VH-EAN was en route from Sydney to Perth, cruising at flight level (FL)350. N99SC was en route from Essendon to Sydney cruising at FL410 and requiring descent. Both aircraft were under the control of the Melbourne Air Traffic Control Centre, Sector 7. It was apparent that the two aircraft would be in lateral conflict near Yass. The controller cleared N99SC to descend to FL360.

Above FL290, for aircraft that are not laterally separated, the minimum vertical separation is 2000 feet. In this instance the controller had only provided 1000 feet. The two aircraft passed with about three and one half miles lateral separation and 1000 feet vertical separation. For aircraft not vertically separated, the minimum lateral separation in this instance was five miles.

The controller realised his error but too late to rectify the situation. Investigation revealed that prior to the incident the controller had been busy with aircraft inbound to Sydney, entering the holding pattern at Bindook. All levels assigned to other aircraft had been in the range of FL210 to FL260 where only 1000 feet vertical separation is required. He believed it was this that had led him to inadvertently apply the same standard to VH-EAN and N99SC

On commencement of the take-off, power was increased normally and at approximately 60 kts a loud mechanical noise was heard from the right engine with the right torque meter showing large deviations.

The take-off was rejected. Smoke was then observed in the cockpit and the cabin rapidly filled with dense smoke. The aircraft was stopped on the taxiway and emergency checklist actions and evacuations procedures were carried out.  All crew and passengers successfully evacuated the aircraft in a timely manner.

Examination of the engine found that the primary reason for the engine failure was the failure of the number three bearing which had allowed the low-pressure impeller to move forward relative to its housing, causing secondary damage and the subsequent engine failure.

The bearing which had been fitted to the engine since manufacture had been removed, checked and reinstalled during the engine turbine modification upgrade about 530 hours before the incident.  Subsequently during the post-incident inspection one of the bearing balls was found to be approximately 1 mm smaller than the remaining balls.  The presence of the smaller ball resulted in alternating tensile stress in the bearing cage, resulting in fatigue fracture propagation through the cage ligaments at the smallest ball.

The quantitative fractography of the cage fracture surface concluded that the fatigue crack appeared to have initiated near or shortly after the time of the engine turbine modification upgrade.

Significant factors

The following factors are considered relevant to the development of this incident.

1. One ball in the number three bearing cage was 1 mm smaller than the remaining balls.
2. The presence of the smaller ball resulted in alternating tensile stress in the bearing cage, causing fatigue fracture propagation through the cage ligaments at the smallest ball.
3. The failure of the number three bearing allowed the low-pressure impeller to move forward relative to its housing causing secondary damage.

The joy-flight was proceeding northwards over the beach of Surfers Paradise and had been cruising at 500 feet for just over a minute when the engine power reduced to idle. After conducting his trouble checks and finding no way to restore power the pilot selected an area for a landing on the beach. The aircraft came to rest in shallow water after manoeuvring to avoid people on the beach.

On inspecting the engine area, the pilot found that a ball joint in a push-pull rod to the carburettor throttle valve had failed, allowing the throttle to close. Further examination by the maintenance organisation revealed that a plate to which the rod was attached was shaped incorrectly. This caused the spring-loaded ball joint to wear abnormally until it fell off. The engine had been overhauled 200 hours earlier.

The pilot was conducting his first solo navigation exercise and had arrived overhead the destination aerodrome. He attempted to identify the wind direction and speed by observation of the windsock. The indications from the windsock were confusing and the pilot continued to observe it as he entered downwind and extended the leg. The resulting landing approach was long and flat, requiring significant power to reach runway 04. As he flared to land and reduced power the landing gear warning horn sounded. He applied power but the aircraft settled onto the runway with the landing gear retracted.

It was later found that the windsock was wrapped around its pole.

While cruising at 3000 ft, the engine suddenly lost power to the extent that level flight was no longer possible. The pilot conducted a forced landing on to a rough field, causing damage to the landing gear, fuselage, and propeller. Investigation revealed a failure in the rocker gear of the number four exhaust valve.

While enroute from Batchelor to Tindal, the pilot encountered deteriorating weather conditions and landed the aircraft in a suitable paddock, on Bonrook Station. The following day he inspected the landing area and judged the length to be 500 metres and that the area was suitable for take-off. However, the weather conditions were still unsuitable for departure.

