Safety summary

What happened

Late in the afternoon on Sunday 19 July 2015, an amateur-built Stoddard Hamilton Glasair SH-2FT two-seat aeroplane, registered VH-HRG and operated in the Experimental category, was seen flying due north, consistent with the downwind leg of a circuit for landing at Wedderburn Airport, New South Wales.

Witnesses stated that they heard the aircraft’s engine surge twice and then silence, prior to hearing the aircraft collide with wooded terrain about 900 m north of the runway threshold. No witness reported seeing the aircraft turn onto the base leg or final approach, nor the aircraft collide with terrain.

The pilot sustained serious injuries, the passenger was fatally injured, and the aircraft was destroyed.

What the ATSB found

The ATSB found that during the turn onto final approach to land, the aeroplane’s engine ceased operating. Following the loss of power, the pilot was unable to control the aircraft’s descent to an appropriate forced landing area before colliding with the ground.

The ATSB also found that the loss of power was probably due to carburettor icing. No defects were identified that would have precluded normal operation of the aircraft or its engine prior to the accident. However, the environmental conditions at the time were conducive to serious carburettor icing at all power settings. The pilot reported using a low power setting during the downwind leg of the circuit to slow the aircraft down and did not use the carburettor heat system.

There was insufficient evidence to support other possible hypotheses for the aircraft’s loss of control, such as an aerodynamic stall as a result of aircraft handling.

Safety message

The ATSB advises pilots of aircraft fitted with a carburettor to check the forecast weather conditions affecting their operations, and consider the risk of carburettor icing prior to each flight. Pilots should be aware that carburettor icing can form over a wide range of outside air temperatures and relative humidity and understand the importance of following aircraft manufacturer guidelines regarding the use of carburettor heat. Further, they should be mindful that certain flight conditions, such as lower engine power settings, may increase the risk of ice accumulating in the engine’s carburettor.

The Civil Aviation Safety Authority’s Carburettor icing probability chart provides helpful guidance for pilots to determine carburettor icing probability before flying. The chart is available at www.casa.gov.au.

Photograph of VH-HRG

Source: NSW Sport Aircraft Club

The occurrence

At about 1625 Eastern Standard Time^[1] on Sunday 19 July 2015, the pilot and passenger of an amateur-built Stoddard Hamilton Glasair SH-2FT two-seat aeroplane, registered VH-HRG and operated in the Experimental category, took off from runway 17^[2] at Wedderburn Airport, New South Wales. The pilot recalled a ‘shallow’ take-off due to the aircraft’s weight (two people on board plus almost full fuel) and the characteristics of the aircraft’s wooden cruise propeller.

At 1628, an unverified aircraft was observed on Airservices Australia surveillance radar for the area, consistent with the position of VH-HRG. The aircraft tracked in a south-south-westerly direction toward Wollongong. At about 1639, when near Wollongong, the aircraft turned left and commenced tracking back to the north (Figure 1).

Figure 1: Radar trace of the aircraft that appeared on Airservices Australia surveillance radar in the area of Wedderburn Airport at 1628 (at the top, with the aircraft’s radar returns shown in lilac) and the radar trace and returns (also in lilac) overlaid on Google earth (right)

Source: Airservices Australia and Google earth, modified by the ATSB

Witnesses stated that, at about 1649, the aircraft was seen flying due north, consistent with the downwind leg of a circuit in preparation for landing on runway 17 at Wedderburn Airport. The pilot recalled that the aircraft’s airspeed was higher than preferred on downwind, being greater than 100 kt. The pilot reported attempting to reduce airspeed by adjusting power and the aircraft’s attitude. However, the airspeed remained high when on late downwind. The pilot indicated that by that time, power had been reduced to the idle position. The pilot further indicated that when on downwind, the aircraft was closer to the runway than normal.

The pilot commenced a continuous turn onto base and final at an airspeed of about 90 kt and applied full flap. The pilot indicated that the power setting remained at idle until approaching the finish of the turn onto final. The pilot then attempted to apply power to arrest the aircraft’s deceleration. However, the pilot stated that when power was applied, the engine coughed and spluttered. The pilot recalled the aircraft’s airspeed at that time as about 80–85 kt. The pilot commented that the engine was left at idle longer than preferred. The pilot could not recall anything further until after the impact with the ground.

Witnesses also stated that they heard the aircraft’s engine surge twice and then silence, prior to hearing the aircraft collide with wooded terrain, about 900 m north of the threshold of runway 17 (Figure 2). None of the witnesses reported seeing the aircraft turn onto the base leg or final approach, nor the aircraft collide with terrain.

The pilot sustained serious injuries and the passenger was fatally injured. The aircraft was destroyed.

Figure 2: Aircraft wreckage. Note that the tail empennage is secured in an elevated position by being tied off to a nearby tree

Source: ATSB

__________

1. Eastern Standard Time (EST) was Coordinated Universal Time (UTC) +10 hours.
2. Runways are named by a number representing the magnetic heading of the runway.

Context

Pilot information

Qualifications and experience

The pilot held a valid Private Pilot (Aeroplane) Licence and a Class 2 Aviation Medical Certificate that required the pilot to wear distance vision correction and have reading correction available while exercising the privileges of the licence. The pilot reported wearing their distance correction glasses at the time of the loss of power.

No medical anomalies were identified during the pilot’s last medical examination. The pilot last completed a flight review on 1 November 2014. As of 9 June 2015, the total flying hours recorded in the pilot’s logbook was 621.8 hours, with 423.8 hours on type.

Recent history

The pilot’s spouse had recently passed away following a lengthy illness and the funeral was held on 13 July. The pilot reported not sleeping well or eating properly during this stressful period. However, the pilot stated that by 19 July they were sleeping and eating normally as the stress of the previous week had abated. More specifically, the pilot recalled obtaining about 9 hours sleep on the night prior to the accident (18 July) and commented that the sleep was normal. The pilot indicated that on the second and third nights preceding the accident (16 and 17 July), normal sleep was obtained, but that they felt tired on Saturday, 18 July. The pilot could not recall experiencing disturbed sleep or any illnesses.

There was no evidence that fatigue or the recent stressful events contributed to the accident.

A witness who interacted with the pilot on the morning of the accident stated that the pilot presented as being fine and well.

Aircraft information

The Glasair SH-2FT is a high-performance, two-seat, low-wing aeroplane constructed from fiberglass composite components. The FT model features a fixed, tricycle landing gear. The airframe was designed to provide a high top speed and efficiency of operation. Glasair indicated that the high-speed performance was complemented by a low-speed, gentle stall.

The aircraft, serial number W121, was home-built and registered VH‑HRG (HRG) in 1985. The pilot was the third owner of the aircraft since its construction.

HRG was fitted with a Lycoming O-320 engine with a hybrid ignition system and a wooden, fixed‑pitch, two-blade cruise^[3] propeller. The aircraft was issued with a Department of Civil Aviation (Australia) approved flight manual on 5 September 1988.

An estimate^[4] of the aircraft’s weight and balance showed that it was within centre of gravity limits at the time of departure and immediately prior to the accident.

Maintenance history

Examination of HRG’s maintenance records indicated that it was maintained to a day visual flight rules standard in the Private Category. The last periodic inspection was completed on 22 January 2015, and a maintenance release was issued at that time. At the time of the accident, all of the required maintenance had been completed.

Meteorological information

The meteorological conditions at Wedderburn for the flight were reported as calm with light winds. The 1700 aerodrome weather report (METAR)^[5] for Camden Airport, about 18 km to the north-west of Wedderburn Airport, stated that the weather was clear with wind from the south-east at 4 kt (about 7 km/h).

The pilot reported that the weather conditions were fine with a slight crosswind from the left on take-off.

A witness, who was airborne when HRG departed Wedderburn but had landed before HRG returned, reported experiencing a ‘little bit’ of sun glare when on the base leg of the circuit to runway 17. Geoscience Australia astronomical information^[6] showed that, at 1649, the sun would have been 2° 51.750’ above the horizon and that sunset was at 1707. The pilot reported they could not recall experiencing any sun reflection on the windscreen during the flight.

