Loss of separation

Simultaneous runway operations were in progress at Sydney when VH-MJH was cleared to land on runway 34, with a requirement to hold short of runway 07. After the aircraft had landed the aerodrome controller (ADC) instructed the pilot to take the first available taxiway to the right. This was taxiway 'Hotel', located about 238m from the runway 07 simops lights. VH-RMN, a Boeing 727 which was lined up on runway 07 was then cleared for take-off. VH-MJH was then observed to have passed the taxiway and was told to hold short of runway 07. The aircraft continued rolling and when the ADC became uncertain that the aircraft was going to stop before runway 07, he instructed the aircraft to stop.

The aircraft did not stop, and the ADC then cancelled the take-off clearance for VH-RMN which was now rolling for take-off. Another aircraft was sent around from a 2-mile final approach for runway 07. VH-MJH came to a stop mid-way between the stop lights (holding point) and the edge of runway 07. After VH-RMN had stopped VH-MJH was cleared to cross runway 07 and vacated runway 34 on taxiway Bravo 6. The first officer was the pilot flying VH-MJH until after the aircraft had touched down. The captain then took the controls during the landing roll, and the first officer turned his attention to the after landing checks. The captain stated that he had not heard the transmissions to stop.

It was not until he noticed that the first officer was transmitting that he realised that he was not monitoring the tower frequency. After he had received a landing clearance, he had intended to deselect the company frequency and had inadvertently deselected the tower frequency. The first officer had subsequently acknowledged the transmissions from the tower and was not aware that the captain was not monitoring the tower transmissions. The first officer realised that the captain was not following the towers instructions when the aircraft passed the taxiway and he instructed the captain to stop using both verbal and visual signals. By this time the aircraft was nearing the runway intersection.

The captain misidentified the position of taxiway Hotel due to the position of a large mound of earth which had previously been located south of runway 07, but which was now located north of runway 07. This gave him a false impression as to the location of the taxiway and was not aware that he was approaching the runway until after it was too late to stop before the holding point lights.

Significant factors

1. The captain inadvertently deselected the tower frequency just prior to taking control from the first officer.

2. After handing control of the aircraft to the captain, the first officer diverted his attention to the after landing checks.

3. There was a breakdown in crew co-ordination procedures.

Qantas 10 was one of four south-east bound aircraft estimating overhead Curtin reporting point at approximately 0000 WST. Lauder Air 2 was north-west bound and estimating Curtin at approximately 0100 WST. Several other aircraft were also due to enter the airspace around Curtin during this period. Lauder Air 2 was maintaining FL350 (35,000 feet). Qantas 10 was maintaining FL310 and had requested FL370. The duty Air Traffic Controller (ATC) cleared Qantas 10 to climb to FL330, at Curtin, with an expectation that a clearance to FL370 would be available later. The complexity of the traffic situation required the duty ATC to co-ordinate traffic with Melbourne Sector and to make a number of calculations associated with the "time of passing" between Qantas 10 and other aircraft.

During this process the ATC, incorrectly, wrote the "time of passing" between Qantas 10 and another aircraft on the Lauder Air 2 flight strip. The correct "time of passing" between Qantas 10 and Lauder Air 2 was 0029 but the duty ATC wrote 0047 on the Lauder Air flight strip. A "time of passing" of 0047 was also entered on Qantas 10's flight strip. Following advice from Melbourne, at 0026, that Qantas 10 would be accepted at FL370 the duty ATC checked the "time of passing" written on the flight strip (0047) to ensure that sufficient time for the climb was available, (ten minutes for the climb and ten minutes for the buffer), and cleared Qantas 10 to climb to FL370. As Qantas 10 commenced climbing the crew observed the lights of Lauder Air 2 directly ahead. Qantas 10 stopped its climb at FL340 and reported the situation to the duty ATC who then instructed Qantas 10 to descend again to FL330. Lauder Air 2 passed directly overhead and 1500 feet above Qantas 10.

