FACTUAL INFORMATION

The pilot in command of a scheduled passenger service from Sydney to Cootamundra, via Cowra, discontinued the flight with the assigned aircraft due to a mechanical problem experienced during engine start at Cowra. The chief pilot had also been on board the aircraft, observing the flight procedures of the pilot in command. The scheduled service was subsequently completed by the pilot in command using a substitute aircraft, VH-XLA, which had been dispatched from Cootamundra.

After landing at Cootamundra the aircraft was taxied to the fuel bowser and shut down. In anticipation that he would be refuelling the aircraft, the pilot in command left a written note for the next rostered pilot which stated that the main fuel tanks were full. At about this time the chief pilot chose to carry out a debrief of the flight with the pilot in command. The two men sat down in the rear of the aircraft to discuss the flight. Whilst the debrief was in progress, a company engineer boarded the aircraft and taxied it to the hangar for routine maintenance. The pilot in command reported that he was distracted by the debriefing and did not subsequently return the aircraft to the fuel bowser to complete the refuelling process. In addition, the maintenance release (MR) did not show that the aircraft had been flown from Cootamundra to Cowra and return, nor were details of the flight, and fuel remaining on board at the completion of the flight, documented in other company records. The operator had planned to use the same aircraft for the scheduled afternoon service to Sydney via Young and Cowra.

The pilot rostered for the afternoon service arrived at 1300 EST to carry out a pre-flight inspection. Company fuel policy for that service required a minimum of full main tanks. Fuel records indicated that the main fuel tanks of VH-XLA had been refilled on the previous evening, information which was supported by the note left by the previous pilot. One of the company engineers, who was queried by the pilot, was not aware that the aircraft had been flown that morning but added that the engines had been run at considerable power for about 10 minutes. The pilot later reported that his observation of the fuel gauges confirmed the tanks were close to full. However, he did not carry out a visual inspection of the fuel levels in the tanks prior to departure. The aircraft subsequently departed Cootamundra at approximately 1525.

During climb after departing Cowra, the pilot observed the fuel gauges indicating less than he would normally have expected at that stage of the flight. He elected to cruise at 7,000 ft, and reduced power in order to decrease fuel flows. The pilot also contacted the company head office about possible fuel gauge anomalies. As the aircraft approached Bindook, the pilot carried out a further check of the fuel gauges and chose to land at Camden as a precautionary measure. A PAN call was transmitted to Flight Service, and the passengers were advised that the aircraft would be landing at Camden due to a low fuel indication. The pilot executed a normal landing following a straight-in approach.

Two hundred litres of fuel were added to the main fuel tanks in order to complete the flight, but the pilot was unable to start the left engine. Ground transport was subsequently used to take the passengers to Sydney. A further 182 L were added to fill the main fuel tanks, bringing the total amount of fuel added at Camden to 382 L. The combined usable fuel capacity of the main tanks is 402 L.

An engineering examination of the aircraft did not reveal any anomalies with the fuel system or fuel quantity gauges which may have contributed to the incident. It was determined that the pilot had used an incorrect start procedure at Camden, causing problems in starting the left engine. Fuel calculations, based on the company fuel policy, indicated that the aircraft had departed Cootamundra with approximately 250 L on board, equivalent to about two-thirds of the combined fuel capacity of the main tanks. The actual fuel required for the flight, including 113 L of standard reserves, was calculated to be 358 L.

The pilot held a valid licence and ratings for the operation being undertaken. He had commenced full-time employment with the company on 24 January 1996 after working part-time with the same company since early 1995. He had accumulated 1,275 hours in Piper Chieftain (PA31) aircraft, all of which had been flown in company aircraft.

The pilot believed that fuel would not be visible in the main tanks, as viewed through the filler caps, if any more than a few minutes of fuel had been burned. This did not appear to be a view held by other company personnel. Moreover, the pilot indicated that had he visually inspected the tanks prior to departure from Cootamundra, and not been able to see fuel, he would still have concluded he had sufficient fuel based on all other available information and the knowledge that the engines had been run for approximately 10 minutes during maintenance checks. A number of operators of Chieftain aircraft were consulted, all of whom said that a total of some 300 L remained in the main fuel tanks when fuel was just visible on the sloping bottom surface of each tank.