Ambient condition-related carburettor icing^[7] probability

Bureau of Meteorology weather observations at Campbelltown (Mount Annan)

Bureau of Meteorology weather observations were available for Campbelltown^[8] (Mount Annan), about 13 km north-north-west of the accident site. At around the time the aircraft departed Wedderburn Airport (about 1630), the temperature was 13.4 °C and the dewpoint^[9] temperature was 5.1 °C. This resulted in a dewpoint depression of 8.3 °C. Around 10 minutes after the accident (about 1700), the temperature was 11 °C and the dewpoint temperature was 4.9 °C. This resulted in a dewpoint depression of 6.1 °C.

Camden Airport METARs

The 1630 METAR for Camden Airport indicated an air temperature of 13 °C and a dewpoint temperature of 5 °C. This resulted in a dewpoint depression of 8 °C.

The 1700 METAR for Camden Airport indicated an air temperature of 12 °C and dewpoint temperature of 5 °C. This resulted in a dewpoint depression of 7 °C.

According to the Civil Aviation Safety Authority (CASA) carburettor icing probability chart (see appendix A), the conditions recorded at Campbelltown (Mount Annan) and Camden Airport were conducive to ‘serious’ icing at all power settings.

Pilot reports

A pilot who was flying at Wedderburn Airport at the time HRG departed reported that, as far as they were aware, they did not experience carburettor icing during their flight. However, that pilot indicated being vigilant in using carburettor heat as their aircraft was prone to such icing, and that they had applied carburettor heat on that flight.

Two other pilots who were flying at Wedderburn Airport at the time also indicated that no carburettor icing issues were experienced and that they too were vigilant in the use of carburettor heat.

The pilot of HRG reported believing the conditions affecting the flight were not conducive to carburettor icing.

Airport information

Wedderburn Airport was an uncertified, unregistered landing area about 13 km south of Campbelltown. The airport had a sealed and adjacent grass runway and was aligned 17/35 (roughly north-south) (Figure 3 inset). The airport was privately-owned and -operated. The Airservices Australia En Route Supplement Australia (ERSA) extract for Wedderburn advised the following:

• pilots should avoid flying over Appin Township (Figure 3)
• circuit traffic should avoid flying over the populous area to the north of the airstrip (refer to the section titled Wedderburn circuit pattern)
• restricted area R555 is located 1.5 NM (3 km) to the east of the airport (Figure 3).

The terrain surrounding the airstrip was heavily timbered, which provided minimal landing opportunities for an off-field forced landing.

Figure 3: Proximity of restricted area R555 (outlined by the red dashed line) to Wedderburn Airport (bottom-left of the restricted area) and Wedderburn runway 17/35 at inset

Source: Google earth, modified by the ATSB

Wreckage and impact information

On-site information

Examination of the wreckage found that, following a steep descent, the aircraft collided with trees before coming to rest about 26 m further on into a wooded area and about 20 m from a road leading to the eastern gate of Wedderburn Airport. The initial point of impact was a tree about 7 m in height. A second tree, measuring about 30–35 cm in diameter, was impacted about 17 m further along the wreckage trail. The aircraft’s angle of descent during the impact sequence with the trees was calculated to have been about 30°.

The impact sequence resulted in the engine oil sump being breached and destruction of the carburettor. The entire wing section, engine and fuselage forward section separated from the aircraft due to impact forces. The aircraft’s wooden propeller blades were broken off at the root and had shattered.

The aircraft’s landing gear collapsed and there was substantial damage to the fuel tanks and the header fuel tank was separated from the airframe. All header fuel tank plumbing was fractured. As a result of that damage, an unknown quantity of aviation gasoline leaked onto the ground under the wreckage and the amount of fuel on board prior to impact could not be determined. However, about 3 L remained in the header tank when examined.

There was no fire.

Fuel quality and quantity

Fuel quality

A number of other aircraft had refuelled from the NSW Sport Aircraft Club fuel source since 9 May 2015. A search of the ATSB’s occurrence database for the period 1 April–31 August 2015 did not identify any occurrences that could be linked to the fuel supply at Wedderburn Airport.

Fuel quantity

First responders indicated that there was a substantial amount of fuel detected at the accident site. A quantity of this fuel was recovered and tested by the ATSB. The test determined that the fuel smell and colour were consistent with Aviation gasoline 100, and that the sample was free from contamination.

The NSW Sport Aircraft Club fuel records indicated that 98 L of fuel was purchased by the pilot for HRG on 12 April 2015. This was followed by four refuels of 26 L, 24 L, 17 L and 20 L respectively on 9 May 2015 – a total uplift of 87 L. These latter fuel uplifts were consistent with a reported fuel vent line issue that resulted in the vent line being replaced. At that time, the aircraft had full fuel.

In terms of fuel quantity measurements, the owner’s manual noted that, due to the wing dihedral and location of the fuel gauge sender in the wing, the wing tank held a larger quantity of fuel than indicated by the full mark on the gauge in the cockpit. The pilot also indicated that the maximum reading on the fuel gauge was 90 L, yet the aircraft held 175 L when full (156 L in the main tank and 19 L in the header tank). Of this, about 150 L of main tank fuel was usable. There was no fuel dipstick for the aircraft.

The pilot recalled making a 40-minute flight 2 weeks prior to the occurrence flight. In contrast, the last two flights recorded on the aircraft’s maintenance release were:

• a 30-minute flight on 7 June 2015
• a flight on 14 April 2015.

That was, the only flight since the aircraft was filled to full fuel on 9 May 2015 was the 30-minute flight on 7 June 2015.

The pilot reported that the aircraft’s fuel burn rate was 30 L/hr but that 35 L/hr was used for flight planning. Using the planned fuel burn of 35 L/hr, the recorded 30-minute flight on 7 June would have burnt about 17.5 L of fuel. Based on Airservices Australia radar surveillance data, the accident flight was about 20 minutes in duration. Allowing for an additional 10 minutes for start-up, taxi and take-off, this meant that about 17.5 L was burnt on the accident flight.

The pilot reported that the main tank was selected for the duration of the flight and that the header tank was not used. The pilot further indicated that the header tank was only used if they believed the fuel was stale, and that this option was only exercised when doing the run-up checks. In this case, the estimated combined fuel burn for the flight on 7 June 2015 and the occurrence flight of 35 L meant that, at the time of the accident, there was an estimated 115 L of usable fuel in the main tank.

Technical examination of recovered items and components

The aircraft’s engine and associated components were recovered from the wreckage and transported to an approved engine overhaul facility for technical examination under the supervision of the ATSB. In addition, a number of aircraft items and components were recovered for technical examination at the ATSB’s technical facility in Canberra, Australian Capital Territory.

Conclusions

The on-site wreckage examination and post-on-site technical examinations found:

• no airframe or engine defects that may have contributed to the accident
• the main fuel tank contained a significant amount of fuel at impact
• the fuel selector was selected to the main fuel tank
• evidence of low-speed propeller rotation at impact.

The ATSB concluded that it was likely that none of engine or airframe defects or fuel contamination contributed to the loss of engine power.

Survival aspects

Shortly after the collision with terrain, witnesses arrived at the site and rendered assistance to the aircraft occupants. Witnesses reported that both occupants were restrained by at least the lap component of their seatbelt assemblies.

The pilot indicated that the passenger’s seat was fitted with a four-point harness, while the pilot’s seat was fitted with a shoulder/lap-type belt. Both seatbelts were appropriately anchored to the fuselage structure. However, due to cabin disruption during the accident sequence, the ATSB could not verify the remaining seatbelt attachment points.

The ATSB examined the survivability of the accident based on estimates of the aircraft’s speed, the impact angle and the level of aircraft disruption following the collision with terrain. The examination showed that the impact forces imparted to the occupants would normally be expected to result in serious to fatal injuries.

Additional information

Carburettor icing

Carburettor icing can occur in temperatures up to about 38 °C and is less likely in very cold climates. Increased humidity increases the likelihood of this icing. If ice continues to accumulate in the carburettor, the flow of air to the engine reduces and eventually, if the process is allowed to continue, the engine will stop.

