The pilot was conducting a series of scenic charter flights in a Kawasaki KH4 helicopter and had already completed several flights during the morning. The pilot reported that he departed on a 30-minute scenic flight and had been airborne for about 25 minutes when the engine suddenly failed. At the time the of the engine failure the helicopter was flying 500 ft above ground level, about 2 NM north-west of the planned landing area. The pilot immediately lowered the collective control for the main rotor and entered an auto-rotative descent.
During the descent the pilot assessed that the helicopter could not safely reach a clearing to the south-west and manoeuvred to land in lightly timbered terrain. During the landing flare the tail rotor was reported to have struck the branches of a tree and the helicopter tipped forward before landing in a slight nose-down attitude. Damage to the tail rotor blades, main rotor mast, right front landing skid, VHF radio aerial and landing light was reported. The pilot and the two passengers did not report being injured.
The helicopter operator reported there was no mechanical reason for the loss of engine power and that fuel exhaustion may have contributed to the engine's loss of power. Approximately half a litre of AVGAS was recovered from the fuel tanks at the accident site and there was no obvious damage to the fuel system. Standard company policy required the pilot to ensure the helicopter carried enough fuel to complete the planned flight, plus an extra 20-minute fixed fuel reserve.
The pilot was a part-time employee of the helicopter operator and would relieve the full-time pilot, usually one day a week. He had been employed on this basis for approximately three months. The operator reported the recent replacement of the calibrated dip stick used to measure the fuel tank contents had contributed to the circumstances of the fuel exhaustion.
The original dip stick was a hollow calibrated hard-plastic tube and was the dip stick supplied by the helicopter manufacturer for dipping the fuel tanks. The tank contents were measured by inserting the dip stick diagonally into the tank, passing it through a hole in the tank baffle and then placing a finger over a small hole at the top of the dip stick. This would cause fuel to be trapped inside the tube, allowing the dip stick to be removed from the tank and reading of the tank contents against a graduated scale.
To ensure the plastic dip stick was inserted at the correct angle, two metal pins protruded from either side, near the top of the dip stick. These pins would rest on the fuel filler neck and ensured the fuel quantity could be measured consistently. Cracks in the plastic tube had made the dip stick ineffective for measuring the tank contents and the helicopter operator had recently replaced it with a wooden dip stick.
The new wooden dip stick had been calibrated to measure the fuel quantity when inserted almost vertically into the tank, without passing through the hole in the tank baffle. Using this technique had the advantage of restricting the angle at which the dip stick could be inserted into the tank and when correctly applied, would not cause large errors in measuring fuel quantity.
On the day of the accident, the pilot reported that he had used the new wooden dip stick for the first time. He had used the same technique to dip the fuel as he had been instructed to use with the original plastic dip stick. He was not aware that this method for measuring the fuel quantity was only valid when using the original manufacturer's supplied dip stick.
This had resulted in the pilot inserting the wooden dip stick into the tank at an oblique angle, passing through the tank baffle and resting on the tank bottom. This technique could result in a significant over estimation of tank contents. The pilot reported that he had not been advised of the change in method for dipping the tanks using the new dip stick, although he previously had used similar dip sticks on other models of helicopter.
The pilot reported that on the day of the accident, he had first dipped the fuel tanks during the daily inspection conducted prior to the first flight of the day. During this inspection, he detected a discrepancy with the helicopter's fuel log, where the closing figure from the previous days flying did not appear to match the reading he obtained from the tank dip. In trying to resolve this discrepancy, the pilot reviewed other entries in the fuel log and noticed what he believed was another similar discrepancy from the day before. With this discrepancy in mind, he elected to proceed on the basis that his dip of the tank was accurate. The pilot continued to over estimate the quantity of fuel contained in the tanks during subsequent dips of the tank.
The operator subsequently reported the pilot had misread the fuel log and there was no discrepancy. The investigation reviewed the calculations of the aircraft's hourly fuel consumption, which supported the accident pilot's interpretation of the entry in the fuel log for the day before the accident. It was not possible to further address this ambiguity or to determine positively what figure had been entered in the fuel log.
The operator did not have a policy for resolving discrepancies with fuel log entries and relied on the pilot using the dip stick to check the fuel quantity before confirming this reading with a visual check of the tank contents. The operator reported that all pilots were trained to verify dip stick readings using this method and that the tank contents could be seen through the opening of the filler neck. This figure would then be verified against the indications of the fuel gauge and the information contained in the aircraft fuel log. The pilot reported that he was not in the practice of making a visual check of the tank contents and he relied on the reading he obtained from the fuel dipstick as being the quantity contained in the tanks. He could not recall being instructed in the technique of comparing dipstick readings with a visual assessment of the tank contents during his training. The pilot did not detect any discrepancy between the fuel quantity measured using the dip stick and the readings from the cockpit fuel gauge. The investigation was unable to verify the apparent training discrepancy.
The pilot fuelled the helicopter twice on the day of the accident. Before the first flight of the day, 40 litres had been added and another 105 litres was added later that morning. The pilot estimated that when the engine failed, the helicopter had flown about 1-hour 25-minutes since the last refuelling. Based on information supplied to the investigation, it was likely the fuel tanks contained between 30 and 50 litres before the first flight of the day and between 25 and 35 litres before departing on the accident flight.
The pilot reported that he had been monitoring the fuel consumption by crosschecking the fuel gauge indications with readings obtained from the dip stick. However, he did not detect the critically low fuel level before he departed on the accident flight. It was also likely the fuel level was critically low on completion of the first flight of the day.
Contributing to the circumstances of the fuel exhaustion was the ambiguity with the fuel log entry from the previous day's flying. The pilot's relatively low-level of experience on this helicopter type and his employment status as a part-time relieving pilot had possibly contributed to his reliance on a dip stick reading to resolve the discrepancy with the helicopter's refuelling log.