The company aircraft checklist required that, as a minimum, the fuel gauges/contents are to be checked prior to start-up and during the pre-take-off procedure. However, the pilot confirmed that following his first observation of the gauges during the pre-flight inspection, he did not check the gauges again until he had departed Cowra. He reported that he had experienced a high workload on this flight, due to a full complement of passengers and a significant amount of baggage to deal with and had overlooked this check. Since he also assumed the main fuel tanks had been full when the aircraft departed Cootamundra, he did not expect the fuel gauges/contents to be a checklist item requiring close and frequent attention.

ANALYSIS

The responsibility of ensuring that there was sufficient fuel to complete the flight rested with the pilot of the afternoon service. However, the incident resulted from a chain of errors and omissions involving a number of company personnel, which ultimately led to insufficient fuel being on board the aircraft to complete the scheduled flight from Cootamundra to Sydney.

In the first instance, the MR had not been certified by the pilot who ferried the substitute aircraft to Cowra. In the second instance, the pilot in command who had completed the scheduled flight in the substitute aircraft, did not complete the flight details nor certify the MR. In addition, the pilot in command was normally in the habit of completing the note, which indicated that the main tanks were full, before leaving his control seat immediately prior to refuelling. However, his normal routine had been interrupted when the chief pilot chose to debrief him before he had time to complete all required administrative and refuelling tasks. His pre-occupation with the chief pilot's critique of his flight subsequently led to these tasks being forgotten.

As a result of the oversights which occurred during the morning, the supporting documentation available to the pilot rostered for the afternoon service did not indicate that a flight had taken place in the aircraft earlier that day. The note left by the pilot of the earlier flight indicated that the main fuel tanks were full. In addition, the pilot's understanding of the fuel system was incorrect. This is considered to be the most significant factor in the chain of events. It was ultimately his responsibility to determine that there was sufficient fuel for the flight. He did not carry out a visual check of the main fuel tanks. However, even if fuel had not been visible, his assessment of the fuel quantity would not have altered. Because of his inadequate knowledge of the fuel system, such an observation would not have alerted him to the discrepancy between the fuel documentation and the actual fuel contents.

The pilot's understanding of the aircraft's fuel system differed with that of his peers within the company, and with other operators of the aircraft type. In addition, this understanding was inconsistent with the general view held by other company personnel that he displayed competence and professionalism in his flying duties. The investigation was unable to resolve this inconsistency.

It was determined that the fuel gauges should have indicated a maximum of three-quarters full at the time of the pre-flight inspection. It is considered probable that the pilot's observation of the fuel gauges reading close to full prior to departing Cootamundra was strongly influenced by the other fuel-related information available to him prior to departure. This expectation was then carried throughout the rest of the flight until he finally became aware of the apparently low-reading fuel gauges after leaving Cowra. Even at that stage he thought the readings were incorrect but wisely decided to land at Camden as a precaution.

SIGNIFICANT FACTORS

1. The maintenance release had not been certified, and details of the morning flight in the substitute aircraft, including flight times and fuel remaining, had not been documented.

2. The pilot in command of the morning flight had written a note stating that the main fuel tanks were full but was distracted from his duties and did not complete the refuelling process.

3. Other documentation indicated that the main fuel tanks were full.

4. The pilot rostered for the afternoon flight did not visually inspect the fuel tanks, nor would he have changed his assessment of the fuel quantity if he had, given the information he had available to him and his incorrect understanding of the fuel system of the aircraft.

5. Due to a high workload, the pilot did not check the fuel quantity gauges in accordance with company procedures prior to departure from both Young and Cowra.

SAFETY ACTION

As a result of this occurrence, and in consultation with the Civil Aviation Safety Authority, the company has reviewed its policies and procedures in relation to the checking and documentation of fuel quantities in PA31 aircraft. The company Maintenance Control Manual was amended to ensure that a visual inspection of the fuel tanks was included in the pre-flight inspection.

FACTUAL INFORMATION

The crew of a Fokker Fellowship (F28) aircraft had flight planned to conduct navigation aid calibration of the Sydney runway 16L instrument landing system (ILS). Runways 16L and 16R were nominated for arrivals and departures. The crew received a clearance for a radar departure from runway 16R. The departures controller's intention was to radar vector the F28 to the east and north of the aerodrome for an intercept of the 16L ILS from the north. The F28 departed from runway 16R, heading 170 degrees, on climb to 3,000 ft. When the F28 was 8 NM south of the aerodrome the departures controller instructed the crew to turn left onto a heading of 120 degrees, to parallel the RWY 16L departure track. The controller had six aircraft on frequency, with a number of pending departures for both runways. He planned to vector the F28 across the RWY 16L departure track between two B737s departing from runway 16L. The first B737 took off with an instruction to turn left onto a heading of 060 degrees. The F28 was also instructed to turn left heading 060 degrees. The controller assessed that the F28 could cross the RWY 16L departure track ahead of the second B737, which had also taken off, and instructed the F28 to turn onto a heading of 020 degrees.