Engines operating at reduced power settings are more prone to carburettor icing as the engine induction temperatures are lower due to the reduced airflow. In this case, the airflow through the carburettor is partially impeded by the throttle butterfly valve. This valve provides more area on which the ice can accrete and increases the partial vacuum downstream of the valve. This causes further chilling of the air and the water droplets, further increasing the likelihood of ice accretion. Unless otherwise stated in the aircraft owner’s manual, full carburettor heat should be applied prior to reduced power or closed throttle operations. The resulting warm air assists in preventing carburettor icing.

For aircraft with fixed-pitch propellers, as ice forms there is typically a small decrease in engine RPM but the engine may continue to run smoothly. As ice continues to accumulate, the reduction in RPM continues and the engine will begin to run rough. If the icing conditions are severe enough, and the pilot takes no remedial action, the engine will eventually stop.

The environmental conditions and time between the accident and the ATSB’s examination of the wreckage meant that any icing in the throat of the carburettor would have melted and not been detectable during that examination.

The pilot indicated that during the circuit, the carburettor heat was in the OFF position and the mixture was full rich. The aircraft was fitted with a carburettor air temperature gauge in the instrument panel. The pilot stated that they could not specifically recall the carburettor air temperature approaching Wedderburn Airport. The pilot also stated that checking the reading may not have been included in their normal instrument scan as icing was felt to not be a risk under the existing conditions and at that stage of the flight.

Wedderburn circuit pattern

One witness reported normally conducting the downwind leg for runway 17 closer to the runway due to the proximity of the R555 restricted area (see the section titled Airport information). Combined with noise restrictions to the north of the runway, this resulted in some pilots doing a continuous turn from downwind to base and onto final. However, another witness believed that the restricted area did not have much effect on the downwind portion of the circuit.

Several witnesses reported that they generally flew the circuit tighter than normal to avoid neighbour complaints about aircraft noise.

The pilot of HRG stated that, at Wedderburn, they normally conducted a ‘restricted circuit’, with a continuous turn from downwind to base and onto final due to the nature of the Wedderburn circuit. The pilot advised that the restricted area did not affect its conduct. The pilot of HRG further stated that members of the NSW Sport Aircraft Club, which was located at Wedderburn Airport, were aware of the need to turn onto base earlier to avoid neighbours on the closest roads to the north of the airport.

Glasair-recommended circuit pattern

The Glasair owner’s manual highlighted that the aircraft was a fast, clean aircraft that took longer to slow down than other light aircraft. Therefore, the manual recommended planning ahead and slowing down prior to entering the circuit pattern. The manufacturer recommended that entry to the circuit pattern occur at a speed between 113–122 kt. At that time, the following actions were recommended:

• the electric fuel boost pump should be turned on and the fuel tank selection changed to the desired tank
• carburettor heat should be applied for 5–7 seconds to check for icing and mixture be selected to full rich
• speed should then be reduced to 87 kt
• when abeam the landing threshold, the pilot should apply the first stage of flap (20°).

From this position, the manual recommended the pilot should:

• continue to reduce the aircraft’s airspeed
• at about 78 kt and no later than on the base leg, apply the second stage of flap (35°)
• commence the turn onto the final leg at an airspeed of about 70 kt and, if required, apply full flap (55°)
• have an airspeed of about 65 kt passing over the runway threshold.

The owner’s manual also provided a suggested pre-landing checklist. This checklist included selecting carburettor heat on and the fuel mixture to full rich.

Stall characteristics

The Glasair owner’s manual indicated that the clean stall speed for the aircraft (solo) was 55 kt, while the stall speed with flaps down at gross weight was 54 kt. The manual also indicated that stall strips were mandated on the wing inboard leading edges to induce the wing roots to stall first. The stall strips were considered by the manufacturer to be the aircraft’s stall warning indicators. Consistent with the manufacturer’s position, the Department of Civil Aviation (Australia)-approved flight manual indicated that stall warning was provided by a buffet brought on by the stall strips, which produced a steady signal about 5–6 kt prior to the stall in all configurations.

Angle of bank calculations

An aircraft in a constant, level turn develops lift greater than its weight and results in increased stall speeds. An increased angle of bank in a turn further increases the stall speed.

The minimum required angle of bank in a turn is worked out geometrically using the average ground speed and radius of turn. Due to the lack of an accurate height and position for HRG when the pilot commenced the base turn, the ATSB estimated approximate angle of bank figures using estimates of the turn radius and the pilot’s recollection of the aircraft’s speed and approximate height entering the base turn. Calculations were also done for the clean aircraft configuration and full flap stall speeds at both solo and gross weights.

All estimations indicated that the reported flying speed was above the stall speed relative to the required angle of bank.

Related occurrences

ATSB investigation AO-2014-149 - Collision with terrain involving Van’s Aircraft RV-6, registered VH-TXF, near Mudgee Airport, New South Wales on 14 September 2014

On the morning of 14 September 2014, the pilot and passenger of an amateur-built Van's Aircraft RV-6, two-seat aircraft, registered VH-TXF and operated in the Experimental category, approached Mudgee Airport. The aircraft had departed Dubbo Airport, New South Wales about 25 minutes earlier.

The pilot approached from the north-west and conducted a non-standard circuit entry including an orbit to the south of the airport. Prior to turning onto the downwind leg of the circuit, the aircraft descended to about 600 ft above ground level. Witnesses stated that the pilot conducted a tight left turn onto final approach at a slow speed and low height. The witnesses also recalled hearing the aircraft’s engine ‘splutter’ and then silence during the turn, followed by a ‘rev’ followed again by silence.

The aircraft continued its high angle of bank left turn and, at about 1053, collided with terrain about 300 m south-west and short of the runway threshold. The pilot and passenger were fatally injured and the aircraft was substantially damaged.

The ATSB found that the engine failure was probably due to carburettor icing. No defects were identified that would have precluded normal engine operation prior to the accident, and uncontaminated fuel was being supplied to the engine at that time. However, the environmental conditions at the time of the accident were conducive to serious carburettor icing at descent power, and the pilot-operated carburettor heat control was found in the OFF position.

ATSB investigation AO-2015-077 - Collision with terrain involving a Robinson R44, registered VH-VOH, 130 km east of Alice Springs, Northern Territory on 14 July 2015

On 14 July 2015, the pilot of a Robinson R44 helicopter, registered VH-VOH, was conducting aerial mustering operations on a property, about 70 NM (130 km) east of Alice Springs. At about 1300 Central Standard Time^[10], the pilot was mustering cattle along a creek system. The helicopter was at about 50 ft above ground level, when the pilot slowed the helicopter to an airspeed of about 40 kt. The pilot felt a small vibration, and initially thought it was due to loose tape on the main rotor blade. The pilot looked for a suitable landing site, but the vibration increased significantly.

As the helicopter descended, the pilot manoeuvred the helicopter through a gap between trees, and pushed the cyclic forward to maintain airspeed. The pilot lowered the collective and noticed the engine seemed to go very quiet. The low rotor revolutions per minute warning horn sounded. The pilot made a radio call to advise another pilot operating nearby that the helicopter was going down. The pilot then flared the helicopter to try to cushion the landing impact. The right skid touched down first, and the helicopter rolled onto its right side. The pilot sustained minor injuries and the helicopter was substantially damaged.

According to the Carburettor Icing Probability chart, the conditions indicated a high probability of serious carburettor icing at descent power.

__________

3. A cruise propeller will achieve maximum efficiency at 75 per cent power during the cruise. Take-off and climb performance will not be as good as with a climb or constant-speed propeller.
4. The ATSB could not establish the exact amount of fuel in the main or the header (or auxiliary) tanks prior to the loss of power. That fuel was estimated based on refuelling records and operational documentation from previous flights.
5. Routine aerodrome weather report issued at fixed times, hourly or half-hourly.
6. Available at www.ga.gov.au/.
7. Carburettor ice is formed when the normal process of vaporising fuel in a carburettor cools the carburettor throat so much that ice forms from the moisture in the airflow, which can restrict the airflow and interfere with the operation of the engine.
8. The elevations of Campbelltown (Mount Annan) and Wedderburn Airport are 368 ft and 850 ft respectively.
9. Dewpoint is the temperature at which water vapour in the air starts to condense as the air cools. It is used, among other things, to monitor the risk of aircraft carburettor icing or likelihood of fog at an aerodrome.
10. Central Standard Time (EST) was Coordinated Universal Time (UTC) +9.5 hours.

