Circumstances:

At 1040 hours EST on 13 April 1990, Beech 95-B55 Baron registered VH-TSM crashed at Brisbane Airport. The pilot was conducting a landing approach when the aircraft struck a light pole near the southern end of a section of Runway 04 of the decommissioned Brisbane Airport and then impacted the runway surface. The pilot and three passengers received serious injuries.

1. FACTUAL INFORMATION

1.1 History of Flight

1.1.1 Beech 95-B55 VH TSM was being operated by an Albury, NSW based charter organization. The aircraft had been privately hired for a flight from Albury to Brisbane via Tamworth.

1.1.2 This was the pilot's first flight into Brisbane.

1.1.3 The aircraft departed Albury at 0610 hours and arrived at Tamworth at 0820 hours. After refuelling, the flight departed for Brisbane at about 0910 hours.

1.1.4 The flight proceeded normally, and the aircraft descended to 2500 feet on the 183 degree Brisbane VOR (Very High Frequency Omni Range) radial for an approach to Runway 01. At about 10 miles from the destination, the pilot switched on the aircraft Instrument Landing System (ILS) receiver as he was unable to locate visually the aerodrome. He noticed that the Course Deviation Indicator (CDI) for the ILS was indicating that the aircraft was right of the Runway 01 centreline and asked the Approach Controller for confirmation of this. When the Approach Controller advised that the aircraft was one mile right of the centreline, the pilot adjusted the aircraft's heading 20-30 degrees left. A short time later, after being advised that the aircraft was now on centreline, the pilot sighted what he thought was the runway directly ahead of the aircraft and proceeded to fly towards it. The aircraft was then transferred to Brisbane Tower frequency.

1.1.5 While initially satisfied that he was tracking for the correct runway, the closer in the pilot flew, the more concerned he became that it might not be the correct runway because it appeared much shorter than 3500 metres which he recalled was the length of Runway 01 and because he could not see the Domestic Terminal to the left of the runway. He questioned whether the Tower Controller had his aircraft sighted and was advised that he could not be seen from the Tower. The pilot then noticed that the CDI was indicating the aircraft to be well left of centreline. He commenced a gentle right turn and was increasing engine power to go around from the approach when the aircraft struck the light pole.

1.1.6 The pilot reported that the aircraft landing light was not selected on for the approach.

1.1.7 In the days preceding the flight, the pilot had studied closely the Instrument Approach to Land Procedure (IAL) Charts for Brisbane Aerodrome because he planned to conduct the flight under the Instrument Flight Rules (IFR). He stated that he was aware of the aerodrome runway layout, the position of the Domestic Terminal and General Aviation parking area but that he did not study in any great detail the International Apron Chart as it did not appear relevant to his operation. Further, he did not study the Brisbane Visual Terminal Chart as he was conducting an IFR flight and considered the Instrument Approach Charts adequate for this purpose. The pilot stated that he did not study the Aerodrome Diagrams Chart for Brisbane as his experience indicated that these charts offered no information additional to that contained in the IAL Charts; nor did he consult the En Route Supplement as he felt that his reading of this publication in relation to other aerodromes had not been particularly helpful.

1.2 Injuries to Persons

Injuries Crew Passengers Other Fatal - - - Serious 1 3 - Minor - - - Total 1 3

1.3 Damage to Aircraft

The aircraft was substantially damaged from impact with the light pole and the runway surface.

1.4 Other Damage

The streetlight pole was destroyed during the accident sequence.

1.5 Personnel Information

1.5.1 Pilot. The pilot in command was aged 58 years. He held a current Commercial Pilot Licence and a Command Instrument Rating for multi-engined aircraft. His licence was appropriately endorsed to allow him to fly in command of Beech Baron aircraft.

1.5.2 At the time of the accident, the pilot had a total flying experience of 673 hours of which seven were on Beech Baron aircraft. His most recent proficiency check was on 15 March 1990 when he completed an endorsement on Beech Baron aircraft. The pilot had previously held an ILS Rating for multi-engined aircraft. However, the rating was not current at the time of the accident.