The investigation determined that the duty ATC, who was assessed as very experienced and competent, had also made other procedural errors during his time at the console. Evidence was available which indicated that the errors were probably the result of accumulated fatigue brought on by a lack of sleep and disrupted sleep patterns, a lack of adequate rest breaks over both the long and the short term and pressure caused by the additional tasks associated with his position as a Team Leader.

The crew of VH-NSD received an amended clearance on the flight from Townsville to Mackay, which required tracking over Proserpine. They experienced difficulties receiving the Proserpine VOR (VHF OMNI Range), in that the aircraft indications on both receivers were not indicating a steady reading. The pilots averaged the indications, but when they transferred to the Mackay VOR, they found that their aircraft was on the 300 degree radial and not the 311 degree radial as cleared by air traffic control.

The aircraft was then cleared to continue inbound on the 300 degree radial. Another aircraft was outbound on the 285 degree radial at the time and a breakdown in separation had occurred. The aircraft were estimated to have come within 8nm (15km) of each other.

The crew of VH-NSD did not report their inability to navigate accurately using the Proserpine VOR. This would have allowed air traffic control to use another means of separation other than confining the aircraft to radials. Mackay Air Traffic Control reported that the Proserpine VOR has not been reported as unserviceable by any other aircraft prior to or after this occurrence.

Circumstances

The Arrivals North Controller [ARRN] had commenced duty at approximately 0705 EST and was experiencing a period of heavy air traffic. He had requested the previous occupant of the position to remain at the console in the monitor position to assist him in a settling-in period. This controller was in the monitor position during the events of this occurrence.

The FLOW controller had decided that he would request the use of Sector 8 airspace for three aircraft to reduce the workload on the arrivals controller and reduce the delay to the aircraft. The Sector 8 controller agreed to the request and raised the required flight progress strips [FPS] for his own reference. After receiving instructions from the FLOW controller to track both VH-TQO and VH-TQQ via Richmond, ARRN proceeded to process the two aircraft onto that track, a track that would take them clear of the main arrival route of Singleton to Sydney.

VH-TQO was proceeding at Flight Level [FL]190 on a flight from Tamworth to Sydney. At 0718 ARRN instructed VH-TQO to track direct to Richmond and thence to Sydney. This resulted in the aircraft tracking approximately 210 deg. VH-TQQ was proceeding at FL160 on a flight from Port Macquarie to Sydney and was on a similar track to that of VH-TQO and approximately 12NM behind. At 0721 ARRN instructed VH-TQQ to track direct to Richmond and thence Sydney.

This resulted in the aircraft tracking approximately 200 deg. At 0722 ARRN handed to Sector 8 the identification on VH-TQO with the assigned altitude of 8,000ft. This was accepted by Sector 8 who wrote this altitude on the FPS for VH-TQO. Both aircraft were then descended to 8,000ft by ARRN and at 0724 the controller, intending to transfer VH-TQO to Richmond Approach, mistakenly instructed VH-TQQ to contact Richmond Approach [Sector 8]. VH-TAF was on descent to FL200 on a flight from Coolangatta to Sydney and was tracking on the main arrival route from Singleton to Sydney.

At 0726 ARRN instructed VH-TAF to descend to 8,000ft. When VH-TQQ contacted Sector 8 the controller acknowledged with the callsign TQQ but notated the details on the FPS for VH-TQO. At 0726 the Sector 8 controller had a departure from Richmond that was conflicting with VH-TQO and elected to vector VH-TQO in order to maintain separation. He issued VH-TQQ, the aircraft on his frequency, with an instruction to turn left heading 140 for separation. The crew obeyed the instruction but were approximately 12 NM north of the point the controller thought the aircraft was passing.

The controller saw that VH-TQO was not turning and issued a further left turn to a heading of 120 to VH-TQQ. This placed VH-TQQ on a track that would conflict with VH-TAF. ARRN did not immediately realise that VH-TQQ had turned back towards the main Singleton to Sydney track and it was the arrivals controller in monitor who first saw the new position of VH-TQQ and initiated recovery action. At about the same time the crews of both VH-TQO and VH-TQQ started to question the control instructions and asked if there had been a transposition of the callsigns.