The crew of the second B737 had earlier been issued with a West Maitland 7 SID, on climb to 5,000 ft. The SID required the aircraft to track 155 degrees until reaching 600 ft, and then to turn left to intercept the 126 degrees radial of the Sydney VHF omni-directional radio range (VOR) navigation aid. The B737 crew were subsequently re-cleared to climb to their planned level of FL210 immediately after departure. The controller observed the reducing distance between the F28 and the second B737, but believed the B737 would climb above the F28, and that the vertical separation standard of 1,000 ft would be achieved prior to the distance reducing to less than the required lateral standard of 3 NM. However, when the B737 reached 3,000 ft the crew levelled the aircraft to accelerate to 250 kts. As the nose of the B737 was lowered the pilot in command observed the F28 at the 11 o'clock position, at a distance of 1 to 2 NM and slightly below their level. Just prior to the distance between the aircraft reducing to less than 3 NM, when the vertical separation was 300 ft, the controller instructed the crew of the F28 to turn right onto a heading of 040 degrees. He then instructed the crew of the B737 to turn right onto a heading of 180 degrees. The aircraft passed with less than the minimum vertical and lateral separation standards. Aircraft departing to the north and east were normally instructed to to depart from runway 16L, if that was operationally acceptable. The F28 had been parked overnight at Sydney aerodrome on the tarmac adjacent to the threshold of runway 16R. There was no operational reason for the aircraft not to have used RWY 16L for departure.

ANALYSIS

The controller relied on his understanding of B737 performance to establish vertical separation between that aircraft and the F28. The lack of application of separation assurance techniques provided no alternative means to maintain separation when the B737 was unable to attain the expected altitude. The controller had a number of aircraft under his control and this aspect should have been a prompt for him to ensure separation was assured between aircraft. Had a clearance been issued for the F28 to depart from runway 16L it is unlikely that the incident would have occurred.

SIGNIFICANT FACTORS

1. The controller's expectation that the B737 would climb 1,000 ft above the level of the F28 before lateral separation reduced to less than 3 NM.

2. The controller's lack of application of separation assurance techniques.

3. The departure of the F28 from runway 16R when runway 16L was available meant that the F28 would have to cross the departure track of runway 16L in order to carry out its intended operation.

The pilot reported that during cruise at 2,300 ft the engine gradually lost power until he was committed to a forced landing. During the forced landing the aircraft flew through the tops of 40 ft trees and then under a powerline. The aircraft then hit and killed two sheep, ran through a depression, became airborne again and finally hit the ground heavily before coming to a stop. The engine is a sealed rotary liquid cooled unit which can only be dismantled at the manufacturer's facility in England. A post-accident inspection revealed a coolant hose had detached from the water pump but it could not be determined if this initiated the engine failure or was a result of the accident. Traces of internal seal material were found in the exhaust system so it is possible that a seal failure initiated the engine failure. The owner does not intend to return the engine to the manufacturer for dismantling.

The pilot had planned the flight for an early morning departure from a private airstrip. He had calculated the take-off weight, with four passengers, to be below the maximum allowable for the take-off length available. The weather was fine and clear, with a light crosswind on the gently down sloping grass strip selected for take-off. The grass had been mown short but was wet from a heavy dew.

During the take-off run the pilot became concerned that the aircraft was not accelerating normally, so he closed the throttles and applied maximum braking. However, he was unable to stop the aircraft before it overran the strip end and collided with a fence, rupturing a fuel tank and causing minor damage to the engine cowlings and landing gear doors.

No defects were found with the aircraft that may have affected its performance. It is likely that the wet grass would have increased the take-off distance required and also decreased the aircraft's braking efficiency when the take-off was rejected.

The aircraft had been operating on the mustering flight for about 45 minutes. With the helicopter at 70 feet above the ground and 50 kts indicated airspeed, the pilot reported that he had just commenced a climbing turn to approach a small mob of cattle when he heard a loud noise from the rear of the cabin. The engine oversped and the rotor speed decayed. A forced landing was conducted but the pilot was unable to retain adequate rotor speed to prevent a heavy landing.

