Loss of separation

FACTUAL INFORMATION

The Saab 340 aircraft was inbound from Orange for a landing at Sydney and the crew had been instructed to track via the Sydenham locator for a left circuit to runway 16L. The Saab had been cleared to descend to 6,000 ft. At about the same time a Mooney aircraft, en route from Bankstown to Tobins Gap, was on climb to 5,000 ft while tracking north from overhead the Sydney Very High Frequency Omni-Directional Radio Range (VOR) navigation aid to Williamtown.

The approach controller was aware of the possibility of conflict between the aircraft due to the limited displacement (approximately 2.5 NM) between the Sydenham locator and the Sydney VOR. Consequently, he used a function of the radar display to highlight the symbol and label of the Mooney to assist in monitoring the aircraft's track. However, due to the close proximity of the terminal radar antenna and the VOR, the Mooney passed through the radar overhead cone of silence (an area immediately above the antenna where the radar is unable to detect aircraft) as it overflew the VOR. Consequently, the Mooney's symbol and label disappeared from the controller’s radar display.  The radar display does not retain controller inputs for lost tracks and subsequently did not re-highlight the symbol and label when the aircraft emerged from the cone of silence. Consequently, the controller was not provided with the visual cue to assist in monitoring the aircraft's symbol after it passed over the VOR.

The controller intended to instruct the pilot of the Saab to turn downwind after passing overhead the Sydenham locator, an action that would ensure separation was maintained between the Saab and the Mooney.  However, this instruction was not passed to the pilot as the controller commenced a handover/takeover to a new controller. As the Saab passed overhead the locator, the pilot maintained an easterly heading which was a converging course with the Mooney. The controller could not confirm whether or not the Saab's symbol and label were continuously displayed on the radar screen. The radar recording system showed a continuous plot for the aircraft; however, such a recording does not replay individual radar screen presentations and consequently, the actual radar display provided to the controller could not be confirmed.

The pilot of the Mooney had been instructed, by a controller on an adjacent control position, to turn north to track to Williamtown and climb to 8,000 ft. As separation between the aircraft reduced, the two controllers conducting the handover/takeover were alerted to the situation by a controller at an adjacent control position. The latter queried the track of the Saab as the aircraft was about to infringe his airspace. The new on-duty controller instructed the pilot of the Saab to turn left while the pilot of the Mooney was given traffic information and instructed to turn right for separation. Traffic was busy but not abnormal for the time of day and the weather was fine with unrestricted visibility.

The aircraft passed with approximately 1.5 NM horizontal separation and 100 ft vertical separation. The required standard was 3 NM horizontally or 1,000 ft vertically. There was a breakdown in separation.

ANALYSIS

Controllers were aware of the problem with the radar display not retaining controller inputs for lost tracks. This had become apparent with the advent of new procedures for parallel runway operations. The actual radar cone of silence is very narrow, and approach controllers sometimes transfer aircraft early to either of the Directors (East or West as appropriate) to ensure aircraft tracking via the Sydenham locator do not enter the cone. The directors operate on a larger scale display and consequently are better able to monitor aircraft clear of the cone and maintain aircraft symbols on the display.

In this incident the controller elected to retain the aircraft and then became distracted while conducting a handover/takeover. The provision of separation assurance techniques would have ensured separation between the two aircraft while they were in close proximity to the cone of silence. This was not done. Alternatively, the controller could have tracked the Mooney via a route which would have ensured that the aircraft did not enter the radar cone of silence and, consequently, the system would have maintained the track on the display.

The situation was compounded by the inability of the radar and display systems to continue plotting the track of the Mooney as it passed through the overhead cone of silence. Secondary surveillance radar data from the Sydney terminal area is available from two sources and the use of multi-radar tracking would eliminate the blind spot due to the cone of silence. However, this facility is currently not available at Sydney. Additionally, the system was unable to maintain controller display inputs to re-highlight the track. While not major factors in the incident, the lack of these functions reduced the defences available to the air traffic system.  New equipment being provided by Airservices Australia will address this deficiency.

SIGNIFICANT FACTOR

Separation assurance techniques were not adequately utilised.

SAFETY ACTION

Safety deficiencies involving handover/takeover aspects are being addressed through Occurrence 9600800.

