The crew of a Boeing 737 (B737) received a traffic alerting and collision avoidance system (TCAS) resolution advisory to reduce climb when passing 800-1,000ft on departure from runway 34R at Sydney. This was immediately followed by a TCAS instruction to descend. As the crew commenced descent, the clear of conflict command was heard and a Kawasaki BK 117 helicopter sighted passing to the left in the opposite direction at a distance of about 1 NM.

The investigator reviewed radar data and air traffic control automatic voice recordings to establish the sequence of events. The investigation found that the aerodrome controller East was controlling the B737 and the helicopter was being controlled by the aerodrome controller West. Through a comprehensive coordination process, both controllers were aware of the other controller's traffic.

The helicopter was inbound from the north on a Medical Category 2 flight to Prince Henry Hospital and was cleared by the West controller to track via the runway intersection at 2,000 ft. The West controller had provided this advice to the East controller, who agreed with the clearance and advised that the helicopter was not required on his frequency.

Prior to clearing the B737 for take-off, the East controller alerted the West controller of his intention to launch the B737. The West controller advised that he would keep the helicopter to the west of the B737. The B737 was departing from runway 34 R and was cleared via an ENTRA TWO standard instrument departure (SID). The SID required a right turn after departure to intercept the Sydney 023 VOR radial. Because the aircraft had been cleared for take-off and was on an unrestricted climb, there was the potential for conflict with the helicopter, which was inbound to Sydney from the north at 2000 ft.

Approximately 30 seconds after the B737 had been cleared for take-off, the West controller asked the helicopter pilot to report sighting a 737 on departure roll on runway 34R. The pilot reported sighting the B737. The West controller then advised the helicopter pilot of the B737's intentions and assigned the pilot the responsibility for separation. The pilot acknowledged this.

The TCAS on the B737 activated with a resolution advisory some 20 seconds later and the two aircraft passed each other with 0.9 NM horizontal and 500 ft vertical separation.

Visual separation of air traffic was a valid method to use in these circumstances. However, the criteria for its application were clearly detailed in the Manual of Air Traffic Services (MATS) Part 4 Section 5. In particular, MATS 4.5.1.10 stated: "In circumstances where an aircraft has been instructed to maintain separation from, but not follow, an IFR aircraft, traffic information shall be issued to the IFR aircraft, including advice that responsibility for separation has been assigned to the other aircraft". The B737 was an IFR aircraft but was not provided with the required traffic information.

A Cessna 340 (C340) was maintaining 8,000 ft while tracking direct to Williamtown from a position bearing 020 degrees at 25 NM. A Royal Australian Air Force Macchi was conducting a Tacan approach and intended to perform a missed approach procedure before climbing to 10,000 ft to continue the training sortie. The sector controller had issued the approach controller with a restriction for the Macchi to maintain 7,000 ft on departure to ensure separation with the C340. The approach controller was a trainee being supervised by a rated officer.

While the Macchi was on final approach to runway 30, a further restriction of 2,000 ft was placed on the departure due to other conflicting aircraft. The crew of the Macchi contacted the approach controller on departure and advised that they were maintaining 2,000 ft. The controller took appropriate action to resolve the confliction and then cleared the crew of the Macchi to climb to their planned level of 10,000 ft. The controller had omitted to issue the 7,000 ft restriction even though it was still a requirement to ensure separation with inbound aircraft.

As the Macchi climbed through 8,000 ft, while approximately 6 NM south of Williamtown, it passed within 1 NM of the C340. There was an infringement of separation standards.

The investigation by the Directorate of Flying Safety - Australian Defence Force revealed that the Australian Defence Air Traffic System (ADATS) was being trialled at the time but that the older surveillance radar (SURAD) equipment was actually in use by the approach controller. The SURAD did not have identification labels or height information (facilities that were available on more modern equipment) and that limitation increased the workload on the controller. Additionally, the SURAD was unreliable in its ability to provide constant, accurate position information within 10 NM of Williamtown. The controllers were aware of those restrictions as they were documented in aeronautical publications.