The floatplane took-off from a coastal lake in strong gusting winds associated with passing squalls. As the aircraft approached the coast, and flew over the beach at low altitude, it descended until the left float struck the top of a wave. The aircraft nosed over into the ocean and began to sink. The pilot and passenger managed to evacuate the semi submerged aircraft without injury.

The pilot subsequently reported that, after taking off toward the south into the strong wind, he had to turn left at low altitude to avoid a small hill adjacent to the shore. The aircraft had then probably encountered windshear on the downwind side of the hill, which resulted in the unintended loss of altitude and subsequent contact with the water.

After an apparently normal touchdown, the aircraft was observed to become airborne again and began to porpoise. On the third bounce, the nose landing gear collapsed, and the propeller struck the runway before the aircraft came to rest.

The pilot subsequently reported that the approach and touchdown were normal, but he was unable to recover from the bounce with elevator control and did not apply any engine power to recover the situation. His experience on the aircraft type was limited as he had only gained his endorsement the previous day.

FACTUAL INFORMATION

NITE 01, a United States Navy F-14 aircraft, had departed from the Atsugi Naval Air Facility, Japan on an Instrument Flight Rules (IFR) plan that included an exercise within Restricted Area R116. The restricted airspace R116 was used for military exercises, the northern end of which was approximately 6NM south east of the track of the B767. The F14 pilot cancelled his IFR status during his initial climb, choosing to proceed under Visual Flight Rules (VFR) procedures.

The crew of the B767 were approaching flight level (FL) 250 while descending into Narita, Japan, when they received a Traffic Alert and Collision Avoidance System (TCAS) resolution advisory (RA) instruction to climb. During the transition from descent to climb, they saw a military aircraft passing in the opposite direction at high speed, approximately 500 ft below them.

At approximately the same time, the air traffic controller observed radar returns from a VFR aircraft departing R-116 at FL250 and passed this information to the crew of the B767. He also contacted the pilot of the F-14 to establish that pilot's intentions. The B767 was climbing through FL255 (approximately) when the military aircraft passed.

The rules for VFR flight in Japanese airspace were such that a pilot could choose to conduct a flight using a 500 ft separation standard up to FL290 without an airways clearance, provided a listening radio watch was maintained on air traffic control frequencies. These procedures allowed the pilot of the F-14 to leave R-116 under VFR and maintain his own separation, of at least 500 ft, from the B767. He stated that he had the B767 in sight and was listening on the control frequency. Therefore, he had complied with the 500 ft separation standard for VFR flight below FL290.

The aircraft passed with approximately 500 ft vertical separation and no appreciable horizontal separation.

ANALYSIS

The TCAS advice to the B767 crew was a genuine alert as the respective flight paths were within the envelope for a RA alert. In this case, the crew manoeuvred their aircraft to a comparatively safe altitude, allowing the military pilot to pass beneath their aircraft.

SAFETY ACTION

As a result of the investigation, the Japanese Civil Aviation Bureau has requested the United States Air Force (the body responsible for overall military operations in the area) to remind all pilots under their jurisdiction of the proximity of IFR air routes and of the general civil airspace structure, and also to remind their pilots of the requirement to maintain adequate separation from civil aircraft.

Additionally, the Japanese Bureau has asked for military training to be kept, as much as possible, within the specified training areas.

FACTUAL INFORMATION

The pilot in command (PIC) of an instrument flight rules (IFR) Piper Chieftain (PA-31-350), enroute from Bankstown to Coffs Harbour, was cleared via Williamtown to Kempsey at 9,000 ft. The aircraft was overhead Williamtown at 1406 EST and estimated to be at Kempsey at 1440. Williamtown approach coordinated the position report with Sydney flight service, the next air traffic services (ATS) unit responsible for the flight. The planned track of the PA-31-350 remained in Williamtown airspace until approximately 10 NM south-west of Taree. North of this position the flight would be in Class G airspace from ground to flight level (FL) 125.

A De Havilland Canada Dash 8 (DHC-8) taxied at Taree and the crew made broadcasts and reports to flight service in accordance with procedures for operations in Class G airspace. The DHC-8 was an IFR regular public transport (RPT) flight to Sydney via Craven, a position 41 NM north-north-west of Williamtown. The crew of the DHC-8 had planned to operate at FL160.