Visual approaches to Sydney Airport were in progress, domestic arrivals being processed for runway 34R, and international arrivals for runway 34L. The arrival sequence was being handled by Approach South (APP-S) and Approach North (APP-N) radar controllers. Traffic density was described as light to medium.

To resolve a traffic management problem in his airspace, APP-S coordinated with APP-N for a Metro 23 to complete a right circuit for arrival to runway 34R. This involved radar vectoring the aircraft to pass overhead Sydney Airport onto a right downwind leg, east of the runway complex, an area which was normally under the jurisdiction of APP-N. However, as there was no traffic east of the runway at the time, it was agreed to keep the Metro 23 on the APP-S frequency.

A SAAB 340 (SF34) was also being processed by APP-S for a landing on runway 34R and was given radar vectors to join on left downwind. During the arrival sequence, the crews of both aircraft were given instructions to descend to 2,000 ft and were vectored onto opposing headings for their respective base legs. Some coordination concerning other aircraft took place at about this time. When the Metro 23 was passing through the final approach path of runway 34R, the crew questioned the controller as to whether they were to turn the aircraft right to intercept final approach. APP-S confirmed this, then vectored the SF34 to remain clear of the Metro 23. Recorded radar data indicated the aircraft passed within 2 NM of each other, with a minimum vertical separation of 200 ft. The required radar separation standard was 3 NM laterally or 1,000 ft vertically.

The decision to leave the Metro 23 on the APP-S frequency, whilst being vectored within APP-N, airspace was considered to be a significant factor in the development of this occurrence. Whilst the APP-S controller believed he had not forgotten to issue the crew of the Metro 23 with a clearance to turn onto final approach, it appeared he had a mind-set, or expectancy, that APP-N would process the aircraft, as would normally happen. This mind-set was experienced despite the fact that the colour of the Metro 23's radar symbol indicated it was under the jurisdiction of APP-S.

ZK-NBI was inbound to the Plenty locator from Wonthaggi on descent to 5,000 ft. VH-UWW departed Essendon runway 17 with departure instructions to turn left heading 090 deg with an unrestricted climb to 5,500 ft. These instructions were issued by the approach controller who was also controlling ZK-NBI.

Approximately four minutes after VH-UWW departed Essendon, the controller realised that the two aircraft were on conflicting tracks with no vertical separation. Both aircraft were given heading changes, but their tracks closed to approximately two miles at the same altitude before radar separation was re-established.

Investigation revealed that the approach controller failed to ensure maintenance of separation between the two aircraft because he was distracted by separation requirements for other traffic. He had intended to turn VH-UWW to the northeast earlier to maintain separation, but due to the distraction this did not occur.

Two other factors contributed to the failure. The first one was that when the approach controller issued the departure instructions for VH-UWW, ZK-NBI was not displayed on his radar screen. The second was that ZK-NBI maintained an unusually high speed on descent. Even though the aircraft had been issued with no speed restriction below 10000 ft, an unusually high groundspeed was maintained.

A breakdown in separation occurred between two VFR aircraft, a twin Comanche cleared Station Pier for Essendon at 1,500 ft and a helicopter cleared World Trade Centre for Essendon on climb to not above 1,500 ft. Essendon Tower passed traffic to each aircraft, but the helicopter pilot claimed not to have heard advice on the Comanche.

The Comanche pilot never saw the helicopter whereas the helicopter pilot saw the Comanche pass about 200 ft overhead.

Significant Factors

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

1. Both aircraft were cleared at the same altitude.

2. The two pilots did not take adequate action to ensure separation was maintained.

3. Air traffic control did not provide sufficient information to the two pilots to assist them to sight each other's aircraft.

The out-going controller stated that the runway configuration made the management of traffic complex, a situation which required a high level of concentration. She had spent approximately one and half hours on duty and felt ready for a break. Consequently, once the new controller commenced in the approach east position, she relaxed. However, she remained at the console monitoring the audio program for approximately another 10 to 15 minutes but did not maintain a full appreciation of the new traffic sequence. She provided only limited coordination assistance to the new controller.