Safety analysis

Introduction

Following the loss of engine power, the pilot was unable to control the aircraft’s descent to an appropriate forced landing area before colliding with the ground. Due to serious injuries during the impact sequence, the pilot was unable to recall anything about the descent following the loss of power. No witnesses saw the aircraft during the base and final turns, nor the loss of control and collision with terrain.

ATSB on- and off-site specialist analyses of the engine, airframe and selected items and components did not identify any issues that could have contributed to the accident. There was also no evidence that pilot fatigue or recent stressful life events were contributory.

This analysis will examine a number of potential factors that can account for the loss of power and discuss a number of possible reasons for the subsequent loss of control.

Potential reasons for the loss of control following the engine power loss

Aerodynamic stall

The ATSB considered the possibility that the aircraft entered an aerodynamic stall prior to colliding with terrain; however, there was not enough evidence to support that hypothesis. Due to the noise abatement procedures in place at Wedderburn Airport, the circuit was reported tighter-than-normal and the pilot conducted a continuous turn from base onto final approach. Generally, a tighter‑than‑normal turn would require an increased angle of bank, increasing the stall speed. In addition, the pilot reported entering the base turn at a speed higher-than-normal.

Based on the aircraft’s estimated airspeed and angle of bank, it was determined that the aircraft remained above the stall speed throughout the base turn. Therefore, an aerodynamic stall was ruled out as a possible explanation for the loss of control.

Fuel starvation

The possibility that the header tank was selected for the duration of the flight was also considered. If full prior to take-off and used for the flight, the header tank may have run dry during the turn on to the final leg of the circuit, contributing to the loss of power and then of control.

On-site examination of the fuel selector found that it was in the mains tank position. However, this evidence is somewhat unreliable due to the potential influence of impact forces during the impact sequence.

The pilot stated that the header tank was only selected for engine run-ups and when, due to the length of time between engine starts, the fuel in the header tank might have been considered ‘stale’. The pilot stated that they always took off and flew on the mains fuel tank.

The header tank was found separated from the airframe and all fuel lines and plumbing had been fractured by impact forces. Whether the header tank was full prior to take-off, and the quantity of fuel remaining in the header tank following the impact, could not be established. In combination, these factors prevented a determination of whether the header tank was selected for the flight and ran dry during the turn onto final.

Carburettor icing

The witness accounts of two engine surges before there was silence, and the low-speed rotational signatures on the propeller, indicated that the engine lost power prior to the collision with terrain. No defects were identified that would have precluded normal engine operation prior to the power loss.

Due to the level of disruption from the impact sequence, it could not be determined if the carburettor heat was selected at the time of the accident. However, the pilot stated that they did not use carburettor heating as they believed the conditions that day were too mild for carburettor icing. The pilot also stated that the downwind leg of the circuit was conducted with the engine at idle or near idle in an effort to reduce airspeed.

In this case, the meteorological conditions around the time of the accident were conducive to serious carburettor icing at all power settings (appendix A). In combination with the reported low power setting and lack of carburettor heat, there was an elevated risk of ice accreting in the throat of the carburettor. Ice in the throat of the carburettor would have reduced the flow of air to the engine and, without pilot action to correct the situation, the engine would have stopped.

In the absence of any contrary evidence, the ATSB concluded that it was probable the loss of power during the final turn was a result of carburettor icing. The loss of power in this position would suggest that any carburettor icing remained undetected by the pilot throughout the initial legs of the circuit.

Findings

From the evidence available, the following findings are made with respect to the loss of power and collision with terrain involving Glasair SH-2FT, registered VH-HRG, near Wedderburn Airport, New South Wales on 19 July 2015. These findings should not be read as apportioning blame or liability to any particular organisation or individual.

Contributing factors

• The meteorological conditions at the time of the accident, combined with the aircraft’s low engine power setting during the downwind and base legs of the circuit, probably resulted in carburettor ice formation.
• The lack of application of carburettor heat increased the likelihood of the formation of carburettor icing and, as a result, the engine losing power.

Sources and submissions

Sources of information

The sources of information during the investigation included:

• the pilot of VH-HRG
• a number of the members of the NSW Sport Aircraft Club, Wedderburn Airport
• the aircraft maintainer
• the aircraft manufacturer
• the Civil Aviation Safety Authority
• the New South Wales State Coroner
• the New South Wales Police Force
• the Bureau of Meteorology.

References

Australian Transport Safety Bureau, (2001). Melting moments: Understanding carburettor icing, Educational fact sheet.

Partie, E. and Peterson, B.D. (2009). Combating carb ice. AOPA Air Safety Foundation - Safety Brief, SB09-10/09.

United Kingdom Civil Aviation Authority. (2013). Piston engine icing. Safety sense leaflet No. 14 – January 2013.

Submissions

Under Part 4, Division 2 (Investigation Reports), Section 26 of the Transport Safety Investigation Act 2003 (the Act), the ATSB may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. Section 26 (1) (a) of the Act allows a person receiving a draft report to make submissions to the ATSB about the draft report.

A draft of this report was provided to the pilot of VH-HRG and the Civil Aviation Safety Authority.

A submission was received from the pilot. The submission was reviewed and where considered appropriate, the text of the report was amended accordingly.

Appendices

Appendix A – Civil Aviation Safety Authority carburettor icing probability charts

Figure A1: Civil Aviation Safety Authority (CASA) carburettor icing-probability chart annotated with Campbelltown (in yellow) and Camden (in purple) temperature information at 1630

Figure A2: CASA carburettor icing-probability chart annotated with Campbelltown (in yellow) and Camden (in purple) temperature information at 1700

Purpose of safety investigations The objective of a safety investigation is to enhance transport safety. This is done through: identifying safety issues and facilitating safety action to address those issues providing information about occurrences and their associated safety factors to facilitate learning within the transport industry. It is not a function of the ATSB to apportion blame or provide a means for determining liability. At the same time, an investigation report must include factual material of sufficient weight to support the analysis and findings. At all times the ATSB endeavours to balance the use of material that could imply adverse comment with the need to properly explain what happened, and why, in a fair and unbiased manner. The ATSB does not investigate for the purpose of taking administrative, regulatory or criminal action. Terminology An explanation of terminology used in ATSB investigation reports is available here. This includes terms such as occurrence, contributing factor, other factor that increased risk, and safety issue. Publishing information  Released in accordance with section 25 of the Transport Safety Investigation Act 2003 Published by: Australian Transport Safety Bureau © Commonwealth of Australia 2016 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this publication is licensed under a Creative Commons Attribution 3.0 Australia licence. Creative Commons Attribution 3.0 Australia Licence is a standard form licence agreement that allows you to copy, distribute, transmit and adapt this publication provided that you attribute the work. The ATSB’s preference is that you attribute this publication (and any material sourced from it) using the following wording: Source: Australian Transport Safety Bureau Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

Safety summary

What happened

After carrying out track resurfacing and tamping works at Allandale, NSW, the Australian Rail Track Corporation Network Control Officer (NCO) granted permission for a Leighton Swietelsky Rail Joint Venture (LS Rail) tamper and ballast regulator to travel in convoy and stop before signal BN 87U near Greta. The tamper led the ballast regulator, with the Traffic Officer (TO) travelling in the rear vehicle. Near Greta, the lead track vehicle operator (operator) contacted the TO advising he was approaching a signal displaying a stop (red) indication, but that he was unable to identify the signal number. Both the TO and the operator thought that this signal was BN 87U.

The TO contacted the NCO for further instructions. During the discussion, both the operator (lead vehicle) and the TO (rear vehicle) independently established that the signal ahead was BN 83U. At about this time the tamper stopped near signal BN 83U, but the rear vehicle was still approaching from behind. The TO continued to talk to the NCO about clearing signal BN 83U, so they could continue on to signal BN 87U in preparation for stabling the vehicles at Greta.