The next day, after obtaining a weather forecast and being satisfied that the weather was suitable, the pilot decided to continue the flight. He and his passenger boarded the aircraft, and the pilot configured the aircraft for a short-field take-off. During the subsequent take-off the pilot rotated the aircraft at 55 kts. It became airborne and commenced to climb clear of the 10-metre-high trees at the end of the strip. However, the pilot became concerned about a tall tree directly ahead and commenced a left turn to avoid the tree. The stall warning sounded, one wing dropped, and the aircraft impacted the trees. The pilot suffered minor lacerations to his head, probably as a result of being struck by sections of the broken windscreen, and the passenger received minor bruising.

Checks of the fuel did not reveal any contamination, and the pilot believes that the engine was operating normally during the take-off.

An inspection of the take-off area revealed that the surface was firm and that the grass had recently been slashed.

The performance charts contained in the flight manual show that the strip distance available was adequate for a take-off from a dry grass runway. However, on this occasion, it is likely that the surface was wet and covered in recently slashed grass and that take-off performance was degraded. The pilot was not familiar with operation from such strips.

The student pilot, returning from a solo training session, carried out a normal circuit and approach for landing, but the aircraft bounced slightly on touchdown. The pilot applied some forward elevator control, but the aircraft veered about 90 degrees to the left and ran off the runway.

The pilot unsuccessfully attempted to control the swing with right rudder, then applied full power to go-around. She then observed another aircraft taxiing and turned the aircraft left to avoid that aircraft. The left wing struck the ground and the aircraft cartwheeled.

The aircraft were operating on the same air route within controlled airspace under the jurisdiction of Jakarta control sector. Although Jakarta was providing control and separation services within the Jakarta Flight Information Region (FIR), Perth Flight Information Service International (FIS INT) was monitoring the common international high frequency (HF) aeromobile facilities being used by both aircraft and Jakarta.

G-BPFV was southbound at FL370 estimating LAMOB at 1914 hours and had established HF Primary Guard with Perth FIS INT.

VH-EAM was northbound, 10 NM right of track, at FL350 until passing waypoint LAMOB at 1858 when a flight planned climb to FL390 was requested from Jakarta.

FIS INT 1 intercepted Jakarta's approval for VH-EAM to climb to FL390 and recognised the potential for a traffic confliction with G-BPFV. FIS INT 1 immediately advised Perth control Sector 1 of the situation and provided traffic information direct to VH-EAM.

Almost simultaneously, Jakarta instructed VH-EAM to maintain FL350, but the aircraft had already climbed to FL365 prior to resuming FL350.

The aircraft had established the required 2,000 feet vertical separation again by approximately 1904, two minutes prior to the point of closest approach, which was approximately 60 NM north of LAMOB, within the Jakarta FIR. It was later established that G-BPFV had been authorised by Jakarta to track 10 to 15 NM west of track.

Neither the amended track for G-BPFV nor the change of level for VH-EAM were co-ordinated with Perth Sector or FIS INT for relay. There was a breakdown of the required separation standards.

The pilot reported that he made a normal approach and landing to the wet runway, touching down at around 60 knots. During the landing roll the aircraft began to veer to the right. Attempts to stop the yaw using both rudder and brake were unsuccessful. The aircraft left the sealed runway and the nosewheel was damaged when it entered a soft area.

The wind was blowing at 30 knots with gusts of up to 15 knots and a crosswind of 8 knots from the right.

The precise reason for the loss of control could not be determined but it is possible that it was the result of the combination of the wet runway and the strong and gusting crosswind.

VH-AJJ was climbing to Flight Level (FL) 370 (37,000 feet) whilst VH-WGJ was climbing to FL350, and the aircraft were under the procedural (non-radar) control of the Perth Sector 1 air traffic service controller. The aircraft were on reciprocal tracks and a vertical separation of 2000 feet was required at least 10 minutes before the aircraft passed. The climb profiles of the aircraft meant that the required separation would not be achieved by 10 minutes prior to passing. The controller did not recognise the potential confliction.

Nine minutes prior to the aircraft passing, VH-AJJ requested an amended level of FL350 but was given FL330 because the controller assumed VH-WGJ was already at FL350.

Seven minutes prior to passing, the controller ascertained that VH-WGJ was passing FL320. He then realised a conflict existed and instructed VH-WGJ to descend to FL310. VH-WGJ started its descent back to FL310 as VH-AJJ reached FL330 on the climb. The aircraft reached the standard vertical separation requirements approximately three minutes prior to passing.

Although the controller was an experienced procedural sector controller he had not been involved in this type of control for about 12 months and had only recently been re-categorised.