At 0728 ARRN instructed VH-TAF to turn left immediately onto a heading of 090 deg. to place that aircraft on a heading away from VH-TQQ. Flight levels were then checked and vertical separation established until radar separation could again be guaranteed. Radar analysis showed that there was no breakdown in separation as vertical separation existed at all times during which horizontal separation was not provided.

The callsigns were very similar and belonged to the same type of aircraft in the same company. Anecdotal evidence and statements from the persons concerned in this occurrence, indicate that this confusion has often arisen because of the common feature of these two aircraft being processed for arrival or departure at similar times. During the course of the interviews five controllers complained of the two callsigns being on frequency together and having had trouble with them at some time or other.

The crew of VH-TQQ were expecting to be given the radio frequency change to Richmond Approach at the time the ARRN controller issued the instruction and, therefore, had no reason to query the instruction. The O and Q do not readily stand out from one another on the digital read out of the radar display and this can lead to a confusion of callsigns in high density traffic situations.

Significant Factors

1. The callsigns TQO and TQQ are very similar and do not stand out from each other on the digital read out showing on the radar screen.

2. VH-TQO and VH-TQQ were given the same tracking instructions while proceeding 12 NM in trail.

3. Both the Arrivals North controller and the Sector 8 controller used the callsign TQQ when intending their transmissions to be for TQO.

Safety Action

Since the above incident a further occurrence has been reported to the Bureau. On 23 March 1994 a controller at Sydney transposed the callsigns and issued a clearance to TQO which was intended for TQQ. On this occasion there was no further incident as the pilot queried the callsign immediately and corrective action was taken.

Recommendation

R930316

With the above occurrences in mind, the Bureau of Air Safety Investigation recommends that Qantas Airways: Change the registration of VH-TQQ to avoid confusion with other callsigns.

Envoy 606 (Falcon 900) was arriving at Canberra from Sydney and was assigned 7,000 ft on descent on the Wollongong - Canberra track. The pilot had been given a requirement to reach 7,000 ft by 20 DME Canberra due to conflicting traffic and reported at 7,000 ft when 25 NM from Canberra. VH-IEL (Cessna 172) had departed Goulburn for Canberra on a VFR training flight with the student performing all the functions and the instructor engaged in an examining role. The aircraft was outside controlled airspace at 6,500 ft and was tracking via Lake Bathurst.

The control area (CTA) step passes through the north eastern corner of Lake Bathurst and is 30 NM from Canberra. However, the reporting point is an island in the centre of the lake, and this is 28 NM from Canberra. The student was approaching Lake Bathurst and prepared to contact Canberra Approach to request an airways clearance but for some reason delayed the call until at the CTA boundary.

The instructor was allowing the student to conduct his own operation as a test was in progress and although mindful of the requirement to remain outside controlled airspace until obtaining a clearance, he elected to let the situation continue up to the last instant. The instructor was about to instruct the student to make an orbit when the student made contact with air traffic control (ATC). At the time of the call by VH-IEL, ATC observed the aircraft to be 28 NM and Envoy 606 25 NM and therefore considered that a breakdown in separation had occurred.

By the time altitudes were checked and VH-IEL positively identified, the aircraft were 5 NM apart and a radar separation standard existed. The tracking tolerances were 4 NM for VFR flight and 1.2 NM for the radar at that distance. The required separation standard was 5 NM by radar or 1,000 ft vertically. To remain in controlled airspace the pilot of Envoy 606 would need to be 9,000 ft at 30 NM. The descent profile of Envoy 606 was steeper than normal due to the requirement to reach 7,000 ft by 20 NM.

Significant Factors

1. The student pilot allowed his aircraft to proceed to the CTA boundary before initiating an airways clearance request to ATC.