1.6 Aircraft information

1.6.1 The aircraft was manufactured by Beech Aircraft Corporation in 1977. It was a low wing, six seat, twin piston engined aircraft with a maximum take-off weight of 2313 kilograms. At the time of the accident, the aircraft was fitted with four seats, two in the front and two in the centre row. It was predominantly white in colour.

1.6.2 Loading The weight and centre of gravity of the aircraft were within specified limits, and there was adequate fuel on board the aircraft for the completion of the flight.

1.6.3 Maintenance and Serviceability The aircraft had a current Certificate of Airworthiness and a valid Maintenance Release. No maintenance was outstanding at the time of the accident. Examination of the wreckage did not reveal any defect that might have contributed to the accident.

1.7 Meteorological Information

1.7.1 At the time of the accident, the surface wind at Brisbane Airport was north-easterly at 4 knots. Visibility was measured at 25 km at 1030 hours and 30 km at 1100 hours. There were 2 oktas of cumulus cloud at 2500 feet and 5 oktas of stratocumulus cloud at 5000 feet. There were showers in the area, particularly to the south-east of the airport and a weather observation at 1045 hours EST noted a recent shower at the aerodrome.

1.8 Aids to Navigation

1.8.1 All navigational aids at Brisbane Airport, including the VOR, Distance Measuring Equipment (DME), and ILS were operational at the time of the accident.

1.9 Communications

1.9.1 All Brisbane Air Traffic Services, including Brisbane Approach and Tower frequencies were operating normally at the time of the accident.

1.10 Air Traffic Control Procedures

1.10.1. The procedures applied by Brisbane Approach and Tower Control to VH-TSM during its approach into Brisbane were standard and no abnormalities were noted. In brief, the aircraft's approach was being radar monitored under the control of Brisbane Area Approach Control until the pilot reported the runway in sight at which time he was directed to transfer to Brisbane Tower frequency.

1.10.2 The Brisbane Runway 01 Approach Lights were on at the time of the accident.

1.10.3 The Tower Controllers followed the normal practice of attempting to sight the aircraft on approach to Runway 01 both with and without the aid of binoculars. When it could not be seen in this area, they began checking the approach to the decommissioned runway but did not sight the aircraft until after it had crashed.

1.11 Brisbane Airport

1.11.1 Brisbane Airport is owned by the Australian Government and operated by the Federal Airports Corporation (FAC) and is situated 27`23'09"S 153`06'59"E. The main runway, Runway 01/19, is 3500 metres long and 45 metres wide and constructed of asphalt.

1.11.2 Four kilometres south-west of Brisbane Airport lies the decommissioned Brisbane aerodrome. The northern 1150 metres of Runway 04 of the old airport fell within the Brisbane Airport boundary. The remaining southern section of the runway had been dug up.

1.11.3 The decommissioned Runway 04 formed part of the taxiway system from the present International Terminal area to the taxiway system for the new airport. It was also used on occasions for aircraft parking. A road crossed the runway 50 metres from its southern end. On the northern side of the road was the Brisbane Airport boundary fence while a series of floodlights bordered the southern side of the road. The pole for one of these light was positioned some two metres right of the decommissioned runway centreline. Both the light pole and the boundary fence were difficult to see from the air against the runway surface background.

1.11.4 There were visible for the first 240 metres on the old Runway 04 surface four white runway centreline markings, each of the standard 30 metres length. The centreline markings for the next 450 metres had been painted over but were still discernible from the air. The remaining northern section of the runway contained a continuous white centreline taxiway marking. There were no runway threshold markings or numbers at the southern end of the runway section. A line of six white cone markers painted with a 0.25 metre wide horizontal red band was positioned some 200 metres from the southern end of the runway section. There were no white unserviceability crosses on the runway.

1.11.5 The Brisbane Airport Control Tower is situated adjacent to the airport new terminal complex. The distance from the Tower to the threshold of Runway 01 is some 1650 metres while the distance to the southern end of the section of Runway 04 is 4000 metres. The Approach Radar Transmitter site is positioned approximately 1100 metres west of the southern end of the section of Runway 04.

1.12 Approach Radar Display

1.12.1 The radar display for the Approach Controller in the Brisbane Area Approach Control Centre allows a fairly accurate assessment of aircraft range to within about one kilometre of the runway threshold. However, accuracy with respect to aircraft position in azimuth is limited at close ranges and precludes the accurate determination of an aircraft's position with respect to the centreline of Runway 01.