The military sector controller was using the Interim Radar Display System (IRDS). Although that system had labels and a Mode "C" height reading capability, the Macchi was not equipped with a Mode "C" capability. Consequently, the sector controller did not have a radar indication of the height of the Macchi.

The airspace was divided vertically between approach/departures and sector control with ground level to 5,000ft being owned by approach/departures, and sector control the airspace above. The coordination had been adequate but flight progress strip management made the task of remembering an additional restriction more complicated. Consequently, the trainee approach controller forgot to issue the 7,000ft requirement to the crew of the Macchi. The rated officer did not pick up the error until the Macchi was actually passing the level of the C340. The training officer said that he looked at the radar and the strips but was feeling tired and may have missed the information.

The management of the flight progress strip, which was very crowded and difficult to read, was different from that used in other Australian locations. Local procedures were being taught where level restrictions were placed in box 11, whereas in all other air traffic control units the box for such notation was box 4, as specified in the Manual of Air Traffic Services (MATS). That local procedure had been in use at Williamtown for some time, but the investigation was unable to find any documentation specifying such action. In addition, the strip had been divided into four quarters to cater for four separate approaches that the crew of the Macchi had intended to carry out. That action made the writing very small and difficult to read. The alternate method was to use separate flight progress strips for each approach, resulting in more writing space being available for instructions and, therefore, allowing for larger print.

While climbing through 1,500 ft after take-off from Hobart, the crew of the Boeing 737 advised air traffic services that the number one engine had failed and was being shut down. The crew returned the aircraft to Hobart. There was no evidence of fire.

Disassembly and inspection of the CFM56-3C1 engine, serial number 856135, traced the failure to the loss of a 15 x 20 mm segment of trailing edge from a single high-pressure turbine blade. The passage of the segment through the turbine resulted in extensive damage to all four stages of the low-pressure turbine assembly, rendering the engine inoperative. The subject blade was subsequently removed and examined by the Australian Transport Safety Bureau.

The failed blade, serial number 849M8, had accumulated 26,576 hours and 17,928 cycles since new. Maintenance records indicated that the blade received a "full" repair in the manufacturer's facilities in Singapore in June 1997 and was subsequently installed into the subject engine where it accumulated 10,226 hours and 5,332 cycles. The repair involved the use of Rene 80 alloy. The blade also received a "mini tip" repair in July 1995. On that occasion Inconel 625 alloy was used.

The reason for the "full" and "mini tip" repairs was to rebuild the blade tip area eroded due to the high operating temperatures and pressures. Any thermal cracks in the area were checked and removed at that stage. The "full" and the "mini tip" repairs were carried out in approved repair facilities in accordance with the appropriate manufacturer's repair documents. There was no difference between the repairs as far as the extent of the blade weld repair, materials and techniques used. The only difference was in the area of the blade coating.

The subject blade and the high-pressure turbine were periodically inspected in accordance with the manufacturer's and operator's maintenance programs. Those programs required the high-pressure turbine to be inspected at intervals not exceeding 1,140 hours. The inspection required detailed examination of the convex and concave mid-chord tip areas for radial cracks and of the blade trailing edge for axial cracks. The procedure specified the permissible crack length and imposed engine service limitations if cracks exceeded the limits.

The subject blade had accumulated 660 hours since it was last inspected on 11 August 2000. Following that inspection the high-pressure turbine was certified as serviceable.

The operator advised of a similar failure to a high-pressure turbine blade from another CFM56-3 engine. That item showed the loss of a similar section from the blade trailing tip corner but without the catastrophic consequences for the rest of the turbine and the engine.

The engine manufacturer indicated that that type of blade failure was not new. Although not every blade failure was reported by the engine operators, the manufacturer received reports of a number of similar cases each year. The manufacturer indicated that no formal records of the failed blades, their time in service, and repairs, were maintained.