2. The instructor on VH-IEL allowed his aircraft to penetrate CTA prior to obtaining an airways clearance.

Pirate section [2 X FA18] departed Williamtown for Townsville on climb to Flight Level [FL] 310 at 1112 hours. The section was on a track direct to the reporting point GWYDIR which is the boundary position between Sydney and Brisbane air traffic control [ATC] areas of responsibility. This departure time was relayed to Brisbane sector control by Sydney sector via the intercom line and the Brisbane flight progress strips were activated. The GWYDIR strip at both AACCs indicated the same estimate of 1144 hours.

VH-OGD [Boeing 767] was proceeding at FL 370 on a regular public transport flight from Cairns to Sydney and was tracking via GWYDIR on the reciprocal track to Pirate section.

Sydney sector provided Brisbane sector with a radar identification of Pirate section when those aircraft were approximately 50 NM from the boundary. The leader of Pirate section then requested climb to FL 350 and, after all the appropriate coordination was carried out, this was approved by ATC.

Pirate section realised that they were making time against their flight plan and attempted to pass an amended estimate for GWYDIR to ATC. VHF communications between the aircraft and ATC were poor and, despite two or three attempts, this message was not received by ATC.

During this period the Brisbane sector controllers [procedural and radar] realised that the type of formation being used by Pirate section may involve an infringement of the 2,000 ft vertical separation standard required for the passing of the section with VH-OGD. Because of the poor communications with Pirate section, it required several transmissions to ascertain the type of formation being used and, as no definite answer was received from the crew, it was assumed that the aircraft were in Standard Formation. This meant that one aircraft could be up to 500 ft above the other and would therefore infringe the separation standard in use. Brisbane ATC decided to initiate a cruise level change to guarantee vertical separation and instructed Pirate section to descend to FL 340 with a requirement to reach that level by time 1140 hours. This instruction was acknowledged by Pirate section and complied with.

Brisbane sector had calculated the time of 1140 hours based on a time of passing which was calculated on the respective estimates at GWYDIR for both Pirate section and VH-OGD. Although GWYDIR was within radar coverage, the track that the aircraft were using left radar coverage approximately 30 NM north west of GWYDIR and therefore a procedural standard was required.

The Brisbane sector radar controller was monitoring the progress of Pirate section following the radar hand off from Sydney but did not inform the procedural controller that the section was significantly early in relation to their estimate for GWYDIR.

When Pirate section reported at GWYDIR at 1139 hours the Brisbane sector procedural controller realised that they were five minutes early and recalculated the time of passing on which the level change requirement was based. This resulted in the correct time for the Pirate section to reach FL 340 being 1137 hours and as they had not reported at that level until 1139 hours [their actual position at GWYDIR] a breakdown in separation may have occurred.

Pirate section were on radar at the time and VH-OGD appeared on the radar screen at about the same time approximately 30 NM north west of the position of Pirate section. This distance did constitute a radar standard, but this could not be guaranteed until VH-OGD was identified by the radar controller [a situation that occurred after Pirate section had passed GWYDIR].

The crew of Pirate section stated that they had actually been maintaining FL 340 prior to GWYDIR and, although they cannot remember exactly how long before, it is probable that a breakdown in separation did not occur.

Significant Factors

1. The poor VHF communications between ATC and Pirate section inhibited the passing of significant operational information.

2. The communication between the Brisbane sector radar and procedural controllers was ineffective.

Factual information

VH-TAJ and VH-OJL were tracking via crossing airways T41 and W46 respectively, VH-TAJ maintaining FL350 and VH-OJL initially maintaining FL330. The Sydney air traffic control sector responsible for the airspace in which the aircraft were flying was being operated by a rated officer under familiarisation following a period of absence. This officer was under the supervision of a suitably rated and current training officer. At 1557, the crew of VH-OJL reported passing PARKES with an estimate for ALBEE of 1640 and requested further climb to FL 350. This placed the aircraft in possible conflict with VH-TAJ where the two routes crossed, and the controller initially refused the request until positive separation with VH-TAJ could be assured.