1.13 Recorded Radar Information

1.13.1 The recorded primary radar information of the track flown by VH-TSM during the final stages of the flight was examined. It indicated that the aircraft was tracking about 360 degrees until nine miles from Runway 01. The track then diverged left to about 335 degrees crossing the Runway 01 centreline at six miles (10 kilometres) final. At about five miles (eight kilometres) final the track gradually veered right to line up on Runway 04.

1.13 Flight Recorders

1.13.1 The aircraft was not equipped with a flight data recorder, nor was there any requirement for it to be so equipped.

1.14 Wreckage and Impact Information.

1.14.1 The right wing root of VH-TSM struck the light pole one metre below the light/reflector housing at the top of the pole and 15 metres above ground level. The seven metre upper section of the pole became embedded in the wing root and remained attached until the aircraft came to rest. The light/reflector housing contacted the right side of the windscreen and the right front side window, breaking both sections of Perspex. The housing also contacted the right propeller, causing the right engine to stop.

1.14.2 The aircraft impacted the runway surface right wingtip first 54 metres beyond the light pole and two metres right of the centreline. The aircraft attitude at impact was approximately 45 degrees nose down, 15 degrees right bank and 15 degrees right yaw. The nosewheel and left main wheel were torn off and the mounts for both engines broken by the impact. There was also substantial crushing damage to the lower forward fuselage. This extended to wrinkling of the wing carry-through spar, indicating the extent of energy absorbed by the fuselage structure. The aircraft came to rest after skidding 52 metres along the runway surface.

1.15 Medical Information

1.15.1 The pilot was in good health at the time of the accident. He had recently had his visual acuity checked and had new spectacles prescribed which he was wearing at the time of the accident.

1.16 Survival Aspects

1.16.1 Seating and Seat Restraint. The front seat occupants were restrained by lap/sash harnesses while the centre row seat occupants wore lap belts only. There were no seat belt failures during the impact sequence, but all seats and/or seat mounts were damaged. The front left (pilot) seat remained anchored, but the seat back failed at the rotation bar position. The front right seat was similarly damaged and also had moved forward with the left foot forced off the seat rail. On the left centre row seat, the two feet were forced off the rails, allowing the seat to pivot about the front attachments. The right centre row seat remained attached to the seat rails, but the rails had been torn from the floor at the rear and centre attachment points. The centre row seats were found to be fitted with sash type shoulder harnesses, but these had not been used because they were underneath seat covers.

1.16.2 Injuries The pilot and front right seat passenger both sustained facial/head injuries as a result of being thrown forward and to the right when the aircraft impacted the runway. The passenger in the left centre row seat suffered a fractured ankle as the seat rotated forward. The passenger in the centre row right seat suffered no serious injuries other than bruising.

1.16.3 Although the results could not be quantified precisely, calculations indicated the aircraft experienced a deceleration level of approximately 17 Gs over a period of about 0.1 seconds at the time of the major impact on the runway surface. The extent of injuries to the occupants supports these figures.

1.17 Maps and Charts

1.17.1 The pictorial layout of Brisbane Aerodrome (including all runways and taxiways) was depicted in a number of Civil Aviation Authority publications current at the time of the accident. These included Aerodrome Diagrams (ADDGM), the Brisbane/Maroochydore Visual Terminal Chart, and Departure and Approach Procedures. On two places on the Visual Terminal Chart there was a "CAUTION ABANDONED RUNWAYS" notice directing attention to the location of the decommissioned airport. In the Enroute Supplement Australia, under Brisbane, Special Procedures, was advice of a decommissioned aerodrome four kilometres SW of the airport.