When the Navajo was north-east of Perth, the approach east controller radar vectored the aircraft to separate it from the Baron and to position it for right base runway 03. This was in compliance with a local instruction that required inbound aircraft which would overfly the departure end of a nominated runway to be maintained above 4,000 ft or radar separated from departing traffic. After radar separation was established with the departing Baron, the pilot of the Navajo was instructed to descend to 2,500 ft. When the Navajo was approximately 20 NM east of Perth, the approach east controller elected to track the aircraft for runway 11. The Navajo was radar vectored on a heading of 270 degrees for a left circuit to that runway.

Shortly after, the approach east controller observed the secondary surveillance radar altitude readouts from both aircraft and determined they would pass without adequate vertical or horizontal separation. The pilot of the Navajo was instructed to maintain 3,000 ft, which provided 500 ft vertical separation from the Baron. This was less than the required vertical separation of 1,000 ft. The approach east controller also instructed the pilot of the Baron to turn right onto a south-easterly heading for separation. As the controller further determined that both the horizontal and vertical separation standards were not going to be maintained, he passed traffic information to the pilot of the Baron. The pilot of the Baron subsequently sighted the Navajo and monitored the flightpath of the other aircraft until it was well clear.

The two aircraft passed with less than 3 NM horizontal separation and less than 1,000 ft vertical separation. There was a breakdown in separation.

ANALYSIS

The Navajo was the second of three aircraft being sequenced for runway 03. The controller determined the Navajo might eventually conflict with following traffic and considered changing the aircraft to runway 11. However, a pending Airbus departure to the east from runway 03 would possibly have conflicted with the Navajo as it tracked downwind and thus it remained sequenced for runway 03.

When the Airbus from runway 03 was airborne, the controller reconsidered the earlier option of changing the Navajo to runway 11. Two larger and faster aircraft, following for runway 03, were closing and would require vectoring to maintain separation and spacing for landing. The controller decided to change the Navajo to runway 11. At this stage the Baron and Navajo were radar separated and assigned the same level. However, when the approach east controller turned the Navajo downwind for runway 11, he placed both aircraft on closing reciprocal radar headings without re-establishing vertical separation in accordance with local instructions.

The turn given to the pilot of the Baron was an attempt to maintain 3 NM separation. The controller was also aware that 500 ft was insufficient separation for IFR category aircraft but believed that under the circumstances it was better than having the aircraft at the same altitude. The provision of traffic information was appropriate and enabled the application of visual separation by the crew of one of the aircraft.

The out-going controller could have contributed more to assist the new controller in the period immediately following the handover/takeover. While the new controller would have appreciated the immediate disposition of aircraft on the radar display it takes some time to appreciate the plan for a developing sequence of air traffic. The basis of the handover/takeover and monitoring process is to ensure the out-going controller maintains the full disposition of aircraft and understands what actions are needed to manage and separate, immediate and pending aircraft until the new controller fully appreciates the disposition and sequencing plan of the control position. While the initial reaction of a controller to relax after handing over is understood, for an effective handover/takeover, all controllers at a control position should maintain a state of awareness commensurate with active controlling until the new controller indicates a complete comprehension of the traffic situation.  Perth Approach Control Centre does not have specific handover/takeover instructions for controller guidance.

During the investigation, aspects of the restructure of Perth airspace scheduled for implementation by December 1996 were provided. The restructure will provide track crossover points outside 30 NM and separation assurance procedures. While these measures will not necessarily alleviate similar occurrences, they should provide an airspace environment better able to cope with similar situations.

SIGNIFICANT FACTORS

1. The aircraft were required to pass on approximately reciprocal tracks due to the runway configuration.
2. The approach east controller did not use separation assurance techniques when there was a possibility the aircraft would conflict.
3. The out-going air traffic controller did not fully monitor the new traffic sequence and only provided limited assistance to the new air traffic controller.

SAFETY ACTION

The Bureau of Air Safety Investigation is evaluating controller handover/takeover procedures. Any forthcoming recommendation will be published in the Quarterly Safety Deficiency Report.

At 0812, VH-CZL departed runway 16 at Melbourne on a 16 Cowes 5 departure with instructions to maintain 3000 ft. This departure involves tracking 170 deg until through 2500 ft or four miles DME and then tracking direct to Cowes which is on the Melbourne 150 deg VOR radial. After CZL made contact with departures control, the aircraft was cleared to climb to flight level 370.