While the TO was talking with the NCO, the TO sensed that the ballast regulator (rear vehicle) was not slowing. He looked ahead to see the gap was closing between the two vehicles. The TO noticed that the ballast regulator operator was looking at him and the TO hand gestured in a forward motion. The operator interpreted the hand gestures to mean ‘keep going’. Instead, the TO was indicating that the operator ‘look forward’. Soon after, the TO called out for the operator to ‘STOP’. The operator turned and quickly realised that the tamper was closer than he had realised and made a full brake application. However, there was insufficient time to stop before colliding with the rear of the tamper.

The collision saw the ballast regulator ride above the tamper coupler and remain supported on the tamper deck. The ballast regulator also derailed the lead axle. The tamper operator sustained a minor injury and there was moderate damage to both vehicles.

What the ATSB found

The ATSB found that the Traffic Officer’s telephone conversations distracted the ballast regulator operator, drawing his attention away from the driving task and the location of the tamper ahead. Although the operator was aware of the slowing tamper ahead, he did not maintain a safe distance between the vehicles. When the operator became aware of the stationary vehicle ahead, there was insufficient time for the operator to react and bring the vehicle to a stop before colliding with the tamper.

What's been done as a result

LS Rail has taken action to mitigate against the risk of further collisions between track maintenance vehicles, including requiring greater attention to worksite planning, briefing, communications, hazard identification and route knowledge.

Safety message

When travelling in convoy, track vehicle operators should remain alert, focused on the driving task and maintain a safe braking distance from other vehicles.

The occurrence

On 14 July 2015, a tamper and ballast regulator had carried out track resurfacing and tamping works near Allandale (New South Wales).

The section of track between Allandale and Greta was on the Australian Rail Track Corporation (ARTC) Hunter Valley corridor and was part of the Middle Hunter Train Control area. (Figure 1).

At about 1712^[1], the Traffic Officer (TO^[2]) contacted the worksite Possession Protection Officer who granted permission for both track machines to travel from Allandale on the Up Main track prior to stabling on the Up Relief track at Greta.

Figure 1: Location map

The tamper and ballast regulator had been working near Allandale before travelling in convoy to Greta.

Source: ARA/Geoscience Australia; annotated by ATSB

The NCO authorised the movement to travel to and stop at signal BN 87U at Greta (Figure 2). The TO then contacted the tamper track vehicle operator (operator) by radio, advising that they were now authorised to travel up to signal BN 87U.

At about 1714 the vehicles departed the worksite near Allandale and travelled in convoy with the tamper leading and the ballast regulator following. In accordance with Rule ANWT 316, the TO was travelling in the rear vehicle, the ballast regulator.

Figure 2: Diagram showing track layout near Greta, NSW

The tamper and ballast regulator were authorised to travel to signal BN 87U before stabling on the Up Relief track at Greta. However, signal BN 83U (about 1450m before BN 87U) was showing a stop indication, requiring both vehicles to stop. Source: Graham Vincent, Track and Signal, annotations by ATSB

As the vehicles travelled towards Greta, the operator of the lead vehicle observed a signal ahead that was displaying two red lights (a stop indication). The operator slowed the vehicle and prepared to stop, as he had not been authorised to pass any signals displaying a stop indication.

While approaching the signal, the lead vehicle operator was unable to see the signal identification sign due to the fading daylight conditions. While the operator assumed it was signal BN 87U, he contacted the TO by radio (at about 1717) and queried whether the signal immediately ahead of his position was BN 87U. The operator received a response from the TO, stating that he too thought the signal was BN 87U. The lead vehicle operator acknowledged the message and continued to approach the signal with the intention of stopping.

At about 1719, the TO contacted the NCO by telephone, to gain permission for both vehicles to pass the signal they had assumed to be BN 87U.

Meanwhile, the lead vehicle was now close enough for the operator to identify the signal, not as BN 87U but as BN 83U. Signal BN 83U was an intermediate signal located about 1450m before signal BN 87U at Greta. The operator radioed the TO, confirmed that the signal was BN 83U and asked for further instructions.

At that point in time, the TO was talking to the NCO by phone, so he could not respond to the radio communication. Consequently, the operator of the following vehicle replied to the radio communication with a ‘copy that’ response. Both operators then awaited instruction from the TO. The lead vehicle continued slowing to stop at signal BN 83U while the following vehicle continued to approach from behind.

As the ballast regulator approached the tamper, the TO continued talking to the NCO. The NCO confirmed that the signal at stop ahead was BN 83U and not BN 87U (track circuits confirmed the tamper and ballast regulator were approaching BN 83U). Having heard the radio communication from the lead vehicle operator, the TO agreed that they were approaching signal BN 83U.

While talking with the NCO, the TO sensed that their vehicle was not slowing and the gap to the vehicle ahead was closing. The TO then looked at the operator of their vehicle and noticed that he was looking back at him, rather than the track ahead.

Still talking on the phone, the TO reacted by using hand gestures to draw the operator’s attention to the vehicle ahead and that it had stopped. Although the operator had started to slow the vehicle, he assumed that the TO had received instruction from the NCO and interpreted the hand gestures to mean ‘keep going’.

It was about 1720 and the NCO was advising the TO that 112 points at Greta were set for the mainline to allow the vehicles to enter and stable on the Up Relief track.

At about the same time, the TO realised a collision was imminent and quickly called out ‘STOP’ to the operator. The operator immediately reacted by applying the service brake, however the brake application was too late and the ballast regulator collided with the rear of the tamper.

The NCO reported hearing the ‘STOP’ exclamation over the phone, followed by a ‘rumbling’ sound. About two seconds later, the phone connection terminated.

The collision occurred on a slight downhill grade, just before signal BN 83U (Figure 3). In the collision sequence, the front of the ballast regulator rode above the coupler on the tamper and came to rest partially on the rear deck of the tamper. The ballast regulator also derailed its lead axle.

Figure 3: Signal BN 83U and the Plasser tamper

Main image: Shows the location of the Plasser tamper following the collision, having previously stopped near signal BN 83U. 
Inset: Signal BN 83U and its corresponding signal identification sign.
Source: Australian Rail Track Corporation and LS Rail.

Suspecting that something was not right or there had been an accident, the NCO attempted four times to call the TO to investigate. The NCO also tried to call the operator of the lead vehicle (tamper) without success. About five minutes after the disconnected phone call, the TO called the NCO to report the collision between the two vehicles. He advised that no persons were injured and that they would place protection measures on track to protect the accident site.

The ballast regulator was fitted with a data recorder that confirmed the collision speed was around 30 km/h. Both vehicles received moderate damage and obstructed the Up Main line.

The vehicles were cleared from the accident site the next day and damage to the track was minor.

The tamper operator later reported that he had a minor injury. The four employees (two operators, the TO and the NCO) were tested for the presence of drugs or alcohol. All returned negative results.

__________

1. Time shown as Eastern Standard Time (EST).
2. The Traffic Officer also performs the role of Protection Officer (PO).

Context

Location

Greta is located at the 210.660 km mark^[3] on the ARTC Hunter Valley Network in New South Wales. The point of collision was near the 208.135 km, about 2.5 km southeast of Greta and about 2.4 km from Allandale. The track alignment at this location is on a sweeping curve with a 1:103 falling gradient.

Track maintenance vehicles

The ARTC had appointed LS Rail^[4] as the principal contractor for the track maintenance works. The effective control and management of the vehicles were the responsibility of LS Rail.

The track maintenance vehicles were a:

• Tamper – Plasser model 08/275 3S (fleet number DR 73114)
  • The tamper weighed about 64 t, was 20.7 m long, and was restricted to a maximum speed of 80 km/h.
• Ballast regulator – Plasser model PBR 203 (fleet number BX-045)
  • The ballast regulator weighed about 22.5 t, was 11 m long, and was restricted to a maximum speed of 50 km/h.