2. ANALYSIS

2.1 Preflight Preparation

2.1.1 The pilot did not study all available information appropriate to the intended operation. It is perhaps ironic that of the two publications he did not study, the Brisbane VTC contained a caution notice indicating abandoned runways at the site of the decommissioned aerodrome, while the En Route Supplement Australia, under Brisbane, in the Aerodrome and Facility Directory, contains a Special Procedure notice referring to "Decommissioned aerodrome 4 km SW of airport"

2.2 The Approach into Brisbane - Pilot Aspects

2.2.1 The pilot's account of the approach indicated that he initially was using the CDI as an aid in locating Runway 01. However, as the aircraft came closer to the aerodrome, his attention seems to have been diverted almost exclusively outside the cockpit towards locating the runway. There could be a number of reasons for this. The weather conditions, in particular visibility, around the time of the accident were good and more than adequate for a visual approach to be authorised. However, the pilot's comment to Brisbane Approach that in-flight visibility was reduced might have been caused by one of the rain showers observed in the area around the time of the accident. Also, because the pilot was not familiar with the area, he did not know what ground features to look for to locate the runway. These influences may have led to some anxiety on the part of the pilot. At the same time, the two requests by Approach to report the runway in sight might have been perceived by the pilot as additional pressure and further channelled his attention outside the cockpit. While his attention was outside the cockpit, he denied himself the information from the CDI as to the aircraft's position in relation to the runway centreline. In fact, in the visual meteorological conditions prevailing, and even though he did not hold a current ILS Rating, the pilot could have conducted a practice ILS approach and thus flown down the Runway 01 extended centreline until he located the runway visually.

2.2.2 What appeared to have been of significance in the pilot locating the decommissioned runway was the heading change left he made when advised by ATC that the aircraft was about one mile right of centreline. The net result of this was, in effect, to position Runway 01 about 40 degrees right of the nose of the aircraft. When the pilot reported the runway in sight, he was looking directly ahead of the aircraft (i.e. towards the decommissioned runway) instead of the 1.30 o'clock position towards Runway 01

 2.2.3 A number of reasons can be advanced for the pilot continuing the approach to the late stage that he did before commencing an overshoot. Firstly, there was the appearance of the decommissioned runway, including its orientation close to that of Runway 01, visible centreline markings, and absence of white unserviceability crosses. Secondly, while the pilot was in some doubt as to whether he was approaching the correct runway, this doubt was not confirmed until very late in the approach when he noticed the CDI indicating the aircraft to be well left of centreline. Finally, there were no obstructions apparent to the pilot in the undershoot area or runway vicinity which would have prompted him to go around from the approach. Notwithstanding these aspects, there were also many visual features which could have made the pilot realise earlier that he was approaching the wrong runway. That these were not seen or acted upon by him is indicative of his attention being channelised towards the runway and not its environs. This is considered a normal mode of pilot behaviour during the final stages of an approach.

2.3 The Approach into Brisbane - ATC Aspects

2.3.1 The pilot reported the runway in sight when the aircraft was about 3.5 miles (6.5 kilometres) from the threshold of Runway 01 or two miles (3.5 kilometres) from the threshold of the decommissioned runway. The aircraft was then transferred to Brisbane Tower frequency from which point there was no further requirement for it to be monitored by the Approach Control. In any case, the Approach Radar Display was not capable of providing a meaningful indication of the aircraft's position at this close range with respect to the Runway 01 extended centreline.

2.3.2 Tower was advised by Approach Control that the pilot did not report sighting the runway until about 3.5 miles from the threshold of Runway 01 as part of the handover of control of the aircraft to Tower. Tower had no reason to expect other than for the aircraft to have been on the normal line of approach for Runway 01. It was reasonable for some time to elapse while attempts were made to sight the aircraft in this area. When no sighting was made, the approach to the decommissioned runway was checked. In the event, the aircraft was not seen in this area until just after it had crashed. However, there were reasons for this.

2.3.4 Looking from the Tower at the last 1.5 miles (2.8 kilometres) of approach to Runway 01 the change in azimuth was from 181` M to 157`M. For the decommissioned runway, the change was from 208`M to 202`M. At one mile from the threshold for Runway 01, the aircraft was about 3.5 kilometres from the Control Tower. At one mile from the threshold of the decommissioned runway, the aircraft was almost six kilometres from the Control Tower. Thus, in attempting to sight the aircraft on approach to the decommissioned runway, the Tower Controllers were confronted by a target with small lateral movement at a substantial distance. The problem of visual acquisition was compounded by the light colour of the aircraft against an urban background, its small size, and the fact that its landing light was not on during the approach.