At 0813, N61CF departed from runway 17 at Essendon heading 170 deg and initially limited to 3000 ft. At 0815 CZL left 5000 ft and N61CF was cleared to climb to 4000 ft. Shortly afterwards N61CF was instructed "at 4000 ft turn right heading 260 deg".

At 0816, N61CF was observed in a right turn and climbing through 4700 ft. At this time CZL was climbing through 5500 ft with N61CF less than two miles away and closing. N61CF was asked to confirm maintaining 4000 ft and passed traffic on CZL. The pilot of N61CF advised the traffic was in sight and apologised for the transgression.

FACTUAL INFORMATION

A Boeing 727 aircraft was en route from Melbourne to Brisbane via Casino and Coolangatta at flight level (FL) 350, while a Boeing 767 aircraft was en route to Brisbane from Sydney via BANDA and Coolangatta at FL370. The Boeing 767 was east of the Boeing 727's track and a few minutes behind on time estimates. There was a holding requirement at Brisbane with instrument landing system approaches in progress due to reduced visibility in passing rain showers.

The Brisbane sector radar position responsible for controlling the two aircraft was manned by a trainee controller undergoing final checks prior to rating. The trainee was being supervised by a current sector controller. However, the supervising controller was not the regular training officer for the trainee, and this was the first occasion that they had worked together. The Boeing 727 was initially vectored in a northerly direction for separation from other traffic but was shortly after instructed to turn right onto a north-easterly heading for arrival sequencing at Brisbane. The trainee controller intended to maintain the Boeing 727 on the north-easterly heading and provide a final heading to Coolangatta to meet the time required for sequencing. The Boeing 767 entered the controller's area of responsibility, and the crew requested descent to FL330 due to moderate turbulence.

The Boeing 767 was recleared to FL330. The new level was read back and the crew reported leaving FL370. The supervising controller discussed with the trainee the requirement to ensure separation between the two aircraft, since they were on converging tracks with the Boeing 767 crew cleared to descend through the Boeing 727's level. Discussion centred on whether the Boeing 767 should be radar vectored to parallel the track of the Boeing 727 until the required vertical separation of 2,000 ft was established. However, as there was more than sufficient horizontal separation (minimum separation required was 5 NM), the trainee controller elected to leave the Boeing 767 under the crew's navigation and to monitor the situation. The supervising controller believed it was reasonable to allow the trainee a level of latitude for independent action greater than he normally would for a trainee as the trainee controller had almost completed training. As separation between the two aircraft reduced to approximately 15 NM, the trainee controller's attention was diverted as he conducted coordination with another control position. As horizontal separation approached 8 NM and vertical separation was approximately 700 ft (with the Boeing 767 below the level of the Boeing 727) the trainee controller instructed the crew of the Boeing 767 to turn right onto a heading to parallel the track of the Boeing 727.

During this transmission the trainee controller incorrectly advised the Boeing 767 crew that the radar heading was for sequencing, instead of separation. The trainee controller requested the crew of the Boeing 767 to expedite descent and to confirm that the aircraft was turning right. The crew advised that the aircraft was turning right and expediting descent and, shortly after, reported sighting the Boeing 727 as horizontal separation reduced to approximately 4 NM with vertical separation of 1,600 ft. Horizontal separation reduced further to just over 3 NM before 2,000 ft vertical separation was re-established. The operating console was not fitted with an air-ground-air communication override facility to enable the supervising controller to intercede in trainee transmissions. Traffic information was not provided to either aircraft as the horizontal and vertical separation reduced to less than the standard. There was no Traffic Alert and Collision Avoidance System (TCAS) report from the B767. ANALYSIS The trainee's use of the term "sequencing" instead of "separation" did not provide the level of notice to the crew of the Boeing 767 that was warranted under the circumstances. Had the crew been advised that the heading change was for separation they may have been more expeditious in complying. However, the situation should not have been allowed to develop to the stage where immediate action was required to maintain separation.