The controller then consulted a lateral separation diagram for the confliction point and decided that VH-TAJ would need to exit the area of conflict before VH-OJL could be given climb. In performing this task the controller misread the reporting point name and the distance specified in the diagram. He instructed VH-TAJ to report 41 NM from reporting point T41B instead of 72 NM from T41A. As a result, the aircraft would still be within the area of conflict with VH-OJL at that point. At 1621 the crew of VH-TAJ reported 41 NM from T41B and the controller, believing that he had established that aircraft clear of the area of conflict, immediately cleared VH-OJL to climb to FL350. As the controllers assessed the current clearances of aircraft under their control, some doubt arose as to the relative positions of VH-TAJ and VH-OJL and distance checks were implemented. At 1622 VH-OJL reported 166 NM west of PARKES, placing the aircraft 21 NM inside the area of conflict and not yet at the crossing point. At 1623 VH-TAJ reported 51 NM from T41B, placing the aircraft 59 NM prior to the exit point, inside the area of conflict and not yet at the crossing point. Calculations placed VH-TAJ at the crossing point at 1624, approximately two minutes ahead of VH-OJL. The point of closest contact was at about 1626 when the aircraft were approximately 10 NM apart. The required standard is 2,000 ft vertically, or one aircraft must clear the area of conflict before the other enters. In this case the area of conflict was 102 NM long on the track of VH-TAJ and 103 NM on the track of VH-OJL. As the vertical separation was less than 2,000 ft and the aircraft were still within the area of conflict, a breakdown in separation occurred.

Neither crew was informed of the situation by air traffic control, no traffic advice was given and no corrective action was implemented by the controllers. VH-TAJ was instructed to report at 72 NM from T41B, a position the controller believed was required to establish that aircraft clear of the area of conflict. In fact this was still within an area of conflict as the controller was still referring to T41B when the position required was in relation to the reporting point T41A. VH-TAJ reported at 72 NM from T41B at 1626 and no further action was taken by air traffic control as they incorrectly believed VH-TAJ to be clear of the area of conflict.

Analysis

Having decided on the Lateral Separation diagram to use, the controller then misread the figures, transposing the 41 in the reporting point name [T41A] to the distance required and the letter A to a B. This placed VH-TAJ at a point prior to the intersection of the routes, not at the area of conflict exit point as required. Even if the controller had correctly read the 72 NM reference, the error in using B in lieu of A would have still placed VH-TAJ in the area of conflict when he gave VH-OJL climb to FL350. After issuing climb instructions the controllers realised that an error may have been made and checked the position of the two aircraft. When this check revealed a confliction and breakdown in separation, no remedial action was taken. Three minutes elapsed before VH-TAJ reported at 72 NM from T41B but there was no advice to either crew of the confliction or the breakdown in separation. Three minutes should have been adequate time for all these actions to have been instigated. The Lateral Separation Diagrams provided at the console were hand drawn and had many numerals, some of which referred to distance, others to reporting points. This could lead to a situation where a controller may become confused under certain workload conditions. It was also apparent that someone had written 135 T41B on the diagram under the official position of 72 T41A. This is the same point in space but measured from the other reporting point and anecdotal evidence suggests that it is used as much as the official entry/exit point. There is no reason why this position cannot be used to calculate the area of conflict, but it is not on the official diagram.

Significant Factor

The controller misread the lateral separation diagram when determining the area of configuration for the two aircraft.

VH-KTF [Beech76] was proceeding on an instrument flight rules [IFR] training flight from Goulburn to Canberra and was being radar vectored by Canberra Approach control [APP] for a right downwind leg for runway 35. The pilot was maintaining 6,000 feet being the last assigned altitude.

VH-CZP [Boeing 737] was proceeding on a regular public transport flight from Melbourne to Canberra and was being radar vectored by APP for a left base to runway 35. The pilot was also assigned 6,000 feet by Air Traffic Control [ATC].

APP was being performed by a trainee controller who had completed four weeks of a proposed twelve week training period. This controller was supervised by a suitably qualified training officer.

