At approximately 500 ft after take-off the engine began to run roughly.

The aircraft was immediately positioned for a landing back on to the departure runway. The engine was still developing sufficient power at this point to complete a climbing turn to 800 ft. Approximately mid-down wind, the engine began to run very roughly and then lost all power. In the subsequent forced landing, the aircraft touched down approximately one third of the way along the runway. The pilot was unable to bring the aircraft to a halt in the distance remaining to avoid overrunning the the end of the strip. The aircraft came to rest part way down the sloping sea wall at the end of the runway. All three occupants were able to evacuate the aircraft safely.

Examination of the engine found that the centre electrodes for both spark plugs in the number four cylinder were excessively lead fouled and unable to fire. The porcelain surrounding the electrode on the top spark plug for number two cylinder was also cracked and indications were that this plug was firing intermittently. No other faults were found that could have contributed to to the power loss.

When the AFTN strip was made up from the aircraft's flight plan, the flight data officer did not designate the route by inserting the correct address code. When another flight data officer came to transmit the AFTN message he noted that the address code was missing. Instead of checking the flight plan he apparently made a best guess using the departure and destination aerodrome information. However, he guessed the wrong route. Consequently, some of the addressees were omitted, including the Townsville arrival's position. When the B747 departed from Brisbane, a departure message was also not sent to the Townsville arrival's controller.

When the aircraft came within the sector seven controller's jurisdiction he omitted to annotate his flight strip with the updated estimated time of arrival for the aircraft. This mistake was not recognised and coordination for the aircraft with the Townsville arrival's controller was not accomplished. The omission was not noted until the Townsville arrival's controller asked about an unknown aircraft entering his airspace from the south-east. The B747 had been maintaining 31,000 ft and there were no conflictions with other aircraft.

The sector was busy at the time with congested traffic in the Mackay/Whitsunday area which included an international aircraft with radio communications problems and unnotified light aircraft traffic asking for clearances at the new airspace boundaries in the Mackay area. While all the controller's attention was focused on this area, the high altitude B747 was overlooked.

The radar and procedural sector seven consoles were manned with a trainer and trainee at each console. The radar controller was undergoing a final check before being rated on the position. The radar console also suffered from some ergonomics deficiencies. The coordination intercom line was not operating, and the hand operated radio transmission switch was unserviceable. The radar operator had to lean across the procedural console to use its coordination line. The training was also a distracting factor because of trainer/student interaction and as a result the controllers had less opportunity to monitor each other.

The procedural console was manned by a trainee and trainer who had not operated together before. The trainee was being shown a method of laying out his flight strips which involved an altitude ranking rather than a chronological order as previously used by the trainee and his full-time instructor. This disrupted his scanning technique.

The complex combination of factors resulted in a lack of coordination between the various air traffic service units responsible for the supervision of the B747 flight.

The pilot reported hearing several unusual noises from the right side of the aircraft while backtracking in preparation for take-off from runway 08. The aircraft then veered sharply to the right and ran off the sealed surface before the pilot was able to bring it to a halt. The right mainwheel had collapsed.

The investigation found that several bolts had came loose from the inner and outer flanges of the three piece wheel assembly. The right main tyre had recently been replaced and only the outer wheel flange was removed to facilitate the tyre change. The torque on the inner flange bolts was not re-checked after the new tyre was fitted.

The pilot attempted to start the engine for the return flight to Darwin, but the starter motor failed to operate. He then applied the handbrake, chocked the nose-wheel and hand-swung the propeller. After several attempts the engine fired then ran at a high RPM speed causing the aircraft to jump over the chock and head towards the airport fence. After unsuccessfully attempting to enter the cabin the pilot tried to grab a main wheel, but missed. He next grabbed at the tailplane but was knocked to the ground. The empty aircraft then ran through the airport fence, across a road and into a ditch, where it came to rest suffering substantial damaged.

