FACTUAL INFORMATION

The aircraft was making its last spray run before refuelling when it suffered a loss of engine power. The pilot climbed the aircraft while attempting to restart the engine.

When the engine failed to respond, a forced landing was carried out into a paddock but during the landing roll the aircraft entered a 2-metre-deep ditch at low speed, causing damage to the propeller and left wingtip.

Subsequent inspection revealed that while there was a small amount of fuel in one wing tank, the header tank feeding the engine was empty.

ANALYSIS

When the aircraft type was originally designed, it was fitted with a radial piston engine. A turboprop engine was fitted to later models, and the fuel system was redesigned with a small header tank in the fuselage between the wing tanks and the engine. The header tank should supply fuel to the engine for a short time if the tank outlets become uncovered by fuel during manoeuvring with small amounts of fuel remaining.

If the fuel supply to a turboprop engine is interrupted and then restored, the engine is not likely to restart without action from the pilot unless an automatic re-ignition system is installed. There was no such system on the accident aircraft.

Other similar engine failures have been experienced with this type of aircraft due to the location of the fuel tank outlet, which can become uncovered when the fuel level in the tank is low, and the aircraft is in a nose-down attitude.  This was the most likely situation which led to interruption of the flow of fuel to the header tank and to the engine of VH-ODR.

SIGNIFICANT FACTORS

1. The aircraft was operating with a small amount of fuel in the wing tanks.
2. Although there was some fuel in one wing tank, the header tank feeding fuel to the engine was empty probably because a nose-down attitude allowed the tank outlets to become uncovered.
3. The engine failed, probably because of fuel starvation, at too low a height for the pilot to effect a restart.

SAFETY ACTION

As a result of the investigation into this and a subsequent occurrence (BASI No. 9601185, VH-XST), the Bureau of Air Safety Investigation forwarded the following Safety Advisory Notice to the Civil Aviation Safety Authority (CASA) on 1 July 1996.

SAN960052

The Bureau of Air Safety Investigation suggests that the Civil Aviation Safety Authority, in consultation with the US Federal Aviation Administration (FAA), review the fuel system design of aircraft conforming to Type Certificate A17SW to ensure the adequacy of the fuel system with all applicable airframe/engine combinations.

The CASA response dated 18 July 1996 stated in part: 'I have written to the President of Air Tractor, and the Small

Airplane Directorate of the Federal Aviation Administration, advising them of the fuel starvation incidents in

Australia and asking for their comments.  I will advise you of the responses when I receive them.'

The aircraft was cruising at a height of about 800 feet above the ground when the pilot noticed a reduction in engine power. Application of carburettor heat and changing fuel tank selection made no difference. The loss of power was accompanied by an unusual engine sound. A total loss of engine power followed, and the pilot made a forced landing in a paddock. During the landing roll the right wing stuck a fence post.

Inspection of the aircraft showed the crankcase had a hole in it.

Significant Factors

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

1. Engine failure.

2. Collision with a fence post on the ground roll.

The helicopter had been chartered to transport persons associated with a Fraser Island resort to the area of the wreck of the "Maheno". After picking up a third passenger at the resort the helicopter headed north along the beach, flying passed the wreck, before heading in a southerly direction prior to commencing the approach to land.

The pilot stated that he was making an approach over the water to land on the beach to the north of the wreck when, at about 30 feet above sea level, he felt a "stiffness" in the cyclic control as he attempted to move it forward. Before he was able to access the situation, the helicopter descended into the water. The aircraft remained upright, and all the occupants evacuated and made their way to the shore. The helicopter was subsequently rolled onto its side as a result of wave action.

Inspection of the wreckage and testing of the hydraulic components of the helicopter could find no fault with either the aircraft or the hydraulic system. The reason for the reported control stiffness could not be determined.

The pilot reported that after selecting the landing gear down he failed to obtain a safe down and locked indication. He notified the Tower of the situation, then retracted the landing gear and attempted a manual extension. He disarmed the landing gear circuit breaker, placed the selector switch to the "DOWN" position, disengaged the motor release arm and inserted the gear extension handle into the right socket, and after reducing airspeed, pushed the handle as far forward as it would go. As he was still unable to obtain a down and locked indication, he landed the aircraft, but the landing gear collapsed during the landing roll.

Investigation determined that the landing gear functioned normally during ground testing. The navigation lights were on at the time. This automatically dims the landing gear indication such that it is difficult to see in daylight. The pilot therefore assumed that the gear had not locked in the "DOWN" position after the first extension and he elected to attempt a manual extension. However, the pilot failed to complete the manual extension procedure which requires that after positioning the handle full forward in the right socket, the handle is then inserted into the left socket and the procedure repeated. This failed to lock the landing gear in the "DOWN" position, which subsequently collapsed under the landing loads. The pilot reported that in the fifteen years he had owned the aircraft, he had never performed a practice manual extension of the landing gear in flight.

The pilot was carrying out a fence inspection at 500 feet above ground level when the engine power reduced to idle. He changed the fuel tank selector to the fullest tank but there was no response from the engine. The pilot then closed and opened the throttle lever a number of times. This resulted in an increase to 1000 revolutions per minute, but power again returned to idle when he stopped moving the throttle. As the aircraft was losing altitude the pilot discontinued his trouble shooting activities and attempted a forced landing in heavily timbered terrain. The aircraft struck a number of trees during the landing.