2.4 Runway 04 Markings

2.4.1 Aeronautical Information Publication Australia (AIP), Aerodromes and Ground Aids (AGA) details the requirements for aerodrome markings. AGA-5-3-2 refers to Unserviceable Areas and states that the limits of unserviceable areas are delineated by white cone markers painted with a 0.25 metre wide horizontal red band. It also states that, except in the case of total (aerodrome) unserviceability or restricted operations, unserviceable areas on the movement area are marked by the display of unserviceability cross markers on the affected area. Such a marker consists of a white cross with arms at least 6 metres long and 0.9 metres wide.

2.4.2 The six white with horizontal red band cone markers across the runway 210 metres from the southern end marked the northern limit of an unserviceable area adjacent to the road and boundary fence. This area should have been marked with white unserviceability crosses. Had it been so marked, the pilot might have seen the crosses and become aware earlier that he was approaching the wrong runway.

3. CONCLUSIONS

3.1 Findings

1. The pilot was medically fit, correctly licenced, and qualified to undertake the flight.

2. No evidence was found that the aircraft was not capable of normal operation at the time of the accident.

3. The pilot did not, before the beginning of the flight, study all available information appropriate to the intended operation.

4. Weather conditions were suitable for the aircraft to conduct a visual approach.

5. The procedures followed by Brisbane Approach Control and Brisbane Tower during the lead up to the accident were reasonable and in accordance with approved procedures.

6. The pilot misidentified the decommissioned Runway 04 at the old Brisbane Airport for Runway 01 of the current airport.

7. The characteristics of the Brisbane Approach Control radar display precluded an accurate assessment of the aircraft's position in azimuth during the latter stages of the approach.

8. Brisbane Tower was unable to visually locate the aircraft during its final approach.

9. There were no white unserviceability crosses on the southern end of the remaining section of the decommissioned Runway 04.

10. The pilot did not see the light pole against the runway surface background.

11. The pilot elected late in the approach to Runway 04 to discontinue the approach and go around.

Significant Factors:

1. The pilot was unfamiliar with Brisbane Airport and its environs.

2. The pilot's preflight preparation was inadequate in that he did not, before the beginning of the flight, study all available information appropriate to the intended operation.

3. Probably because of concern or apprehension on his part, the pilot's attention was channelised outside the cockpit towards sighting the runway. He was thus denied cockpit information via the CDI indicator as to the aircraft's position in relation to the Runway 01 centreline.

4. The southern end of the remaining decommissioned section of Runway 04 was not marked with white unserviceability crosses.

5. The pilot did not see the light pole.

The aircraft had been chartered to convey seven passengers, their luggage and a quantity of freight from Brisbane to Trinidad Station. When the pilot arrived at Archerfield Aerodrome on the morning of the flight, he found that the aircraft had a fuel load that was too great to allow the expected passenger and freight loading to be carried. He therefore arranged for the aircraft to be defuelled. When the passengers and freight arrived, the pilot learned that the freight load was less than had been originally advised and he stopped further defuelling. The aircraft departed Archerfield at 0900 hours and arrived at Charleville at 1045 hours, where additional fuel was added before completing the flight to Trinidad Station. After arriving in the vicinity of Trinidad Station the pilot experienced some difficulty in locating the strip. He consulted several passengers about its position in relation to a seismic survey camp that had been seen on descent. The strip was located after about five minutes and the pilot positioned the aircraft to join the circuit on the downwind leg for landing into the northwest. On final approach, a passenger who was seated in the third row of seats observed that the landing gear position indicating lights on the instrument panel indicated that the gear was up. He shouted and drew the attention of the passenger seated in front of him to the problem, who in turn tapped the pilot on the shoulder and pointed to the landing gear position indicator lights. Almost immediately both propellers struck the ground. The pilot applied power and rotated the aircraft. The left engine stopped as a result of the propeller striking the ground and the pilot was seen adjusting the engine controls, apparently either attempting to feather the left propeller or restart the left engine. The aircraft climbed to between 150 feet and 300 feet above the strip, where it levelled out before gradually yawing and banking to the left. The yaw and bank increased and the aircraft descended, struck the ground on the left engine and nose section 310 metres to the left of the strip and slid backwards for about 20 metres before coming to rest. Inspection of the wreckage revealed that the landing gear was up and had not been selected down. The aircraft was fitted with a gear warning system that activated an aural alarm when the engine throttles were retarded below 14" manifold pressure and the gear was not in the down position. Although this system was serviceable at the time of the accident, it was company policy that the pilots not reduce the throttle settings below 15" manifold pressure until the aircraft had touched down. It was estimated that during the go-around attempt the available power from the right engine would have been reduced by at least 25 per cent due to propeller blade damage.