The pilot reported that he established the aircraft on final approach at 60 kt, with full flap selected. He assessed that the aircraft was slightly high on the approach, so he lowered the nose of the aircraft and flew it on to the ground to touch down near the runway threshold. The aircraft bounced and the pilot added some power to cushion the second touchdown. The aircraft then bounced again, more severely than the first bounce, so the pilot elected to go around and applied full power. He stated that the aircraft seemed sluggish and stabilised at about 2 m above the strip, so he retracted the flap to 20 degrees. He then became concerned about the high ground beyond the airstrip and flew towards a gap between trees about 30 m to the right of the strip. He banked the aircraft to avoid a fence and, on seeing more trees ahead, levelled the wings and closed the throttle. The left wingtip dug into the sloping terrain and yawed the aircraft. As the aircraft slowed further, the nose gear was torn off and the right gear bent backwards. The pilot could not recall hearing the stall warning horn operate at any stage during the sequence.

Witnesses reported that, while aircraft belonging to the operator involved in the accident always landed towards the north-west, other operators who used the strip landed towards the south-east when wind conditions dictated, by flying a curved approach to avoid the high ground beyond the end of the strip. The witnesses reported the surface wind as being from the south/south-east when the aircraft flew the approach. The aircraft was described as being close to the runway on base leg and to then fly a steep approach compared with other aircraft they had observed land at the strip. They stated that the aircraft touched down some distance into the strip, up to halfway between the end of the strip and the windsock, and to bounce four or five times before attempting to go around from a position past the windsock.

There was no apparent fault with the aircraft which might have contributed to the accident.

Neither the touch-down or attempted go-around positions could be determined accurately. The performance of the aircraft during the attempted go-around, particularly after the partial flap retraction, indicates that the aircraft probably was operating in ground effect.

Three local considerations were identified which could have contributed to the occurrence:

1. the high ground beyond the south-eastern end of the strip could have created an illusion that the aircraft was low and led the pilot to fly a steeper than normal approach path;
2. the actual wind at the approach end of the strip could have been different to that indicated by the windsock which could have been subjected to local effects caused by the high ground adjacent to the strip; and
3. the pilot was aware that he was operating a charter flight and perceived some pressure to land from the approach, rather than fly another circuit.

The factors considered relevant to this accident were:

1. The pilot's relatively low level of flying experience.
2. The position of the windsock was such that it might not have provided a realistic indication of the surface wind at the south-eastern end of the strip.
3. The terrain under the approach path could have contributed to the pilot experiencing an illusion relating to the aircraft’s approach angle.
4. The pilot perceived pressure to land from the approach.
5. The aircraft was probably high and fast on final approach.
6. The aircraft landed downwind.
7. The pilot made a late decision to go around.

When the landing gear was selected down for landing at Bundaberg, the landing gear motor was heard to operate only momentarily before the landing gear circuit breaker popped, and a burning smell entered the cabin. The circuit breaker was reset and again the landing gear motor appeared to function only momentarily before the circuit breaker tripped. The pilot advised that he then tried unsuccessfully to lower the landing gear using the manual emergency extension. A decision was then taken to return to Brisbane where a "gear up" landing was carried out. All five occupants were able to exit the aircraft safely.

VH-DBB was one of four aircraft conducting circuit training on runway 26L. The pilot was advised by air traffic control that her aircraft was number three in sequence and to follow an aircraft on mid-downwind, but she mistook a landing aircraft as the aircraft to follow and turned base in front of the number two aircraft. The air traffic controller noticed the conflict and alerted the aircraft so that avoiding action could be taken. The aircraft passed within 200 feet laterally of each other.

The owner-pilot of VH-DBB had been conducting refresher training and was familiar with the aircraft and Parafield and therefore conducting reasonably tight circuits. The other aircraft were being flown by ab initio students under dual instruction and carrying out wider than normal circuits. This led the pilot of VH-DBB to mistake the aircraft on final as the number two aircraft, which she had lost sight of in the ground clutter, and made her turn onto final approach to follow that aircraft. The correct aircraft was on a long final approach at the time. Both of these aircraft were of the same type and similar colour.

The helicopter arrived at Mildura at about 1745 EST. The pilot was marshalled to a parking spot by the refueller. After the helicopter was refuelled, the pilot offered to reposition the helicopter but the refueller suggested it remain parked where it was, as had been common practice in the past. The pilot complied with the refueller's advice. The helicopter was not parked on a taxiway.