The only fault found during an inspection of the wreckage was a fuel tank vent line containing a wasps nest. Testing of the line indicated that the nest could move causing either a partial or full blockage. The vent outlet was covered with a piece of gauze to prevent this type of occurrence. The fuel tanks were fitted with vented fuel caps.

The reason for the loss of power was not determined.

The crew on VH-EWA were conducting circuit training at Melbourne Airport. On the base leg turn for runway 34 the crew reported sighting another aircraft which had passed above them.

The other aircraft was VH-TXN, which had entered the Melbourne control zone without a clearance. The pilot of VH-TXN was proceeding from Tyabb to Shepparton and the pilot was intending to track around the southern and western boundaries of the zone. No contact was made with the aircraft until the pilot later called on the Radar Advisory Service (RAS) frequency at Rockbank.

Subsequently the aircraft again entered the zone without a clearance near Sunbury. The aircraft was not on the RAS frequency at the time. Subsequently RAS made radio contact, and the pilot was given radar headings to clear the zone near Beveridge.

Later checking of the plan for VH-TXN showed the pilot had made calculation errors in estimating the appropriate headings to fly and also errors in the time intervals. Enroute navigation errors also occurred.

Significant Factors

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

1. Poor flight planning by the pilot of VH-TXN.

2. Poor inflight navigation procedures by the pilot of VH-TXN.

While conducting solo circuit training at Broken Hill the student pilot made a bounced landing, then failed to apply sufficient back pressure on the control column as the aircraft touched down again.

The nose landing gear partially collapsed allowing the propeller to contact the runway. The pilot was not injured and later received additional dual circuit training from a club instructor.

The glider was being winch launched. On the initial climb at a height of about 300 feet the winch cable failed. The pilot made a modified circuit, with the base leg halfway along the strip. Touchdown was well into the strip. The pilot realised the glider would hit the fence at the end of the strip, so he attempted to lift it over the fence. The glider came down on top of the fence, catching the left wing. It then swung left, and the right wing hit a tree.

Significant Factors

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

1. Winch cable failure.

2. Insufficient altitude after the winch cable failure to ensure a successful circuit and landing.

Enroute from Curtain Springs to Alice Springs, while flying low and slow to observe cattle, the pilot heard a loud noise followed by an immediate increase in engine RPM and decrease in rotor RPM. An autorotational approach and landing was carried out into the rough hilly terrain, but the helicopter landed heavily and rolled over.

A subsequent investigation revealed that the clutch assembly upper bearing had failed allowing the engine drive to disconnect from the rotor system.

The bearing was a commercial bearing, not a genuine Robinson bearing, although the manufacturer's instructions clearly state that only genuine bearings, having the correct internal clearances, must be installed. A specialist report indicated that the failure was due to brinelling of the bearing races, possibly caused by the bearing being dropped or damaged during assembly of the clutch. There was also evidence that the bearing had not been sufficiently lubricated which would have decreased its time to failure.

Although the accident was survivable, severe spinal and lower back injuries were suffered by both occupants. It was reported that the space below the seats contained a number of hard items including a hand fuel pump. The flight manual for this type of helicopter, and placards placed near the under-seat compartments caution against placing hard objects in them as they could cause injury to occupants if the seats crush while absorbing energy during a heavy landing.

SAFETY ACTION

As a result of the investigation, the Civil Aviation Authority issued Airworthiness Directive AD/R-22/39, which requires an inspection of the upper clutch actuator bearing of all R22 helicopters to ensure only approved parts are fitted.

The aircraft taxied for departure to Sydney about 90 minutes behind schedule due to fog at Wagga. Although the First Officer was the handling pilot for the flight, the Captain taxied the aircraft as it was not fitted with dual nose wheel steering controls.

During taxi, the First Officer read the pre-take-off checklist. At the flight and navigation instruments check, he incorrectly read the outbound track from the flight plan as 118, which was the distance in nautical miles from Wagga to the TAPIO reporting point enroute to Sydney. The correct track was 053 degrees magnetic and as neither pilot had recognised the error, they set 118 on both HSI's.

After issuing an airways clearance for the aircraft to proceed to Sydney via Bindook at 9,000 ft, the Wagga tower controller cleared the aircraft to take-off on runway 05 and to make a right turn.

The First Officer assumed the handling pilot duties when the Captain handed over control of the aircraft during the take-off roll. The tower controller, who was monitoring the departure, said he lost sight of the aircraft due to low cloud while it was maintaining runway heading. Following a request from the tower controller the Captain advised that the cloud base was 1,000 ft above terrain. When the aircraft was established on the 118 radial of the Wagga VOR, the Captain passed the departure message. He advised the tower controller that the aircraft was tracking 053, which he read correctly from the flight plan, and was climbing to 9,000 ft.

At cruising altitude the aircraft was above two layers of cloud, through which the crew had occasional glimpses of the ground. About 10 minutes after departure the Captain decided to climb to FL 110 due to moderate turbulence, and to avoid cloud build-ups ahead. ATC requested the crew to squawk code 1000 with ident prior to issuing a clearance for the change of level. The aircraft was then located 70 NM to the right of the planned track, to the south west of Canberra. This error was subsequently recognised by the Captain who advised ATC. The aircraft was issued with a clearance to track via Canberra and Bindook to Sydney. There was no confliction with other IFR traffic.

Significant Factors

• The First Officer misread tracking information from the flight plan during the pre-take-off checks.
• Neither pilot verified the orientation of the selected track by reference to the appropriate map or chart.
• Weather conditions prevented the tower controller from adequately monitoring the aircraft's outbound track.