Similarly, the inclusion of a console facility that would enable the supervising controller to override the trainee's transmissions to establish appropriate separation measures would have been beneficial. But again the situation should not have developed to the extent that such facilities were essential. Both controllers had discussed separation requirements for the two aircraft, and the trainee was satisfied with monitoring the situation. The supervising controller deferred to the trainee's judgement because he understood the trainee was close to achieving a rating. However, the trainee became distracted, and the supervising controller did not adequately monitor the trainee's subsequent control actions.

The supervising controller could have emphasised to the trainee the need to give priority to the radar display and to maintain a high scan rate to ensure separation. However, the situation would have been still totally dependent on the performance of the trainee and the supervisor. As the trainee became distracted by coordination to the detriment of control, the supervising controller was unable to intervene adequately, and separation was lost between the aircraft. Monitoring of the flight paths did not provide an adequate level of separation assurance. Early implementation of an altitude requirement or a radar heading (to parallel the track of the Boeing 727) to the crew of the Boeing 767 would have been an adequate separation assurance technique. Use of separation assurance techniques would have greatly improved the possibility of the two aircraft remaining separated.

SIGNIFICANT FACTORS

1. The supervising controller did not adequately monitor the trainee controller's actions or ensure that a suitable separation assurance technique were employed.

2. The use of inappropriate radiotelephony phraseology by the trainee controller did not impart to the crew of the Boeing 767 the need for expeditious compliance with instructions.

3. The lack of a communications override facility deterred the supervising controller from implementing timely remedial action.

SAFETY ACTION

The Bureau of Air Safety Investigation is continuing its investigation into the provision of adequate communications override facilities for air traffic control training officers. This is intended for all consoles where on-the-job training of air traffic controllers is likely to occur.

The Cherokee departed from runway 14 to track to Flat Top Island, 4 NM east of Mackay, to carry out Airwork. Soon after, a Brasilia reported ready to take-off from the same runway, with the intention of tracking to Rockhampton via the 131 VOR radial. The aerodrome controller cleared the Brasilia to take-off after observing the Cherokee appeared to be established on track to Flat Top Island.

The pilot of the Brasilia reported that as he turned left to intercept the 131 VOR radial, he noticed the VOR had apparently failed and attempted to resolve the problem. Soon after, the copilot reported another aircraft immediately ahead. The pilot saw a Cherokee about 200 metres away at the same altitude, but slightly to the left, and passing from left to right. He took immediate evasive action by turning left to pass behind the other aircraft.

The investigation established that the instructor in the Cherokee had been concentrating on the instructional sequence and had not continued on track to Flat Top Island as cleared. The Cherokee had been subsequently turned onto a south-westerly heading, which conflicted with the flight path of the departing Brasilia.

FACTUAL INFORMATION

A BAe146 (VH-NJN) was maintaining flight level (FL) 280 on a flight from Cairns, Qld to Ayers Rock, NT.  The crew reported over Alice Springs at 1236 CST with an estimate for Ayers Rock of 1312.

A second BAe146 (VH-JJW) departed Alice Springs at 1237 for Ayers Rock and climbed to FL240.

Both aircraft were under the control of Melbourne air traffic control (ATC) sector 5 which was staffed by a trainee under the supervision of a team leader. The aircraft were subject to procedural control procedures.

The trainee controller coordinated the two aircraft positions with flight service, giving the estimated times of arrival (ETA) at Ayers Rock as 1312 for NJN, and 1318 for JJW.   An error was made in the time of arrival for JJW in that the trainee incorrectly added the flight plan time intervals.  The correct estimate was 1316.  This error was not detected by the team leader.

When both aircraft were on the control frequency, the trainee checked their respective DME distances, this resulted in NJN being 22 NM ahead of JJW. As the aircraft were of the same type, a suitable standard (15 NM) had been established to allow the descent of NJN through the level of JJW.

At 1253, the crew of NJN requested descent.  This was approved by Melbourne control and the aircraft commenced descent at 1255.  At 1256, the crew established contact with flight service and advised their ETA as 1311.

At 1259, the crew of JJW requested descent.  This was approved by Melbourne control and descent was commenced at 1302.  At 1303, the crew contacted flight service and advised their ETA as 1312.  The flight service officer then passed traffic information on three aircraft, including NJN, to the crew of JJW.