The pilot had purchased the aircraft some 12 months previously but had carried out only limited and infrequent training. This training had not progressed to the stage where turns were performed, and was limited to straight and level hops along the strip. On the morning of the accident the pilot had performed several hops, taking off and landing into a light easterly breeze. On about the fifth flight the aircraft climbed to a height of about 85 feet, levelled off momentarily and then descended with reduced power as if for a normal landing. When the aircraft was about 50 feet above the ground and close to the end of the landing area, power was applied, a climb was made to about 200 feet and turns were made to position the aircraft on a downwind leg. Witnesses noted that the aircraft was under control until it suddenly adopted a nose-high attitude, followed by a rapid pitch-down which continued into a tumbling motion. Two in-flight impact noises were heard above the sound of the engine and shortly afterwards the aircraft struck the ground in a steep nose-down attitude while cartwheeling to the left. Wreckage examination showed that there had been two strikes on the tail fin and rudder by the rotor blades, causing the rudder to become detached from the aircraft in flight. It was considered that the pilot endeavoured to carry out a circuit of the strip after he perceived that there was insufficient distance remaining to safely land the aircraft. The reason for the sharp change of attitude on the downwind leg could not be positively established, but was probably due either to turbulence or an incorrect control input by the pilot. When the nose pitched down, the airflow through the rotor disc would have been significantly reduced. This would have led to a loss of rotor RPM and a consequent loss of performance of the rotor to the stage where control of the aircraft could not be regained.

The aircraft was being operated on a regular freight carrying run between Dysart and Rockhampton. On the evening preceding the accident, the aircraft departed Dysart at 2045 hours, arriving at Rockhampton at 2139 hours. At 0222 hours on the following morning the pilot telephoned Brisbane Flight Service Centre to obtain an update on the weather for the return flight to Dysart. The forecast indicated that at the time of arrival at Dysart, light rain with 5 oktas of cloud at 700 feet above ground level, could be expected. The pilot nominated Rockhampton as the alternate for Dysart in the event that weather conditions at the latter aerodrome precluded a safe approach and landing. The aircraft was refuelled to give a total fuel endurance of 290 minutes, and after loading departed Rockhampton at 0354 hours. At 0435 hours, the pilot reported that descent had been commenced into Dysart and because of poor reception on High Frequency radio channels further communications with VH-CLU were relayed by another aircraft in the area, VH-EEF on Very High Frequency channels. After communications with VH-CLU, the pilot of VH-EEF advised Brisbane Flight Service Unit (FSU) at 0453 hours, that the runway lighting was not yet displayed. The runway lighting at Dysart was provided by a number of hand-lit flares. The lighting of these flares was carried out by an employee of the company operating VH-CLU, and normally took about ten minutes. On this occasion the employee had overslept and arrived at the aerodrome at about the same time that the aircraft flew overhead. At 0501 hours further attempts by the pilot of VH-EEF and Brisbane FSU to contact VH-CLU were unsuccessful. The wreckage of the aircraft was later located about 800 metres to the west of the aerodrome. The aircraft had struck trees while heading in a direction aligned with runway 14 but displaced to the west of the runway. It had been destroyed as a result of the impact forces. Witnesses reported that when VH-CLU arrived at Dysart it was not raining, however, low cloud was present. The aircraft was observed to complete three orbits of the aerodrome and at times during these orbits it was obscured by cloud. The only fault found with the aircraft during the investigation was a failed fuel pump on the right engine. It was established that the right engine was operating on impact and the failure of the fuel pump is not considered to have substantially affected the operation of the aircraft. The reason the aircraft crashed was not established.

