The pilot departed on a mail run to his first port of call at Yalcowinna. He was unable to land, as the strip was covered with water from recent heavy rains. He diverted to the next strip at Mawarra, and recalled overflying the strip which he assessed as being wet but suitable for landing. The strip owner usually contacted the aircraft by radio to advise on the surface conditions, but on this occasion the pilot was unable to make contact before landing. The aircraft became bogged after turning off the strip following landing. After being extricated and the turnaround completed, the aircraft again became bogged while taxiing for takeoff. After once again freeing the aircraft, the pilot positioned it at one end of the strip ready for takeoff. He then shut the engine down, left the aircraft and inspected the strip from a vehicle. Having assessed the surface as suitable, the pilot attempted a takeoff with two stages of flap extended. At a previously selected decision point, the pilot elected to continue the takeoff and attempted to lift off at an indicated airspeed of 50 knots. The aircraft settled back onto the strip on the main wheels, and remained in this nose high attitude until the pilot decided to abandon the takeoff. The aircraft came to rest about 1400 metres beyond the end of the 700 metre long strip, after passing beneath telephone wires which stood some 200 metres past the strip end. The nosegear collapsed towards the end of the ground roll. The pilot did not recall being briefed by his employer that he should not land at Mawarra if Yalcowinna was unserviceable due to surface conditions.

The pilot was conducting agricultural spraying operations on a paddock which had a powerline crossing it at an angle and a fence along the western boundary. On completion of a spray run to the north, the pilot returned for a run to the south along the fence line. The marker was required to climb over the fence into the next paddock and move away from the flight path of the aircraft, prior to the commencement of the run along the fence. As he approached the powerline, with the right wing almost over the fence, the pilot noticed that the marker had delayed climbing over the fence. He took evasive action to avoid striking the marker and during this attempt the aircraft contacted the powerline. The aircraft deflector cable failed, allowing the power line to sever the rudder from the aircraft. The pilot carried out a precautionary landing on a road. During the landing roll, the left brake failed and the aircraft ground looped. The investigation revealed that the pilot misjudged the clearance between the aircraft and the powerline. The deflector cable was severed by the impact with the powerline and the rudder was torn from the aircraft. The left main landing gear leg was found to have failed from a combination of bending and twisting. It is considered likely that the brake failure occurred as the leg and attached brake line were torn from the aircraft, depriving the pilot of directional control. There were no pre-existing defects with the aircraft which could have contributed to the accident.

The pilot was landing at a strip aligned 070 degrees. He estimated the wind to have been gusting 20-25 knots from 090-120 degrees. Neither drift nor turbulence were excessive on finals. However, as the aircraft was flared, it suddenly rolled rapidly left to what the pilot judged was more than 60 degrees of bank. Despite the application of full power and opposite rudder, the pilot was unable to prevent the left wing contacting the upper branches of trees to the left of the strip. This resulted in the aircraft striking the ground nose first and coming to rest 80 metres left of the strip. The conditions existing at the time of the accident were such that strong wind gusts were possible. Other pilots who had used the strip reported experiencing wind gusts at a similar stage of the approach.

The pilot had earlier landed in the paddock which was adjacent to the homestead to seek assistance regarding a broken down vehicle being used in the spraying task. Prior to commencing the takeoff run the pilot did not inspect the proposed takeoff path. Initially the aircraft takeoff run acceleration was more sluggish than normal, due to the traversing of patches of soft wet ground. Acceleration improved. However, the aircraft was approaching some trees so the pilot turned it to provide a greater ground roll distance. The aircraft became airborne for a short time, but then touched down in long wet grass which degraded its acceleration. It then hit a fence and nosed over.

The pilot was conducting a bank-run flight. It was raining heavily when the aircraft arrived at Canberra, and the pilot took shelter while freight agents attended to the load. When this operation was completed, the pilot ran back to the aircraft, quickly boarded and prepared for DEPARTURE. Although Runway 35 was the runway in use, he was given a taxy clearance for DEPARTURE from Runway 12, as the cloud base and visibility were better in that direction, and the wind was light and variable. It was still raining as the takeoff roll was commenced, with water accumulation on the runway retarding the acceleration of the aircraft. At the start of the roll the pilot heard a tapping noise coming from the left hand side of the aircraft, which he thought may have been a strap from his seat belt hanging out of the door, but a quick check confirmed that his seat belt was correctly fastened. As speed increased the noise subsided, and as both engines appeared to be operating normally, the takeoff was continued. However, at the point of rotation the noise returned, louder than previously, and the pilot suspected that there was a fault in the left engine. As it was apparent that the aircraft would enter cloud shortly after becoming airborne, the pilot made a decision to abandon the takeoff. The aircraft touched down approximately 550 metres from the end of the runway, and during the application of heavy braking the aircraft aquaplaned along the wet runway. After crossing the threshold it continued through the boundary fence, across a road, then came to rest in an adjoining paddock. A length of the cargo restraint webbing was found hanging out of the cabin door, allowing it to flap against the side of the fuselage. Canberra Runway 12/30 has a known drainage problem during rain periods, and other aircraft have also experienced aquaplaning following abandoned takeoffs.