The flight service officer also passed traffic information on JJW to the crew of NJN who then transferred to the Mandatory Broadcast Zone (MBZ) frequency.

Shortly after, the crew of JJW also changed to the MBZ frequency and commenced a DME check with the crew of NJN.  This resulted in the aircraft being 12 NM apart while JJW was passing FL180. Based on this information, the crew believed that NJN passed through their flight level in controlled airspace with only 12 - 14 NM separation and therefore a breakdown in separation had occurred.

As the aircraft approached Ayers Rock, in the MBZ, a pilot initiated distance check revealed that the aircraft had reduced the separation to 4 NM.  Both aircraft landed without further incident.

ANALYSIS

Air Traffic Control

The two minute error made by the trainee in the ETA for JJW at Ayers Rock was not considered to be a factor in this occurrence because the flight service officer passed traffic information based on the pilot reports not the ATC estimates and the DME check was independent of the ETA.

The controller established a correct separation standard (15 DME) for aircraft of the same type and issued control instructions appropriate to the circumstances.  It is probable that the correct distance for this standard was maintained throughout the period of flight within controlled airspace.

The flight progress strips that the controllers had for reference did not differentiate between the various series of BAe146 aircraft.  Additionally, the Manual of Air Traffic Services only published some of the various profile speeds for the BAe146 series of aircraft and these were displayed in the radar control section, not specifically in the procedural control sections. Flight Service

The flight service officer provided traffic information in a correct manner appropriate to the circumstances.

Aircrews

Both crews acted in accordance with ATC instructions and operated their aircraft within company profiles.

Flight Plans

The flight plan held by the crew of JJW varied to that held by ATC in that the ETA Ayers Rock on the crew's plan was 1312 and on the ATC plan was 1316. The reason for this discrepancy was that ATC took the overall block time provided by the flight plan and had their computer calculate the ETA at each reporting point based on the various distances from one point to the next.  As the company provided a two minute manoeuvring time at the departure and destination aerodromes, a four minute difference existed between the times provided to ATC in the flight plan and the time the pilot held in the cockpit.  The pilot based his reporting calculations on the time of departure and had, therefore, already accounted for the two minute manoeuvre prior to advising his actual departure time to ATC. ATC then added the computer generated time intervals to this reported time of departure and the erroneous situation was complete.

Aircraft Performance

ATC reference data for the BAe146 aircraft indicated that there was a similarity of performance between the operating profiles of the various series of BAe146 aircraft operated by all companies.  This was not the case, as the operating profiles indicated that a difference of up to .08 Mach can exist while two aircraft are on descent.

This performance differential could lead to a degradation in certain separation standards such as the 15 DME standard used in this case.

Additionally, the reference material was only contained in the radar sections of the Manual of Air Traffic Services.

SIGNIFICANT FACTOR

The air traffic controllers had insufficient information to indicate that significant closing between aircraft may occur.

SAFETY ACTION

As a result of the investigation the following action was taken:

Local ATS management introduced an amendment to the Manual of Air Traffic Services (MATS) Local Instructions which specifies the full range of performance options for BAe146 series aircraft.

Airservices Australia has undertaken to amend MATS to ensure that the information on BAe146 aircraft is adequately displayed for controller reference.

Airservices Australia and the operating company of JJW are consulting over the format for flight plan submission so that both parties understand the requirements and operating methods of the other.  This process should ensure that controllers and aircrew have the same initial information on which to base their operational requirements.

Factual Information

The DHC8 aircraft VH-TNG was cleared for take-off from runway 32 and was assigned a heading of 315 degrees, which is the runway direction. After take-off the pilot turned right onto a heading of 315 degrees, and this reduced separation and brought the aircraft into conflict with the Boeing 737 VH-TJD which had become airborne from runway 01 at the same time. The pilot of VH-TJD had been instructed to turn left at 3 NM onto a heading of 350 degrees. The pilot of VH-TNG called Brisbane Approach at about 2 NM established on a heading of 015 degrees. He was instructed to immediately turn left onto a heading of 310 degrees to re-establish separation.