North of Williamtown, the PA-31-350 was radar vectored to the west of track due to other traffic. When the aircraft was approximately 30 NM north of Williamtown the approach controller, cancelled the radar vector and instructed the PIC to track direct to Kempsey. The aircraft was 2 NM west of track and was within the tolerance of the primary tracking navigation aid. The approach controller was not required to advise flight service that the aircraft was west of track under the circumstances and there was no further coordination with flight service.

Prior to being instructed to transfer to the flight service frequency and while the aircraft was in Williamtown airspace the PIC of the PA-313-350 established radio communications with flight service. The flight service operator requested an estimate for abeam Taree. The PIC advised flight service to standby while he calculated the estimated time. At about the same time the crew of the DHC-8 reported the departure from Taree to flight service and was instructed by the flight service operator to contact Brisbane Centre approaching FL110.

The PIC of the PA-31-350 reported to flight service that the aircraft would be abeam Taree at 1428. Neither crew were given traffic information on the other aircraft although the DHC-8 was estimated to be at Craven at 1429. The PIC of the PA-31-350 heard the departure report from the crew of the DHC-8 but was not concerned because the flight service operator did not provide traffic information about that aircraft.

The PIC of the PA-31-350 was instructed by Williamtown approach to transfer to the flight service frequency. He reported to flight service that the aircraft was maintaining 9,000 ft and this was acknowledged by the operator. Shortly after, the PIC requested the callsign of the regional airline aircraft departing Taree from flight service and advised that it had just climbed through his level, about 400 ft in front of his aircraft. The crew of the DHC-8 did not see the PA-31-350.

The flight service operator was required to pass traffic information on IFR flights in accordance with criteria from the Manual of Air Traffic Services (MATS), to the pilots and crews of all IFR flights operating in Class G airspace. The operator used procedural methods and was also required to use the criteria from MATS to establish whether the aircraft would conflict or not. There was radar coverage in the area above approximately 7,000 ft which was used by ATS for the management of aircraft above FL125, in controlled airspace, but there was no radar display or access to radar information in the flight service centre.

The workload for the flight service operator was light and he was nearing the completion of an eight-hour shift. The operator assessed that the PA-31-350 would pass about 5 NM to the south-west of Taree and this aspect coupled with the similar estimates for the next positions (Taree and Craven) for both aircraft led him to believe that they would not conflict. The operator twice re-assessed the information on the two aircraft but did think that they would be in conflict. He did not plot the tracks of the aircraft on the chart display at the console. The Williamtown to Kempsey track and the Taree to Craven track cross 12 NM to the south-west of Taree. It was estimated that the aircraft passed at a position approximately 15 NM south-west of Taree.

The actual position and separation between the aircraft could not be ascertained from radar data recordings. The cloud in the area at the time of the incident was a layer of broken cloud between 7,000 and 9,000 ft.

ANALYSIS

The flight service operator did not plot the intended tracks of the aircraft. He relied on his mental ability to assess the potential for conflict. If he had plotted the tracks he would have appreciated that the DHC-8, was going to cross the track and would probably climb though the level of the PA-31-350, and that there was a potential for conflict. He was also nearing the end of his shift period and may have become complacent due to the low level of traffic. This may explain why he did not apply the MATS criteria to assess the potential for conflict.

The flight service operator was using procedural methods to monitor traffic in his area of responsibility. Some of the airspace within his area, near Taree, was within radar coverage. However, flight service did not have a radar display. The greater use of the available radar coverage in Class G airspace would have enabled ATS to more readily assess potential traffic conflicts.

The Taree departure report for the DHC-8 was broadcast during the period when the PIC of the PA-31-350 was calculating the abeam Taree estimate. The PIC of the PA-31-350 heard the Taree departure report but did not appreciate the significance of the information. Flight service did not provide information about the DHC-8 as possible conflicting traffic and because he was concentrating on the estimate calculation, the PIC of the PA-31-350 may have been distracted from monitoring aircraft reports for potential traffic conflicts. One of the reasons for reports in Class G airspace was to provide information to pilots to assist them to maintain situational awareness with regard to potential traffic conflicts. This was additional to the traffic information that flight service was required to provide to the pilots of IFR flights. Maintenance of situational awareness by pilots was part of the defence mechanism of the airspace system.