The aircraft had been constructed by the pilot in his engineering workshop from plans supplied by the designer, Bushby Aircraft Incorporated of Minooka, Illinois, USA. During some stages of the construction the pilot had been assisted by Mr R.S. Howarth. The first flight of the aircraft had been achieved on 5 April 1982 and by the day of the accident some 50 hours had been flown, with Mr Waggott being the pilot on each occasion. The only known problems of significance that had been encountered during this period was wind noise and rattling associated, apparently, with the canopy. On 3 July 1982 the pilot had been carrying out adjustments to the wing root fillets, again apparently to reduce wind noise. The pilot was then joined by Mr Howarth and after boarding the aircraft the pilot advised Bankstown Tower that they would be proceeding to the Bankstown training area. The aircraft was cleared for take-off and the response to that clearance at 1433 was the last recorded transmission from the aircraft. The aircraft is not known to have been seen by any other persons until about 1508 when ground witnesses observed it about two and a half kilometres south-east of Luddenham, heading in a south-westerly direction. The witnesses reported that the engine noise sounded uneven and that the aircraft was rolling to the right. Other witnesses further along the flight path observed what they referred to as "confetti" floating down from the aircraft. It crossed the Luddenham - Bringelly road and as it passed over Vicary's Winery pieces of plexiglass, the cockpit canopy and the occupants headsets, glasses and caps fell to the ground. The aircraft continued in a south-westerly heading, descending toward open country in substantially a level attitude, then rolled to the right and impacted the ground in an inverted attitude near a house. Inspection of the wreckage found no defect or malfunction that would have affected the safety of the operation existed, except that the canopy had become detached in flight. Examination of the canopy revealed that it had not been manufactured in accordance with the aircraft designers plan. Different specification material had been used in the frame of the canopy and this had in turn been reduced in strength by the drilling of holes to allow the access to the nuts of the plexiglass attachment screws. The attachment of the canopy keeper brackets was also altered so that only one screw, instead of the designers intended three, held brackets to the roller assembly. Also the canopy latch mechanism had been altered so that it did not provide positive restraint in a vertical direction. The aircraft designer indicated that the loss of the canopy in flight should not have affected the pilot's ability to safely control the aircraft. Marks on the occupants headsets showed that they had been struck by the canopy, and it was considered probable that the occupants had been incapacitated as the canopy detached. ((1))

Circumstances:

The pilot hired the aircraft for a short flight to practise aerobatics as he intended to participate in a club aerobatic competition later in the day. The fuel quantity was checked by the pilot with a dipstick. There was a total of approximately 120 litres evenly distributed between the left and right wing tanks. The engine run up revealed no abnormalities prior to take off. The aircraft was climbed to 4000 feet in the training area where the pilot practised aerobatics for about five minutes. He then regained the height lost during the aerobatics before the aircraft was deliberately put into a spin. Recovery was effected at about 3000 feet. From 3000 feet the pilot closed the throttle and descended to circuit altitude. He said he cleared the engine on two occasions during the descent by opening the throttle briefly. The aircraft was positioned on a right downwind leg for runway 23 and normal cruise power set.

During the before landing checks the pilot said the engine commenced to run roughly after the fuel boost pump was switched on. The pilot turned onto a base leg where the engine lost all power. As there was insufficient height for the aircraft to glide to the aerodrome the pilot attempted a forced landing in a confined area about 750 metres from the northern boundary. The aircraft struck the ground with a high rate of sink and low forward speed. The pilot suffered back injuries and required assistance to vacate the cockpit. An examination of the aircraft and an engine test run failed to reveal the reason for the loss of power. Witnesses reported the mixture control knob was approximately 25 millimetres out from the full rich position, but it could not be determined if the mixture control was in this position prior to impact.

There is no provision other than internal friction in the mixture control cable to prevent movement of the mixture control. The atmospheric conditions were not conducive to intake icing in a fuel injected engine.