Recorded Radar Data

Examination of recorded radar data was undertaken to determine the proximity of the aircraft at the time of the incident. VH-TNG was observed to take off on runway 32 and commence a right turn at recorded mode C pressure altitude of 900 ft. At 2000 ft the aircraft had stabilised on a magnetic track of 015 degrees. At 2700 ft VH-TNG turned left and commenced tracking 327 degrees magnetic.

VH-TJD was observed to take off on runway 01 and commence tracking along the extended centreline of the runway. At approximately 3.5 NM DME at a pressure altitude of 2800 ft, VH-TJD turned left and commenced tracking 348 degrees magnetic.

A minimum horizontal separation of 2.2 NM occurred at approximately 0832 UTC and the vertical separation at that time was 700 ft. The minimum separation standard is 3 NM and 1000 ft.

Examination of VH-TNG Crew Actions

The Company inquiry reconstructed the likely sequence of events leading up to the incorrect heading being taken up by the pilot of VH-TNG.

Pre-flight Preparation

The cockpit pre-flight preparation proceeded normally. An airways clearance was obtained, a transponder code was not immediately available but was this was issued later during taxi. The pilot in command (PIC) was the pilot flying for this sector. he conducted the standard turn-around, pre-take-off and departure briefing according to the airways clearance received. This briefing was given relative to an anticipated departure from runway 01. Standard procedure in this circumstance is to set 016 (runway magnetic heading) on both horizontal situation indicators (HSIs) with an allowance for crosswind drift.

Taxi Out

When taxi clearance was obtained, both runways 01 and 32 were available for departure, and runway 32 was assigned. When clear of the inner apron areas the PIC confirmed the instrument serviceability checks, briefed the change of HSI heading setting to 315 degrees (runway 32 magnetic heading) plus 5 degrees for crosswind allowance. The PICs HSI was set to 320 thus acknowledging the change of runway from the runway anticipated in his previous briefing. The PIC could not be certain that the copilot reset his HSI to the appropriate heading. The copilot believes he did reset the HSI, but there was inadequate cross checking at this point, and it is possible that the copilots HSI may have remained set at 016.

The PIC then continued with a revised take-off and emergency return brief for runway 32. No revised brief for the Standard Radar Departure (SRD) relative to the change of runway was made. This was at variance with company standard operating procedures (SOPs) and the brief would have ensured that both HSIs were set correctly to 315 plus wind allowance.

Take-off

When all checklists were completed as far as possible the radio was transferred to tower frequency, and the copilot called ready some distance back from the holding point. The aircraft was cleared to line up on runway 32. As the aircraft rolled onto the runway to lineup the tower advised Tango November Golf assigned heading 315, clear for take-off. No reference was made to turning left or right and as evidenced from a play back of the tower tape, the instruction was clear and concise. The copilot hesitantly acknowledged the clearance with Tango November Golf, right onto 315.  The words right onto were not queried by the tower or the PIC. At this stage the copilot, as non-flying pilot should have set and confirmed his heading on the HSI to be 315. The PIC remembered the copilot reaching for the adjusting knob on the HSI but did not confirm that the HSI was correctly set. This a further breakdown in SOPs.

After take-off, checks were completed and passing 600 feet the PIC asked the copilot to confirm the right turn. The copilot confirmed right onto 015. The copilot the transferred to Brisbane Approach Frequency, and the aircraft was then promptly vectored away from the resulting confliction.

Analysis

Air Traffic Control Phraseology

The instruction was given assigned heading 315, cleared for take-off. The use of the phrase "maintain runway heading" may have prevented this incident. The Civil Aviation Safety Authority Head Office had recently issued an instruction to Air Traffic Services to discontinue the long-established practice of issuing the instruction " maintain runway heading."

Crew

The PIC had only recently gained command status and was relatively inexperienced on the DHC8. The copilot was senior in age and experience on type which may indicate a flat cockpit command gradient.

Relevant Factors

1. The PIC did not adequately re-brief on the standard radar departure or the change of runway to 32.
2. The PIC did not adequately monitor or cross check the copilot’s instrument settings.
3. The PIC did not detect the error in readback of the departure clearance by the copilot and was unclear as to the terms of the departure clearance before take-off.
4. The crew did not comply with Company SOPs and two crew procedures.
5. Inappropriate Air Traffic Control phraseology.