The pilot was conducting the final stage of a night freight flight from Melbourne to Hobart and return. Because of traffic in the Melbourne/Essendon area, the aircraft was radar vectored to the south of Essendon. In the vicinity of the Westgate Bridge the pilot was instructed to carry out two orbits, in order to facilitate his approach to Melbourne. These orbits were conducted with the aircraft in cloud. The pilot was then instructed to take up a north-westerly heading, and to descend to 2000 feet (above mean sea level). On gaining visual contact with the ground, the pilot saw an aerodrome ahead, which he assumed was Melbourne. The aircraft was too high for a direct approach to this aerodrome, and the pilot requested approval to carry out a further orbit. The aircraft was advised to take up a westerly heading, and the pilot was instructed to call the Melbourne Tower controller. The Tower controller at Essendon was aware that the aircraft was enroute to Melbourne. After he observed the aircraft passing abeam Essendon, he cleared a waiting aircraft to take-off into the south. Shortly afterwards, he saw the first aircraft commence a descending turn to the left and track towards Essendon. The controller alerted the Melbourne Tower controller to the situation, and instructed the aircraft which had just become airborne to turn to the left. The pilot of this aircraft saw the approaching aircraft, and adjusted his flight path accordingly. The two aircraft passed with about 200 feet of vertical separation, and were some 200 metres apart horizontally. Melbourne Tower advised the inbound pilot he was tracking for the wrong aerodrome, and instructed him to carry out a go around. At this time the pilot realised his error, and the remainder of the flight was uneventful. It was determined that the pilot had become disorientated as he gained visual contact with the ground. When he sighted an aerodrome, he had a fixation that it was Melbourne, and he had disregarded visual and aural cues that he was proceeding to the wrong aerodrome. He was not familiar with the area, and it is highly likely that he was suffering the effects of fatigue.

The culler had just made a kill and the pilot prepared to land so that the beast could be bled. He selected an area of short grass and set the helicopter down with the engine over the barest area. Prior to landing the pilot briefed the culler on the need to deplane quickly to minimise the risk of a ground fire from grass ignited by the hot exhaust. The pilot intended to take-off as soon as possible and return for his passenger when the ground task was finished. Soon after the passenger had deplaned, the pilot was attempting to build up the rotor RPM when he became aware of smoke and flames. He reported that the engine was not responding to throttle input, and when he pulled the collective for liftoff, the engine lost all power and the helicopter settled onto the ground from a few feet. The aircraft was enveloped in flames as the pilot left the cockpit, and it was destroyed by fire. A fenceline cleared of grass existed some 70 metres away, but the pilot was intent of remaining clear of dust areas to prevent wear in the tail rotor assembly. He judged that there was a minimum risk of fire in the area selected for landing, and it was clear of the obstruction posed by the fence. There were no reported aircraft defects or fuel leaks.

The aircraft was spreading Mexican Sulphur, a mixture of sulphur and a carrier base. Operations were being conducted from a one-way strip in a 2 to 3 knot tailwind, and an appropriate reduction in payload had been made by the pilot commensurate with the conditions. It was reported that operations had proceeded slowly as the spreading medium would not flow freely, tending to clog in the clamshell gate. This required the pilot to work the gate continuously to achieve an even discharge. A witness reported that during the sixth takeoff of the day, a strong wind gust substantially increased the tailwind. The aircraft did not become airborne, and the pilot pulled the hopper handle in an attempt to reduce the weight of the aircraft in order to clear a boundary fence. The load failed to dump normally. The pilot then selected additional flap and the aircraft became airborne in a tail-low attitude. Although the mainwheels cleared the fence, the tailwheel struck a fence post. The aircraft remained airborne, but the pilot found it necessary to hold the control column in the fully-forward position. After some altitude was gained, the pilot made a further and successful attempt to dump the load. Although difficulty was experienced in controlling the aircraft, a safe landing was made at another aerodrome nearby. The aircraft sustained substantial damage to the rear fuselage and empennage. Empennage damage included buckling of the tailplane and rudder trailing edge. It was also found that spreading medium had clogged the hopper gate, preventing emergency dumping of the payload, and the hopper gate mirror was incorrectly aligned so that the pilot was unable to observe load release. This accident was not the subject of an on-site investigation.

