up to date.
- The minimum length/width of midship helicopter landing sites and the final approach and takeoff areas has been increased to 20 m.
- No marine pilot transfers are permitted unless rescue equipment on board the vessel (including fire-fighting equipment, rescue boats etc.) is in position and ready for immediate use during helicopter transfers.
- Subject to re-assessment, the operator, in conjunction with the marine pilots, has decided that marine pilot transfers will only be performed when the helicopter's approach and departure can be made from ships with cranes stowed in their normal sea-going position or, if swung, are within the lateral confines of the vessel and the pulley block is fastened to the vessel.
|Date:||26 February 1997||Investigation status:||Completed|
|Time:||0113 hours EST|
|Location:||32 km E Gladstone, Aero.|
|State:||Australian Capital Territory||Occurrence type:||Ditching|
|Release date:||16 December 1999||Occurrence category:||Accident|
|Report status:||Final||Highest injury level:||Fatal|
|Aircraft manufacturer||Hughes Helicopters|
|Type of operation||Charter|
|Damage to aircraft||Destroyed|
|Departure point||32 km E Gladstone, QLD|
|Departure time||0113 hours EST|
|Role||Class of licence||Hours on type||Hours total|