Both vehicles were fitted with flashing lights. The ballast regulator’s lights operate continuously and the tamper’s operate in work mode and during brake applications. The flashing lights were operating on the vehicles at the time of the collision.

An inspection of the ballast regulator brakes at the collision site found all four brake shoes were poorly adjusted. Although adjustment was required to reduce the gaps between the brake shoes and wheel treads to conform to the maintenance specification, the brakes remained operable, with the larger than specified gaps expected to have produced only a small increase in stopping distance.

Traffic Officer

The ARTC Network Rules (NSW), Work on Track – Track Vehicles ANWT 316 required the safe-working qualified worker to travel in the rear vehicle of the convoy. Duties of the Traffic Officer (TO) included notifying the NCO when track vehicles had entered or cleared a track section or running line. The rules stated that when track vehicles were travelling in convoy they:

• must travel as closely as is safely practicable
• must maintain effective communication, and
• must close up if the leading vehicle stops.

In accordance with Rule ANWT 316, the TO was travelling in the ballast regulator (the rear vehicle). The TO regularly communicated with the NCO and the tamper operator (in the lead vehicle) about the safe working arrangements.

Before the vehicles departed the Allandale worksite, the TO briefed both operators that they would be travelling up to and stop before signal BN 87U near Greta. The TO did not mention the presence or location of BN 83U signal even though this signal was shown on the Worksite Protection Plan. This omission may have lead the crews of both vehicles to believe that there were no other network-controlled signals before signal BN 87U at Greta.

Safe working arrangements

The works carried out by the tamper and ballast regulator near Allandale were part of broader track maintenance activities. The safe working method applied was a Local Possession Authority (LPA).

An LPA closes a defined portion of track for a specified period. A Possession Protection Officer (PPO) is then responsible for coordinating all the works and track machine movements. Trains, other than those authorised by the PPO, are excluded from the LPA section of track.

In this case, the TO had contacted the PPO and requested permission to travel to Greta. Permission was granted subject to the TO liaising with the NCO with regard to passing lineside signals. The TO subsequently communicated with the NCO to facilitate travel from the worksite to Greta. These actions were all consistent with the ARTC rules and procedures.

Track vehicle operator distraction

Distraction can be described as a type of inattention, where a person’s attention is diverted by a particular event or object. Operator distraction has been more specifically defined as ‘the diversion of attention away from activities critical for safe driving, toward a competing activity (occurring) voluntarily or involuntarily’^[5].

The ARTC Network Procedure ANPR 748 advises track vehicle operators to remain vigilant^[6]. That is, an operator should not engage in any activity that distracts from their attention to safety. In this case, the ballast regulator operator was aware that the vehicles would be stopping at signal BN 87U near Greta before stabling overnight. Having overheard a conversation between the TO and the tamper operator, he was aware that the tamper ahead was approaching a stop signal and that the signal was BN 83U.

At about the same time, the TO had commenced a conversation with the NCO, to facilitate the continued travel and stabling of the track machines at Greta. Expecting further instructions, the ballast regulator operator had diverted his attention to the conversation between the TO and the NCO. He also misinterpreted the TO’s hand gesturing before his attention was brought back to the driving task and looking ahead.

When the operator became aware of the stationary vehicle ahead, he immediately applied the brakes. However, given the speed of the vehicle, there was insufficient time for the ballast regulator to stop before it collided with the tamper.

Analysis of data log information from the ballast regulator showed the travel speed was 30 km/h, about two minutes before the collision. The collision was recorded as a sudden deceleration from 30 km/h to 0 km/h in less than 2 seconds.

A graph of the data also showed vertical accelerations of up to 2 g at the time of the collision – consistent with the ballast regulator riding above and then then falling onto the rear platform of the tamper (Figure 4).

Figure 4: Detail of ballast regulator resting on the rear deck of the tamper.

The ballast regulator remained overlapping and resting on the tamper’s rear deck where ancillary equipment was damaged during the collision. The ballast regulator suffered a broken front towing ‘A’ frame and a deformed front chassis cross beam. Image Source: The Australian Rail Track Corporation.

__________

3. All track distances in this report are referenced from the Sydney Central Railway Station.
4. LS Rail is a joint venture between Leighton Contractors and Swietelsky Australia.
5. Regan, M.A., Hallett, C. & Gordon, C.P. (2011). Driver distraction and driver inattention: Definition, relationship and taxonomy. Accident Analysis and Prevention, 43, 1771-1781.
6. ARTC ANPR 748, Transferring Track Vehicles, Network Procedures, Track vehicle operators, Section 4.

Findings

From the evidence available, the following findings are made with respect to the collision between ballast regulator (BX-045) and tamper (DR 73114) about 2.5 km southeast of Greta, on the Middle Hunter Valley rail corridor in New South Wales.

These findings should not be read as apportioning blame or liability to any particular organisation or individual.

Safety issues, or system problems, are highlighted in bold to emphasise their importance. A safety issue is an event or condition that increases safety risk and (a) can reasonably be regarded as having the potential to adversely affect the safety of future operations, and (b) is a characteristic of an organisation or a system, rather than a characteristic of a specific individual, or characteristic of an operating environment at a specific point in time.

Contributing factors

• The Traffic Officer’s telephone conversations distracted the ballast regulator operator, drawing his attention away from the driving task and the location of the tamper ahead.
• When the operator became aware of the stationary vehicle ahead, there was insufficient time for the ballast regulator operator to react and bring the vehicle to a stop before it collided with the tamper.
• The operator of the ballast regulator interpreted the Traffic Officer’s forward pointing hand gesture as ‘keep going’ rather than a gesture intended to draw attention to the vehicle stopped ahead.
• The Traffic Officer’s instruction and communication to the regulator operator by way of an ambiguous hand gesture was not explicit in its intended message.

Other factors that increased risk

• The briefing given by the Traffic Officer before the movement of vehicles at the Allandale worksite did not mention controlled signal BN 83U, later leading the operators to believe that the signal showing a stop indication was signal BN 87U at Greta.

Additional safety actions

Whether or not the ATSB identifies safety issues in the course of an investigation, relevant organisations may proactively initiate safety action in order to reduce their safety risk. The ATSB has been advised of the following proactive safety action in response to this occurrence

Proactive safety action taken by Leighton Swietelsky Rail Joint Venture (LS Rail)

Action number:RO-2015-012-NSA-011

LS Rail have advised that actions undertaken to mitigate against the risk of further collisions between track maintenance vehicles include:

• Worksite Planning

Amend the existing planning procedure for working on multiple sites where the Traffic Officer undertakes a detailed review of the planning documents before travelling to the next worksite.

• Communication of safe working arrangements

Amend communication procedures for safe working arrangements to ensure track vehicle operators are fully aware of the travel route using track diagrams that identifies key features including controlled signals.

• Work hazard identification

Undertake a review of Pre-start/Pre-work Briefs to ensure the identification of all relevant high-risk work/hazards.

• Workforce awareness briefing

Provide awareness briefings to the LS Rail workforce to ensure the contributing factors of this occurrence is well understood and to reinforce the requirements for detailed pre-work briefings and the effects of distraction in the workplace.

• Route knowledge

Establish a LS Rail route knowledge database and the assessment criteria for Traffic Officers.

Response to Proactive safety action taken by LS Rail

Action number: RO-2015-012-NSA-011

ATSB comment in response

The ATSB is satisfied that the actions advised by Leighton Swietelsky Rail Joint Venture will reduce the risk of further collisions between track maintenance vehicles.

Sources and submissions

Sources of information

The sources of information during the investigation included the:

• Leighton Swietelsky Rail Joint Venture,
• The Australian Rail Track Corporation,
• The Office of the National Rail Safety Regulator.

References

The Australian Rail Track Corporation, Rules (NSW), ANGE 204, Network Communication, General Rules.

The Australian Rail Track Corporation, Rules (NSW), ANPR 748, Transferring Track Vehicles, Network Procedures, Track vehicle operators.

The Australian Rail Track Corporation, Rules (NSW), Network, ANWT 316 Work on Track – Track Vehicles.