The layer of cloud under the PA-31-350 would have hindered the PIC in sighting the climbing DHC-8. Additionally, as the layer of cloud commenced at approximately 7,000 ft the crew of the DHC-8, during their climb, would have been restricted in their ability to maintain a lookout for aircraft.

SIGNIFICANT FACTORS

1. The flight service operator did not have access to radar information and was required to use procedural ATS procedures.

2. The flight service operator did not plot the tracks of the aircraft on the chart.

3. The flight service operator did not assess the traffic in accordance with the Manual of Air Traffic Services criteria.

4. The flight service operator did not provide traffic information in accordance with the Manual of Air Traffic Services.

5. The PIC of the PA-31-350 did not have sufficient information to maintain situational awareness with regard to potential traffic conflicts.

6. The DHC-8 was operating in an area of restricted visibility prior to and climbing through the level of the PA-31-350.

SAFETY ACTION

Bureau of Air Safety Investigation safety action

Because of this and other occurrences the Bureau reviewed aspects of the increased use of available radar coverage for improved safety and forwarded Interim Recommendation 970112 to Airservices Australia on the 14 July 1997. The Interim Recommendation stated:

"The Bureau of Air Safety Investigation recommends that Airservices Australia review the provision of air traffic services to maximise the use of the currently available radar coverage particularly on routes used by regular public transport aircraft."

Airservices Australia replied on the 17 October 1997 and stated that:

"Airservices Australia is reviewing the provision of air traffic services with regard to maximising the use of radar services both within and outside controlled airspace.

As you are aware, the Airspace 2000 proposal which Airservices planned to introduce on the 26th February 1998, comprehensively addresses the extension of radar services. These radar enhanced services include:

- Radar Class E airspace from Cairns to Melbourne above 8500 feet.

- A Radar Information Service (RIS) in Class G airspace within radar coverage.

The Board of the Civil Aviation Safety Authority (CASA) has deferred making a decision on the proposal.

Regardless of the outcome of the Airspace 2000 review by CASA, Airservices intends proceeding with three initiatives to enhance radar services on the 26th of February 1998.

1. Radar Class E airspace will be introduced between 8500 feet and FL125 outside existing Class C airspace from Grafton to Canberra within radar coverage.

2. Brisbane Enroute will provide radar services within the Class C control area steps over Coffs Harbour down to 4500 feet.

3. Sydney Terminal Control Unit will provide radar services to 45nm Sydney in non-controlled airspace on a discrete frequency.

These initiatives will increase Airservices use of existing radar coverage for air traffic services. Further expansion of radar services is limited pending decisions on Airspace 2000 by CASA."

On 26 February 1998 Airservices Australia implemented radar Class E airspace services in the area between Canberra and Grafton within radar coverage.

Local safety action

The regional airline safety department was planning to use the incident in their safety journal to highlight problems that may be encountered in Class G airspace.

FACTUAL INFORMATION

History of the flight

The helicopter was ferrying oil company personnel from the Griffen Venture oil rig to Karratha aerodrome. The weather was clear and the wind light. The co-pilot, occupying the left-pilot seat, was undergoing a line check by the pilot in command who was a company managing-captain and a qualified line-check captain. Although both pilots were experienced captains on the helicopter type, the co-pilot had the greater overall helicopter time, and the pilot in command the greater time on the helicopter type. Cockpit voice recorder (CVR) evidence indicated that there was some discussion between the pilots about a particular company matter during the flight and ground-taxi.

During the arrival at Karratha the co-pilot flew the approach, landing at a point known as "Heli 26" which is on taxiway F, parallel to runway 26. The pilot in command assumed control of the helicopter for the ground-taxi as control of the wheel brakes was possible only from the right-pilot seat. The co-pilot retracted the nosewheel-lock pin to allow the nosewheel to castor during taxi. CVR evidence indicated that the pilot in command did not invoke sterile-cockpit procedures and that the checklist challenge-and-response procedures were not conducted after the helicopter had landed. Both pilots reported however, that the pilot in command had completed the after-landing checks. The pilot in command taxied the helicopter along taxiway F, completing a 90-degree right turn into taxiway A and a left turn onto the terminal tarmac area. A company-LAME eyewitness reported that the nosewheel appeared to be castoring normally.

The pilot in command reported that he applied the park brake before disembarking the passengers and that he released the park brake before taxiing the helicopter to the company lines. The challenge-and-response procedure required for the pre-taxi checklist was not conducted before taxiing from the disembarkation point.