Significant Factors:

The following factors were considered relevant to the development of the accident:

1. The engine lost power inflight for reasons which were not positively established.
2. The loss of power occurred when the aircraft was outside gliding distance of the aerodrome.
3. The pilot was forced to land in an unsuitable area.

Before DEPARTURE, the pilot removed the fuel filler caps and checked that both tanks were full. He calculated the fuel endurance as 300 minutes. The flight proceeded normally with fuel selection being alternated between the left and right tanks. After a flight time of about 207 minutes, and with the left tank selected, the engine began to surge in a manner described by the pilot as typical of a tank running dry. The pilot immediately selected the right tank and turned the auxiliary fuel pump to "LO". The surging continued so he reselected the left tank and then the right tank, operating the auxiliary fuel pump on "HI" in short bursts. The engine did not restart. By this time, the aircraft had descended to about 2500 feet above ground level so the pilot ceased trouble-shooting and concentrated on landing the aircraft. The aircraft was landed in a cleared area but sustained damage when it nosed into a ditch late in the landing roll. Calculations indicated that 123 litres of fuel had been used from the left tank and 84 litres from the right tank. This was the first occasion since the pilot had been flying the aircraft that he had used more than 90 litres from the left tank. Inspection of the aircraft revealed that the right tank was about one third full. The left tank, however, was dry and five press studs attached to the upper surface of the fuel cell had become detached from the wing upper skin. This allowed the roof of the cell to sag, thus reducing the cell's capacity. The fuel gauge continued to indicate normally. There was, therefore, no way the pilot could have detected the fault other than to run the tank dry. The auxiliary fuel pump is operated by two switches situated side by side on the instrument panel. The right switch is labelled "LO" and is used for engine starting. It will only operate when the ignition switch is turned to start. The left switch, labelled "HI", is used for engine operation should the engine driven pump fail, and also when switching from an empty tank to one containing fuel. Tests determined that the auxiliary fuel pump in the aircraft was operating correctly on both settings. The pilot was under the impression that to leave the auxiliary pump on "HI" would flood the engine. For this reason, he selected "LO" and operated the "HI" switch in short bursts. These actions, along with changing back to the (empty) left tank, were insufficient to purge the air from the fuel lines in the short time the pilot was able to devote to trouble shooting the problem. A placard on the instrument panel lists the procedures to be followed in the event of a major fuel flow fluctuation/power surge. The first item on the list is "AUXILIARY FUEL PUMP ON" but there is no reference to the "HI" or "LO" switches. Had the placard read "AUXILIARY FUEL PUMP - HI", the uncertainty felt by the pilot concerning use of the "HI" setting may not have arisen.

Circumstances:

The aircraft was operating a scheduled service from Brisbane to Port Macquarie with planned intermediate stops at Coolangatta and Coffs Harbour. Weather conditions over the route were influenced by a widespread unstable airmass. The terminal forecast for Coffs Harbour indicated a surface wind of 360/15, visibility in excess of 10 km, 5 octas stratus at 1000 ft, 5 octas cumulus at 2000 ft. Thunderstorms, associated with visibility reduced to 2000 metres were also forecast for periods of up to 30 minutes. The actual weather conditions at Coffs Harbour were generally consistent with the terminal forecast. Runway 03 was in use throughout the evening. Coffs Harbour airport was equipped with NDB, VOR and domestic DME radio navigation aids. A VOR/DME procedure was published for runway 03 approaches. For aircraft not equipped with DME, a VOR or NDB approach was available using common tracking and minimum altitude criteria. Runway 03 was also equipped with a 6 stage T-VASIS and 3 stage runway lighting. All facilities were reported as functioning normally, with the exception of the VOR which was experiencing intermittent power failures due to the effects of heavy rain. The VOR was able to be reset manually from the Coffs Harbour control tower. Although the tower was scheduled to be unmanned before the arrival of VH HOX, the duty air traffic controller elected to man the tower until the aircraft had landed. The controller also called out a technician to attend to the VOR. The aircraft was equipped with dual ILS/VOR and ADF receivers, plus International DME. Domestic DME equipment was not fitted to the aircraft, although required by ANO 20.8. After descending in the VOR/NDB holding pattern, the aircraft was cleared for an instrument approach. The pilot had been told of the intermittent operation of the VOR and had said he would revert to the NDB. At that time, the weather conditions were fluctuating about the circling minima of 950 feet (QNH) and five km visibility. The controller advised the aircraft of a heavy shower to the south of the field. The aircraft subsequently completed the approach and the pilot reported "visual". The controller said he saw the lights of the aircraft in a position consistent with a right downwind leg for a landing on runway 03. The aircraft was then cleared to land. Shortly after, the controller saw the lights of the aircraft disappear briefly, consistent with the aircraft passing through a localised area of rain/cloud. The lights then reappeared briefly, as though the aircraft was turning onto finals, before disappearing. This was immediately followed by short series of "clicks" on the tower frequency. The aircraft was called immediately but failed to respond to any calls. The accident site was located about 1070 metres short of the landing threshold, and about 750 metres to the right of the extended runway centreline. The aircraft was found to have initially struck a nine metre high tree in a nose low attitude, steeply banked to the right, on a track of 050 degrees. After striking the tree with the outboard section of the right wing, the aircraft struck other trees before hitting the ground and overturning. A fire broke out shortly after the aircraft came to rest. As a result of his remaining on duty, the controller was able to provide immediate notification of the accident to the emergency services. This action facilitated the rescue of survivors. A subsequent examination of the aircraft structure, systems and components, found no evidence of any pre-existing defect or malfunction which could have contributed to the accident. The pilot was properly licenced and qualified to conduct the flight. Evidence was provided to show that the pilot had probably flown a total of 930 hours in the previous 365 days, thereby exceeding the ANO 48 limitation of 900 hours. Other breaches of Flight and Duty Limitations were found to have occurred during the previous 12 months, however, during the three months prior to the accident no significant breaches of ANO 48 were found which could have contributed to the accident. Specialist medical advice considered the 30-hour exceedance of the 900-hour limitation was not significant in this accident. Other specialist advice was obtained concerning the possibility of the aircraft being affected by low level windshear or a microburst during the final stage of the night circling approach. It was considered this was not a factor in the accident. Considerable evidence was presented during a subsequent Coroners' Inquest concerning allegations of irregular operating practices by the operator over a period of several years prior to the accident. Much of this evidence was only provided after the granting to witnesses of immunity from prosecution. Despite this, no new evidence was presented which related to the accident flight. The investigation concluded that, on the evidence available, the aircraft was turning onto a short right base leg when it entered a localised area of rain and low cloud. The pilot was required to look out of the right cockpit window to enable him to maintain visual reference with the approach end of the runway. It is considered probable that the pilot briefly diverted his attention from the flight instruments while attempting to maintain that visual reference as the aircraft passed through an area of reduced visibility. During that period, the aircraft continued to roll to the right, resulting in an inadvertent loss of height. The pilot was unable to effect a recovery before the aircraft struck trees.

Significant Factors:

The following factors were considered relevant to the development of the accident:

1. Low cloud base, with localised rain squalls and reduced visibility.

2. Low level, right hand, night circling approach.

3. Pilot lost visual reference at a critical stage of the approach.

4. Pilot did not initiate missed approach.

5. Pilot probably diverted attention from the flight instruments.

Recommendations:

1. During the course of the investigation it was recommended that the Civil Aviation Authority review its policies and procedures concerning the surveillance of commercial flying operations in general aviation. The objective of the review would be to facilitate the identification and rectification of improper operating practices within the industry. The Authority subsequently advised that the level or frequency of surveillance of commercial aviation operations was being increased.

The pilot reported that soon after takeoff from Cairns the right engine rapidly lost oil pressure accompanied by a rise in the oil temperature. There was also a noticeable drop in manifold pressure and engine rpm. The pilot feathered the propeller and advised the tower of his intention to returned to Cairns where a single engine landing was carried out. A local Licensed Aircraft Maintenance Engineer (LAME) was then engaged to investigate and rectify the defect. After checking the oil quantity and completing a visual inspection of the engine bay the LAME carried out a ground run to verify the extent of the oil pressure loss. All engine parameters including oil pressure and temperature indicated normal during the prolonged ground run. The LAME then advised the pilot that he considered the aircraft to be serviceable and explained to him that the probable cause of the pressure loss was that a piece of carbon had become temperarily lodged under the seat of the oil pressure relief valve. No further "trouble shooting" was carried out by the LAME. Soon after takeoff on the next flight the pilot noticed that the right alternator was not showing a charge. This defect was entered on the maintenance release on arrival in Townsville. The subsequent inspection of the aircraft discovered that the alternator drive "clutch spring" was badly distorted and had caused extensive scoring damage to the internal crankcase casting. The metal filings produced by this scoring action had contaminated the engine oil system. The extent of this contamination necessitated a bulk strip of the engine and accessories. It was found that the crankshaft all main bearings conrod bearings oil pump and oil pump housing had been substantially damaged.