The helicopter crew was responding to an emergency call. The aircraft was cruising at 1500 feet above sea level, at about 120 knots, with maximum continuous power applied. As descent was commenced there was a loud bang, followed by severe vibration. The helicopter was briefly out of control and the vibrations were such that the pilot was unable to read the instruments. Power and speed were reduced and the pilot was able to regain control. Because of expected difficulties with control near the ground, the pilot elected to carry out a water landing, some 80 metres from a beach. A subsequent investigation determined that the left engine access cowl had become unlatched, lifted and been struck by the main rotor blades. The hinged section had been struck at least twice and the upper fixed cowl, which had been torn from its fastenings when the hinged cowl lifted, at least once. Apart from minor impact marks to other main rotor blades, a piece of cowling had slid along one blade, causing 2.1 metres of the outboard trailing edge to separate, resulting in severe unbalance to the rotor disc. The pieces of engine cowling and main rotor blade were found approximately two kilometres south of the accident site. As two lower main cowl latches were not recovered, the reason for the cowl becoming unlatched could not be positively determined. However, during the investigation, it was found that the lower cowl latches, which appeared to be correctly locked and engaged, could release when the cowl was subjected to simulated inflight loads. The manufacturer has suggested modifications to improve the cowl latching system.

Reccomendations:

It is recommended that the Civil Aviation Authority evaluate the suggested modifications and, if satisfactory, direct their incorporation in Australian registered Kawasaki BK117 aircraft.

The aircraft underwent a 100 hourly inspection six days prior to the accident. During that inspection three oil leaks were rectified. When the aircraft commenced operations again after the 100 hourly inspection, the engine again started to leak oil at a significant rate. The operation was remote from the maintenance organisation. The pilot did not determine the source of the oil leak nor monitor the oil consumption rate during the days prior to the accident. On the day of the accident the pilot had been spreading superphosphate since late morning. He had finished that task and had commenced spreading clover seed. On the second seeding sortie he sensed a loss of performance. He pushed the throttle lever forward but the engine did not respond. He checked the RPM and noted it was 2200 and decreasing. He then noted that the oil pressure was indicating zero. At this stage he was about 50 feet above the ground and operating in mountainous terrain. He turned the aircraft to the right to avoid an area of trees and try to land on a steep upslope. The aircraft stalled into the upslope and came to rest after a 25 metre ground-slide. The investigation revealed that the oil filter bolt and sleeve had failed as a result of high cycle low stress fatigue. This was consistent with the retaining bolt being under-torqued. This resulted in the oil filter separating from the engine, a consequent loss of engine oil and finally, loss of engine power.

According to his log book, the pilot had flown the aircraft on two previous occasions of about 15 minutes duration each. The pilot was briefed by the owner of the aircraft on power settings and speeds. He then took-off into the south from the strip which is aligned approximately north-south. There was a light southerly breeze blowing at the time. The aircraft was seen to make one right-hand circuit, but it could not be determined if the aircraft actually made a landing from that circuit. When the aircraft was on final approach from a second right-hand circuit at an estimated altitude of about 200-300 feet, it entered a turn to the right. As the turn progressed onto a northerly (downwind) heading, the angle of bank became progressively steeper until it was about 90 degrees. The nose then dropped and the aircraft dived to the ground. The first persons on the accident scene experienced difficulty in removing the pilot's motor cycle type safety helmet (with full face enclosure) before resuscitation could be applied. The investigation did not reveal any pre-existing defects that may have caused the accident. Other pilots reported some turbulence and windshear in the area where the accident occurred. The area was inspected by an officer from the Bureau of Meteorology who specialised in micro-meteorology. His opinion was that turbulence or windshear effects at the time of the accident would have been minimal. The pilot had been receiving dual flight instruction from an ultralight flying school. After a dual instructional flight on the morning of the accident, he had been strongly advised by his instructor not to fly solo until he had completed more dual instruction as the instructor believed that the pilot was not sufficiently competent to fly solo. The pilot lost control of the aircraft for reasons which were not determined.