Regan, M.A., Hallett, C. & Gordon, C.P. (2011). Driver distraction and driver inattention: Definition, relationship and taxonomy.Accident Analysis and Prevention, 43, pp.1771-1781.

Submissions

Under Part 4, Division 2 (Investigation Reports), Section 26 of the Transport Safety Investigation Act 2003 (the Act), the Australian Transport Safety Bureau (ATSB) may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. Section 26 (1) (a) of the Act allows a person receiving a draft report to make submissions to the ATSB about the draft report.

A draft of this report was provided to:

• Leighton Swietelsky Rail Joint Venture,
• The Australian Rail Track Corporation,
• The Office of the National Rail Safety Regulator
• The operator of the ballast regulator,
• The operator of the tamper.

Submissions were received from:

• Leighton Swietelsky Rail Joint Venture,
• The Australian Rail Track Corporation,
• The Office of the National Rail Safety Regulator.

The submissions were reviewed and where considered appropriate, the text of the report was amended accordingly.

Purpose of safety investigations The objective of a safety investigation is to enhance transport safety. This is done through: identifying safety issues and facilitating safety action to address those issues providing information about occurrences and their associated safety factors to facilitate learning within the transport industry. It is not a function of the ATSB to apportion blame or provide a means for determining liability. At the same time, an investigation report must include factual material of sufficient weight to support the analysis and findings. At all times the ATSB endeavours to balance the use of material that could imply adverse comment with the need to properly explain what happened, and why, in a fair and unbiased manner. The ATSB does not investigate for the purpose of taking administrative, regulatory or criminal action. Terminology An explanation of terminology used in ATSB investigation reports is available here. This includes terms such as occurrence, contributing factor, other factor that increased risk, and safety issue. Publishing information  Released in accordance with section 25 of the Transport Safety Investigation Act 2003 Published by: Australian Transport Safety Bureau © Commonwealth of Australia 2016 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this publication is licensed under a Creative Commons Attribution 3.0 Australia licence. Creative Commons Attribution 3.0 Australia Licence is a standard form licence agreement that allows you to copy, distribute, transmit and adapt this publication provided that you attribute the work. The ATSB’s preference is that you attribute this publication (and any material sourced from it) using the following wording: Source: Australian Transport Safety Bureau Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

What happened

Early in the morning on 15 July 2015, the crew of a Boeing 737800, registered VHYIU and operated by Virgin Australia International, prepared for a flight from Christchurch, New Zealand, to Brisbane, Queensland. The captain was the pilot flying and the first officer (FO) was the pilot monitoring.^[1]

The flight usually departed Christchurch at 0650 New Zealand Standard Time (NZST), but the scheduled departure was delayed on this occasion to 0815, due to a crew change. The crew change, which had been planned by the operator during the evening prior, required the captain of the flight to fly to Christchurch as a passenger that morning. The captain arrived in Christchurch at about 0730 and proceeded directly to the waiting aircraft.

Meanwhile, the FO had arrived at the airport at about 0700 and checked the flight plan package,^[2] including the flight plan, weather and NOTAMs.^[3] The FO then ordered the required amount of fuel for the flight, and proceeded to the aircraft. The FO had noticed two NOTAMs dealing with runway works at Christchurch, but assessed that neither NOTAM would affect the flight.

After arriving at the aircraft, the FO commenced normal pre-flight duties. As part of preparation for the flight, the FO prepared the take-off reference data for departure from the runway 02/A6 taxiway intersection, anticipating that the full length of the runway would be available. The FO used the 24K (24,000 lb) engine thrust rating^[4]/flaps 5 take-off reference data from the runway 02/A6 intersection table in the operator’s Airport Analysis Manual (AAM).

The captain went straight to the aircraft and met the FO. The captain then checked the flight plan, fuel load and weather information, and conducted a pre-flight inspection of the aircraft. The captain did not read the NOTAMs but was advised by the FO that there was nothing significant. The aircraft was pushed back from the gate at about 0815.

At about the time the aircraft was pushed back from the gate, air traffic control (ATC) advised the crew that there was a change in the ATIS^[5] and that runway 02 was operating at a reduced length. The reduction in runway length was associated with works in progress (WIP) that reduced the runway length available from 3,288 m to 1,920 m, with the northern 1,368 m of the runway closed (Figure 1).

Figure 1: Christchurch Airport showing runway 02 works in progress - north

Source: CAA NZ - annotated by ATSB

Before starting the engines, the crew reviewed the take-off reference data considering the revised ATIS and the reduced runway length (due to the runway works). The crew again referred to the AAM, expecting to find inserted yellow pages that provided take-off reference data to be used while runway works were in progress (see section titled Airport analysis manual). The crew found that there were no yellow pages available for Christchurch.

In the absence of reduced runway length data related to the runway works (yellow pages), the crew elected to use full thrust during the departure, and commence their take-off from the threshold of runway 02. The crew then used the 26K (26,000 lb – full rated thrust)/flaps 5 take-off reference data from the AAM that was based upon the full length of the runway being available. The FO determined the amended take-off reference speeds from the AAM, and in accordance with company procedures, the figures were cross-checked by the captain.

During taxi and while lining up on the runway, the crew did not see any personnel, equipment or obstructions on the runway. At 0827, the aircraft departed without incident.

Following departure, the crew heard ATC advise the crew of an aircraft that was inbound to Christchurch, that the full length of the runway would be available for their arrival. This prompted the captain to review the NOTAMs that had earlier been reviewed by the FO. The captain found NOTAM B3805/15 NZCH (Figure 2) referring to runway works at Christchurch, which had relevance to their flight.

From the NOTAM, the captain ascertained that the runway length at the time of their departure was reduced to 1,920 m due to WIP. The NOTAM was effective from 14 July 2015 at 2000 UTC (15 July 2015 at 0800 NZST) until 15 July 2015 at 0225 UTC (1425 NZST). The captain also noticed that there was an associated relevant company remark (immediately following the NOTAM and highlighted in Figure 2) regarding a requirement to request On-Board Performance Tool (OPT)^[6] take-off reference data during works in progress.

Figure 2: NOTAM B3805/15 dealing with runway 02 reduced length (and closure of runway 20) due to works in progress^[7]

Source: Aircraft operator, highlight added by ATSB

While still en route, the crew contacted company flight dispatch staff and requested OPT take-off reference data that should have been used during operations while runway works were in progress. The OPT take-off reference data revealed that different take-off reference speeds should have been used under those circumstances (Table 1).

The flight continued uneventfully to Brisbane. On arrival in Brisbane, the captain notified relevant airline staff of the occurrence.

Table 1: Differences between OPT take-off reference data and the data used by the crew

-------------

Take-off reference speeds

Take-off reference speeds or V speeds assist pilots in determining when a rejected take-off can be initiated, and when the aircraft can rotate, lift-off and climb. The definitions of V speeds can be quite complex, but in broad terms: V1 is often referred to as the critical engine failure speed or decision speed.

• V1 is the maximum speed at which a rejected take-off can be initiated. If an engine failure is detected above V1, the take-off should be continued.
• VR is the speed at which the rotation of the aircraft is initiated to the take-off attitude. The speed cannot be less than V1, and takes into account a number of other critical speeds that relate to aircraft performance and handling.
• V2 is often referred to as the take-off safety speed. It is the minimum speed at which a transport category aircraft complies with those handling criteria associated with climb, following an engine failure. V2 is normally obtained by factoring other critical speeds, to provide a safe margin with respect to aircraft controllability.

-------------

Airport analysis manual (AAM)

The crew commented that a recent change in the holder of the Air Operator’s Certificate (AOC) from Virgin Australia Airlines – New Zealand (VANZ) to Virgin Australia International Airlines (VAI) had seen numerous procedural changes. The changes related primarily to the integration of VANZ and VAI procedures, to establish consistency across the company’s operations. With reference to this incident, the crew commented that prior to the change in AOC holder, AAMs included yellow pages that provided take-off reference data to be used when operating from a reduced length runway (such as during runway works). When the crew discovered that the departure runway was operating at a reduced runway length due to runway works, they initially referenced the AAM with an expectation of finding a relevant yellow page, but without that page, they elected to use 26K (full rated thrust)/full runway length data instead.