At 2045 EST, while taxiing to the terminal, in very dark conditions, the Cessna 404 pilot followed a yellow taxi guideline with the aircraft's taxy light illuminated. He was momentarily distracted by the windsock lights flashing and inadvertently allowed the aircraft to drift about one metre to the right. As he returned his attention to the yellow line, he suddenly saw the parked helicopter in his peripheral vision. He attempted to stop but the right propeller contacted one rotor blade and the right wing tip impacted the windshield of the unattended helicopter. The collision occurred about 25 metres from the usual parking bay for the Cessna. The Cessna 404 pilot was not expecting a helicopter to be parked where it was.

Sequence of Events

The aircraft was on a night departure from runway 22. Shortly after a normal take-off it struck trees 439 metres from the departure end of the runway. The aircraft then impacted the ground and slid to a stop approximately 360 metres further on. The aircraft was destroyed by a post impact fire. All occupants were able to evacuate the aircraft although they received significant burn injuries.

This summary looks at the factors leading to the accident, some of the events surrounding the post impact evacuation and safety action already taken.

Accident Analysis

The aircraft was in controlled flight, in a shallow descent with wings level, at impact. All aircraft systems were reported to be operating normally at the time.

Evidence indicates that the pilot flew the aircraft to an initial 10 degree nose up attitude after lift-off. However, he then selected the landing gear up, turned the landing light off, introduced forward elevator trim, to relieve an apparent nose up tendency, and attempted to make a radio call. It was during these activities that the aircraft descended and struck the trees.

The pilot, who was a very experienced instructor, met all night flying recency experience requirements at the time of the accident. Evidence indicates that the pilot had commenced work at approximately 0330, on the day prior to the accident. The aircraft arrived at Halls Creek around 1130, and post flight activities were completed by 1200. The pilot then proceeded, by vehicle and in the company of the passengers, to inspect a drill site, arriving back at the motel about 1630.  He attended a barbecue, completed planning for the next day's flying and loaded some of the baggage on to the aircraft before retiring for the night. He had about six hours sleep immediately prior to commencing duty at 0330 on the day of the accident. The pilot had planned to depart after breakfast however he had been asked, by his company, to bring the flight forward as a piece of defective machinery was required in Perth as soon as possible. He calculated an earlier departure time based on his assessment of minimum crew duty times as set out in Civil Aviation Order (CAO) 48.

The outside environment (no ambient lighting, no visible horizon) was such that the pilot, immediately after lift-off, had to make a transition from outside visual references (runway, runway lights etc.) to total reliance on the aircraft flight instruments for the aircraft to be flown safely away from the runway environment.

If an effective cross reference of the instruments was not established immediately, potential existed for the aircraft to descend. To establish an effective cross reference, a pilot must accept that dark night take-offs need to be conducted in accordance with the procedures normally used for a departure in instrument flying conditions. The actions of the pilot in becoming involved in other activities including the attempted transmission of a departure report, so soon after liftoff, indicates that he had not fully accepted the instrument flying conditions criteria. In addition, the application of nose down trim probably started a descent that went unnoticed by the pilot whilst he was distracted by his other activities.

The pilot advised the investigation team that he was not aware of the somatogravic illusion commonly associated with dark night take-offs. The somatogravic illusion, as outlined in Note 1, is a condition that could also have made the pilot's concentration on a safe departure difficult. The possible presence of fatigue would have increased his susceptibility to the illusion.

Aviation safety research indicates that accidents often occur at a low point in a pilot's circadian rhythm (which often occurs in the early hours of the morning) and/or if a pilot's intake of nourishment is inadequate. The pilot reported that he ate normally except that he did not have any breakfast on the morning of the accident. Whilst no direct evidence was available to support a positive conclusion, the flight timings and other activities indicate that both of these factors may have been relevant to the accident sequence.

The known facts concerning the pilot's post lift-off activities, the aircraft's flight profile and the environment to the south-west of the airfield (the prevailing visibility and darkness conditions) suggest that the pilot did not apply the correct departure techniques and that he probably experienced a somatogravic illusion. It is possible that fatigue, a low circadian rhythm and irregular nourishment made him more susceptible to the illusion.

Evacuation Analysis

The investigation determined that the pilot did not conduct a pre-departure safety briefing at either Jandakot, for the flight to Halls Creek, or immediately prior to the departure for the accident flight. Although it was the pilot's normal practice to conduct such a briefing, and it was a company operation's manual requirement to do so, no information was available which would explain the omissions on these occasions. Consequently, none of the passengers were aware of the correct procedure for evacuating the aircraft in the event of an emergency. There were three emergency exits in the cabin area of the aircraft. Two over-wing window exits and a rear door exit. The safety briefing normally covers the use of the emergency exits. In the case of VH-CFG, the release mechanism was such that knowledge on how to operate the emergency exits was essential for a passenger to be able to escape through these openings under conditions of high stress.

