Bundaberg

No fault was found with the aircraft or its systems which might have contributed to the accident. It could not be determined who was manipulating the controls of the aircraft at the time of the accident. Evidence was obtained that it was the habit of the check pilot to have pilots undergoing endorsement or check to fly two circuits landing into wind and then to carry out crosswind landings. The check pilot and the pilot under check had previously operated at the dam and alighted on to both the east/west and the north/south channels. Having been observed to fly two into wind circuits and then head north and not be sighted again, it is possible that the aircraft then commenced crosswind operations onto the north/south arm of the lake, landing in a southerly direction with a crosswind from the left. Information from the Bureau of Meteorology indicated that the surface wind in the area at the time of the accident was 090` magnetic at 15 knots. This information was confirmed by witnesses at the dam wall who observed white caps on the surface of the dam. The north/south channel of the lake was bounded on its east side by steep hills rising to 70 metres above water level. The effect of this high ground was to partially blanket the north/south channel from the easterly wind. The position of the wreckage was in the area where the wind shadow effect would have ended and where the wind would have blown at full strength along the main east/west channel of the lake. The crosswind limitation for the aircraft as stated in the flight manual was 8.7 knots. Commenting in early 1988 on an enquiry regarding the raising of this limit, the aircraft manufacturer emphasised the 8.7 knot limit and advised that any test work to raise the limit should proceed cautiously starting at or below the current (8.7 knot) limit. If the aircraft was conducting crosswind operations in the north/south channel, and suddenly encountered a 15 knot crosswind on exiting the wind shadow area, the control difficulties confronting the pilot could have been significant. The aircraft wreckage was intact except for the floats which had been torn off by water impact forces. The right float was severely damaged while the left was intact. The forward tip of the right float had been severed by the propeller. The remaining forward section had then been forced upwards and outboard and had broken off. This weakened the float support structure, causing it to fail, and allowing the remaining section of the right float to strike the right side of the fuselage just aft of the cabin. Damage of this type an magnitude was most probably caused by the nose of the right float digging into the surface of the lake at relatively high speed. For this to occur, the aircraft was banked to the right at float impact - a possible consequence of encountering a strong crosswind from the left. There was no evidence that the aircraft had hit a submerged object. The factors associated with the development of this accident could not be determined.

The extensive fire damage hampered the investigation of the accident. The surviving passenger believed that the aircraft was on fire before the collision with the tree. No other evidence of an in-flight fire could be obtained, and it was considered possible that the survivor's recall of the accident sequence had been affected by the impact and the fire. Such discrepancies in recall are not uncommon among accident survivors. The elevator trim control jack was found to be in the full nose-down position, but it was not possible to establish whether the trim was in this position prior to impact. Such a pre-impact position could indicate either a runaway electric trim situation or that, in his hurry to depart, the pilot had not correctly set the trim for takeoff. The aircraft was known to have had an intermittent fault in the engine fire warning system. The fault apparently caused the fire warning light to illuminate, and the fire bell to sound, usually just after the aircraft became airborne. The pilot was aware of this fault. It was considered possible that, if the fault occurred on this occasion as the aircraft entered the fog shortly after liftoff, the pilot's attention may have been focussed temporarily on the task of cancelling the warnings. During this time he would not have been monitoring the primary flight attitude indicator, and would have had no external visual references. It was also possible that, if for some reason the pilot was not monitoring his flight instruments as the aircraft entered the fog, he suffered a form of spatial disorientation known as the somatogravic illusion. This illusion has been identified as a major factor in many similar accidents following night takeoffs. As an aircraft accelerates, the combination of the forces of acceleration and gravity induce a sensation that the aircraft is pitching nose-up. The typical reaction of the pilot is to counter this apparent pitch by gently applying forward elevator control, which can result in the aircraft descending into the ground. In this particular case, the pilot would probably have been more susceptible to disorientating effects, because he was suffering from a bronchial or influenzal infection. Although all of the above were possible explanations for the accident, there was insufficient evidence available to form a firm conclusion. The precise cause of the accident remains undetermined.
It is considered that some of the following factors may have been relevant to the development of the accident
1. The pilot was making a hurried DEPARTURE. It is possible that he did not correctly set the elevator trim and/or the engines may not have reached normal operating temperatures before the takeoff was commenced.
2. Shortly after liftoff the aircraft entered a fog bank, which would have deprived the pilot of external visual references.
3. The aircraft had a defective engine fire warning system. Had the system activated it may have distracted the pilot at a critical stage of flight.
4. The aircraft might have suffered an electric elevator trim malfunction, or an internal fire, leading to loss of control of the aircraft.
5. The pilot may have experienced the somatogravic illusion and inadvertently flown the aircraft into the ground. The chances of such an illusion occurring would have been increased because the pilot was evidently suffering from an infection.

Fuel starvation.