At 1406 hours VH-KTF was given a left turn onto a heading of 320 degrees with the intention by ATC of using this turn to provide 5NM separation with VH-CZP who was to be given a left turn onto the localiser ahead of VH-KTF.

As VH-CZP was approaching the centre line the training officer realised that the trainee was leaving the issuing of the turn instruction a little late and decided to instruct the trainee to turn VH-CZP immediately. At 1408 hours the trainee issued the turn instruction using the callsign CZF. The captain of VH-CZP then queried this instruction and with this further delay the ATC training officer decided to take over and issued instructions to both VH-CZP and VH-KTF which placed the two aircraft on diverging headings. During these manoeuvres the horizontal distance between the aircraft reduced to 2.5NM with no vertical differential.

The weather conditions were such that both crews were in cloud throughout the occurrence and no pilot saw another aircraft.

The required separation standard was either 5NM horizontally or 1,000 feet vertically.

Significant Factors

1. The trainee approach controller allowed VH-CZP to get too close to the localiser before implementing a turn onto final.

2. The trainee approach controller used the wrong callsign when issuing VH-CZP with a turn instruction.

3. The training officer allowed the traffic situation to develop to a point where he was unable to guarantee separation once the trainee made an error.

VH-MFS was inbound to Darwin and cleared to maintain 1,500 feet by Darwin Approach Control, then subsequently instructed to call Darwin Tower.

When the pilot contacted the tower, he was advised of traffic, VH-BYK, a Cessna 210 on downwind for runway 36. This aircraft was sighted about two kilometres in front and 100-200 feet below. VH-MFS passed directly over VH-BYK.

The pilot of VH-MFS was advised by the RAAF controller that an error of judgement on their part had resulted in the confliction.

Safety Action

Deficiencies identified during this and other investigations involving procedures by Darwin RAAF controllers are the subject of further analysis in conjunction with the Directorate of Flying Safety - Air Force.

Circumstances

VH-TJR [Boeing 737] had departed Melbourne for Canberra at 1529 hours and was on a 10NM final for runway 35 at Canberra airport. The crew had been cleared to make the approach by Air Traffic Control [ATC].

VH-TCO [Cessna 150] was performing circuit training with a student pilot at the controls being supervised by an instructor. The aircraft was instructed by ATC to make a right orbit when on downwind leg and as the trainee was completing the manoeuvre ATC instructed the crew to sight and follow a B737 on an 8NM final. They looked to their left and saw a twin engined jet passenger aircraft that looked like a B737 and acknowledged the ATC instruction as they turned their aircraft onto base leg. The aircraft they saw was an Airbus Industries A320 which was carrying out an approach to runway 35 ahead of VH-TJR.

As VH-TCO was proceeding on its base leg the instructor realised that the aircraft they saw was not at the 8NM final position that was indicated by the Aerodrome Controller [ADC]. He then looked to his right and saw a second twin engined jet passenger aircraft, VH-TJR and calculated that his present track and altitude would take his aircraft above and behind that aircraft. At this point the ADC again issued traffic information to VH-TCO as he observed the two aircraft closing on one another.

As the relative tracks of the two aircraft closed the captain of VH-TJR looked to his left and saw VH-TCO in a position that appeared to be on a collision course and immediately took evasive action by turning his aircraft to the right. After diverging about 0.5NM, the aircraft then re-joined final approach and made an uneventful landing.

The crew in VH-TCO did not take evasive action, as the instructor watched the B737 pass clear, in front and below his aircraft. He had maintained visual contact with VH-TJR continuously from the time he had first seen it. The aircraft came within 300 metres and 300ft of each other.

Significant Factors

1. The crew of VH-TCO initially sighted the wrong aircraft and selected a flight path based on this incorrect sighting.

2. The crew of VH-TCO realised their error while on base leg and elected to continue with that track having correctly sighted the B737.

3. Separation between the aircraft reduced to the point where the crew of VH-TJR considered that evasive action was necessary to maintain adequate clearance from VH-TCO.