One passenger indicated that, after the aircraft had stopped and become engulfed in fire, he observed the sign indicating the location of an emergency window exit but was unable to determine how it operated or to force his way out through it. He had then attempted to escape by moving to the rear of the cabin area only to find that there was no apparent way out. An emergency door exit was located at the rear of the cabin, but he was not aware of this fact. A second passenger also attempted, unsuccessfully, to open an emergency window exit.

All occupants, including the two passengers sitting in the middle row of seats adjacent to the emergency exits, were forced to escape through the normal entrance door located on the forward, right side of the combined cabin/cockpit. This led them through the main fire zone.

Safety Action

As a result of concern over a number of previous accidents with circumstances generally similar to this accident, the Bureau of Air Safety Investigation published a research report, in early 1995, titled Dark Night Take-off Accidents in Australia (SAB/RP/95/01).

Research report SAB/RP/95/01 included the following recommendation:

R940219.

The Bureau of Air Safety Investigation recommends that the Civil Aviation Authority:

1. integrate and expand human performance and limitation considerations into the day VFR syllabus at and above the level of the GFPT for the private pilot licence;
2. review the policy on the testing of human performance and limitation considerations and include this area as an examinable part of the syllabus above the level of the GFPT for the private pilot licence;
3. review theory requirements of the instructor, night VFR and instrument ratings, with emphasis on the specific operational and human-factors considerations that the use of these ratings require as compared to day VFR flight;
4. review the practical requirements of the syllabus leading to an instrument rating, to ensure that a candidate has experienced conditions of IMC and flight (including take-offs and landings) at night, in areas with limited lighting, before being granted a rating;
5. review the policy applicable to the renewal and recency requirement of the night VFR rating, to ensure renewal and recency requirements are similar to other instrument ratings; and
6. educate pilots and operators of the effects of fatigue and the need to establish flight and duty times that are commensurate with the demands of their flight operations. In particular, it should stressed that the limits imposed by CAO 48.0 are maximum limits and lower limits may be appropriate to some types of operations.

Note 1. - The Somatogravic Illusion

'If one considers an aircraft flying straight and level and accelerating along the direction of flight because of an increase in power, for example, then the direction of the inertial force due to the acceleration is to the rear of the aircraft and for the purposes of this discussion can be assumed to be along the longitudinal axis of the aircraft. This inertial force combines with the force of gravity to produce a resultant which is inclined to the rear of the aircraft. If this resultant is then used by the pilot as the vertical reference, then the pilot will incorrectly sense that the aircraft is in a nose-up attitude. If the pilot then trims or eases forward on the control column to correct for this nose-up perception, the nose of the aircraft will drop, and the airspeed will increase. This change in attitude will change the direction of the resultant force vector in such a manner as to maintain and perhaps magnify the illusory perception of a nose-up attitude.

Significant errors in perception can develop within the first few seconds of a change in the force environment. Experiments carried out in flight have shown that there is little lag in the onset of the illusion and that there is a relatively rapid increase in its magnitude during the initial six to eight seconds. This illusion is known as the somatogravic illusion, and it is particularly dangerous when it occurs on take-off or when overshooting, especially at night or in poor visibility. An aircraft deceleration will result in the opposite effect, that is, a perceived nose-down attitude. '

Transportation Safety Board of Canada, Report 89H0007

The pilot was a student with limited experience. He was attempting a flapless approach and landing.

The pilot reported that the airspeed was higher than required as the aircraft crossed the threshold and when he attempted to land the aircraft it touched down nosewheel first. Following a number of bounces the pilot opened the throttle and attempted to go around for another circuit. As he did so the stall warning sounded. The pilot closed the throttle and attempted a final landing. A combination of low airspeed and inexperience caused the pilot to loose directional control and the aircraft turned approximately 30 degrees off runway heading. The aircraft touched down on the grass flight strip, to the left of the runway, and ran across the grass for approximately 80 metres before turning through 180 degrees and stopping.