The helicopter had to be taxied slightly left around a parked BAe146 aircraft before executing a left 90-degree turn. This turn was succeeded by a right 90-degree turn to follow the marked taxiway, past a helicopter parking area. Having turned approximately 10 degrees left then 10 degrees right to go around the BAe146, the pilot in command attempted to turn the helicopter left to follow the marked taxiway. The helicopter's nosewheel steering did not fully respond. Although the pilot in command quickly realised that the nosewheel lock-pin had re-engaged and that he was unable to follow the marked taxiway, he decided to continue taxiing his helicopter through the occupied helicopter parking area. An eyewitness reported that the nosewheel was not castoring and that it was being dragged sideways on the tarmac as the helicopter turned slightly. Tyre-scuff marks on the tarmac surface confirmed that the nosewheel was not castoring.

The co-pilot reported that when the helicopter would not turn, he checked the handle, finding it slightly proud of the full-down position. He attempted to withdraw the lock-pin by pushing down on the nosewheel castor-lock control handle, first with one hand, then with both hands. He also reported that as his attention was diverted to disengaging the nosewheel lock-pin, he did not initially realise that the helicopter was entering the helicopter parking area. The pilot in command stated that he did not attempt to bring the helicopter to a light-on-wheels condition or low-hover to unload the nosewheel to release the lock-pin, due to the proximity of parked helicopters. He also reported that he did not stop as he thought that the Super Puma would clear a nearby parked S76. The CVR recorded a comment by the co-pilot, 13 seconds after it was noticed that the pin was still engaged, which appeared to indicate that he also thought the Super Puma would clear the parked S76. A further four seconds later, the CVR recorded clipping sounds made by the main rotor contacting the S76 tail rotor blade.

Both pilots heard the clipping sound which the pilot in command thought was a blade strike, although the co-pilot was unsure. The helicopter had travelled approximately 35 metres from where the lack of nosewheel castor was first noticed, to the impact point. Tips of two blades of the main rotor clipped a tail rotor blade of the nearby S76 helicopter. The tail rotor blade sheared at the contact point. The pilot in command then continued to taxi the helicopter across the parking area to the taxiway. Once clear of the parking area, he brought the helicopter to a light-on-wheels condition to unload the nosewheel. The co-pilot was then able to retract the nosewheel lock-pin which then allowed the helicopter to ground taxi normally.

Cockpit control configuration

The parking brake and nosewheel castor-control handles are located on the right side of the centre console, just to the left of the cockpit's right seat. The park brake-control handle is forward of, and in line with, the nosewheel castor-control handle. Both handles are similar in appearance and feel. The park brake is applied by pulling up the park brake-control handle and turning it 90 degrees. The brakes are released by turning the handle 90 degrees and pushing down.

The nosewheel is locked fore/aft by a lock-pin. The lock-pin is retracted by turning the castor control-handle 90 degrees then pushing down. The pin is re-engaged by pulling up on the castor-control handle then turning the handle 90 degrees to lock it in position. The locking of the nosewheel and application of the park brake are similar actions. The unlocking of the nosewheel and release of the park brake are similar actions. Other than the position of the handles, there are no cockpit indications to indicate the status of the park brake or nosewheel castor. Other company pilots reported they have, on occasions, unintentionally applied the park brake instead of engaging the nosewheel lock-pin during ground taxi.

The nosewheel lock-pin has a design weak point. This allows the lock-pin to shear if sufficient side-loads are applied to the nosewheel. The nosewheel lock-pin can be jammed in place if the nosewheel is not centred or undergoing even slight side-loads. Centring and off-loading the nosewheel is achieved by using tail rotor inputs and by bringing the helicopter to a light-on-wheels condition or hover. With the nosewheel lock-pin engaged, it is possible to achieve approximately 30 degrees of heading change whilst ground-taxiing. Flight Data Recorder (FDR) evidence indicates that the helicopter had turned approximately 10 degrees to avoid the BAe 146. The nosewheel lock-pin did not break during this incident despite the tyre-scrub marks on the tarmac surface. These marks indicated that the nosewheel underwent significant side-loading. A post-flight engineering inspection could not fault the nosewheel system.

Company procedures

The company operations manual prescribes which checklist actions are challenge-and-response items. These include the pre-taxi and after-landing checks. The pilots reported that they were aware of these requirements. The operations manual also directs the helicopter commander (pilot in command) to other publications including Base Instructions - Individual Unit Orders (IUOs) and Operations Information Circulars (OICs).

An OIC, detailing cockpit resource management procedures, introduced the use of a cockpit "sterile environment" within ten miles of a takeoff or landing point. This OIC had not been incorporated into the operations manual at the time of the incident, although both pilots later stated that they were aware of the requirement. Sterile-cockpit procedures require that only communications and actions essential to the safe conduct of a flight are conducted at certain periods in a flight. United States Federal Aviation Administration Regulation 135 requires the practice of such procedures during certain phases of flight, including ground taxi. The failure to practise this procedure has been implicated in several fatal accidents in the United States. Such a requirement does not exist under present Australian civil aviation regulations and orders.

IUOs are held at each company base. To reduce the amount of material being read by pilots, the company introduced a procedure whereby new orders were noted on the publication's cover sheet. If they were familiar with the prior orders, the pilots would then just read the new orders annotated on the cover sheet.

As company helicopters had been observed to diverge from the taxiway and cross the parking area, Karratha base introduced an order reminding pilots, "to stay on the taxi lines in the Woodside apron area and to be particularly vigilant". This order was dated 23/01/97, however the last entry on the cover sheet was 21/01/97 and referred to another order. There is no evidence to indicate that either pilot had read the order although the co-pilot did recall being advised by the company's chief pilot of a requirement to stay on the taxiway and the pilot in command stated that he was aware of the order. Some base pilots, however, stated that they were not aware of the order until it's existence had been noted during the investigation.

Both pilots had undergone company-provided cockpit resource management training (CRM).

ANALYSIS

Although company procedures required the use of sterile-cockpit procedures within ten miles of the landing point, the pilot in command did not invoke the requirement. Whilst the company procedures also required the conduct of challenge-and-response checklist items for after-landing and pre-taxi, no challenge-and-response items were conducted once the helicopter had landed or before taxiing from the disembarkation point. It was considered possible that the pilot-in-command may have inadvertently raised the nosewheel castor-control handle instead of applying the park brake before disembarking the passengers, although there is no evidence to confirm this and the pilot stated that he did not.

Having achieved two 90 degree turns after landing, physical and eyewitness evidence confirm that the nosewheel lock-pin had re-engaged some time after the helicopter had manoeuvred onto the main tarmac area. The lack of nosewheel lock-pin damage, despite the considerable side-loads generated, indicated that the pin had not fully engaged thereby not permitting the weak point to achieve is designed purpose. With no cockpit indication of the nosewheel pin position, the pilots were unaware of any potential problem with the steering until it was time to manoeuvre the helicopter within the constrained turning area.

Although he was aware of the IUO requiring company pilots to follow the taxi lines, the pilot in command decided to continue into the occupied helicopter parking area. The co-pilot did not raise an alert, possibly because his attention was focussed on attempts to release the nosewheel lock-pin, although he implied that he thought they were clear of the nearby, parked S76. It is unclear as to why, having thought that the Super Puma's main rotor blades had clipped the S76 helicopter, the pilot in command then decided to continue and bring the Super Puma to a light-on-the-wheels condition.

Despite the in-situ organisational defences and the pilot in command's considerable experience, he made a succession of decision-making errors. Both pilots were aware of the instructions, yet neither enforced the requirements. The co-pilot did not promote the sterile-cockpit requirement when the helicopter was within 10 NM of the landing point, he did not question the lack of challenge-and-response checks during the ground taxi, or challenge the crossing of the taxiway lines when the nosewheel refused to castor. The status of the pilot in command and the type of check being conducted may have encouraged the co-pilot, despite his CRM training, to assume a passive role in the crew's decision-making process.

The pilot in command's experience and his standing within the company indicated that the decision-making errors were uncharacteristic. The only contributing factor the investigation found was the CVR evidence that indicated that the pilots appeared to be preoccupied with other company matters. The preoccupation with issues outside the conduct of flight and the co-pilot's passive role within the cockpit probably contributed to the pilot in command's decision making errors. The design and location of the nosewheel castor-control and park brake handles associated with the lack of cockpit warnings may have also contributed to the incident.

SAFETY ACTIONS

As a result of this incident, the Bureau of Air Safety Investigation is evaluating safety issues concerning the regulation of cockpit procedures and design aspects of the Super Puma helicopter cockpit. Any forthcoming recommendations will be published in the Quarterly Safety Deficiency Report.