The operator’s draft report dealing with the incident commented that since the transition from VANZ to VAI, yellow pages in the AAM dealing with runway works have been removed. A note has been added to relevant NOTAMs regarding the requirement to request OPT reference data.

The flight crew commented that while a two-day training course was provided to transition crews from VANZ to VAI, the scope of the training was limited, and crews were continuing to discover procedural variations in the months following the transition. The operator advised that the course included a section on aircraft performance as well as training in AAM use. As part of their investigation, the operator reviewed relevant material presented to the flight crew and found no deficiencies, but they could not assess the efficacy of the training.

Flight plan package

The absence of yellow pages in the AAM aside, the crew expected that if OPT take-off reference data was required, it would be provided with the flight plan package. In the experience of the crew, OPT take-off reference data was usually provided with the flight plan package when required, without specifically being requested by the crew. The only reference to the requirement for the crew to request OPT take-off reference data on this occasion was a remark at the end of the NOTAM dealing with the runway works on that day. There were no other relevant prompts in the package that might have alerted the crew to the requirement to request OPT take-off reference data. Contrary to the expectations of the crew, flight dispatch staff considered that it was the responsibility of the crew to request OPT take-off reference data, when it was required.

Flight crew operational notices

During the positioning flight to Christchurch, the captain reviewed the operator’s Flight Crew Operational Notices (FCON),^[8] including the notices relevant to Christchurch. Even though there were NOTAMs in place addressing runway works, there was no reference to any runway works in the Christchurch FCON entry. In contrast, the FCON entry for Cairns, Queensland (directly before the Christchurch entry) included reference to runway works at Cairns. The Cairns entry included a statement that during the works, AAM take-off and landing data was not valid. The entry also stated:

There are no scheduled departures during the works period, however if take-off data is required request OPT …

If the FCON had included a similar reference to the runway works at Christchurch, it may have prompted the crew to review the possible implications of the runways works more closely prior to departure.

The operator advised that runway works at Christchurch were not addressed in the FCON because the anticipated time of the works referred to in NOTAM 3528/15 NZCH (see following section dealing with NOTAMs), did not conflict with the normal departure time for the flight. The operator’s investigation found that for the Cairns entry, there were also no scheduled departures during the works period, but it was close to scheduled aircraft arrival times, which required associated landing data. The operator’s draft investigation report stated that the temporary landing data for Cairns would potentially have been required daily during the works period, whereas for Christchurch the data was only required on an ad hoc basis.

The operator’s investigation found that while the aim was to avoid repeating information in a NOTAM remark and the FCON, it was not clear which was the primary source of information for the flight crew.

Notice to Airmen (NOTAM)

Pre-flight NOTAM review

The operator’s procedures required that both crew members review the relevant NOTAMs prior to a flight. The crew commented that in practice, review of flight plan material including the NOTAMs, is typically done as a team. Following a review of the material, the crew members discuss factors of relevance as part of their preparation for the flight.

Normally, the captain and FO would have met in a crew room facility to discuss the flight, before proceeding to the aircraft. However, to minimise the delay, the flight crew met at the aircraft on this occasion. The captain had limited recent familiarity with Christchurch and was unaware of the runway works, until advised by ATC during push-back.

Despite the arrangements that required the captain to travel to Christchurch during the morning of the flight, and the associated late departure, the flight crew reported that they did not feel rushed as they prepared for the flight.

NOTAMs

In addition to NOTAM 3805/15 (Figure 2), a second NOTAM B3528/15 NZCH (Figure 3), stated that works were expected to commence at 0930 UTC (2130 NZST) in the evening, and finish at 1630 UTC (0430 NZST) each morning, with a NOTAM to be issued advising of activation times. The departure time of the flight during which the incident occurred, was outside those times (both the normally scheduled and delayed departure times). As the scheduled departure time also fell outside the times specified in NOTAM 3805/15, this may have influenced the FO to expect that the works would not affect their departure.

Figure 3: NOTAM (B3528/15) dealing with runway works, identifying the expected times of runway works, and advising that activation times would be notified by separate NOTAM with the location of the works (north or south)^[9]

Source: Aircraft operator

Safety message

The operator’s investigation found that the ability to reject the take-off or maintain obstacle clearance safely in the event of an engine failure was compromised by the use of the incorrect take-off reference speeds. Inaccurate take-off reference data has potentially serious consequences. ATSB Aviation Research and Analysis Report AR-2009-052 (Take-off performance calculation and entry errors: A global perspective) documents a number of accidents and incidents where take-off performance data was inaccurate. The report analyses those accidents and incidents, and concludes:

… it is imperative that the aviation industry continues to explore solutions to firstly minimise the opportunities for take-off performance parameter errors from occurring and secondly, maximise the chance that any errors that do occur are detected and/or do not lead to negative consequences.

This incident highlights the importance of a consistency in the expectations of flight crew and the services provided by an operational support system. A disconnect on this occasion substantially diminished the defences that might otherwise have prevented the incident. In a broader sense, the incident provides an example of how changed procedures can introduce latent procedural deficiencies or misunderstandings. Robust crew training and follow-up standardisation are critical to the safe and effective introduction of new or revised operational procedures.

Aviation Short Investigations Bulletin Issue 46

About this report Decisions regarding whether to conduct an investigation, and the scope of an investigation, are based on many factors, including the level of safety benefit likely to be obtained from an investigation. For this occurrence, a limited-scope, fact-gathering investigation was conducted in order to produce a short summary report, and allow for greater industry awareness of potential safety issues and possible safety actions. Publishing information  Released in accordance with section 25 of the Transport Safety Investigation Act 2003 Published by: Australian Transport Safety Bureau © Commonwealth of Australia 2016 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this publication is licensed under a Creative Commons Attribution 3.0 Australia licence. Creative Commons Attribution 3.0 Australia Licence is a standard form licence agreement that allows you to copy, distribute, transmit and adapt this publication provided that you attribute the work. The ATSB’s preference is that you attribute this publication (and any material sourced from it) using the following wording: Source: Australian Transport Safety Bureau Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

__________

1. Pilot flying and pilot monitoring are procedurally assigned roles with specifically assigned duties at specific stages of a flight. The pilot flying does most of the flying, except in defined circumstances. The pilot monitoring carries out support duties and monitors the actions of the pilot flying and the aircraft flight path.
2. The flight plan package was produced by the operator’s flight dispatch department at 0642 on the day of the incident.
3. A NOTAM (Notice to Airmen) advises personnel concerned with flight operations of information concerning the establishment, condition or change in any aeronautical facility, service, procedure or hazard, the timely knowledge of which is essential to safe flight.
4. 24K is a derated thrust setting. Engine thrust settings less than the maximum available thrust are often used during take-off. Take-off operations conducted at thrust settings less than the maximum take-off thrust available may provide substantial benefits in terms of engine reliability, maintenance and operating costs (FAA Advisory Circular 25-13).
5. The ATIS (Automatic Terminal Information Service) is an automated broadcast of prevailing airport weather conditions that may include relevant operational information for arriving and departing aircraft.
6. For the purpose of this report, an OPT means that the crew were required to request takeoff reference data (for departure under conditions where the runway length was reduced due to the works in progress) from the operator’s flight dispatch staff. That request could be made using on-board aircraft communication systems, or by telephone.
7. NOTAMs and other aeronautical information typically use Coordinated Universal Time (UTC) as a time reference. NZST is UTC plus 12 hours.
8. FCONs are company NOTAMs which are issued to flight crew by the flight operations department to convey new operational and technical information which is of an urgent nature. Flight crew are required to obtain and review a copy of the current FCONs at the commencement of duty each day.
9. The reference to yellow pages in this NOTAM relates to relevant Aeronautical Information Publication NZ aerodrome charts (that depict the works in progress and provide associated operational information), not the operator’s AAM yellow pages referred to elsewhere in this report.

What happened

On 14 July 2015, the pilot of a Robinson R44 helicopter, registered VH-VOH, was conducting aerial mustering operations on a property, about 70 NM east of Alice Springs, Northern Territory.

At about 1300 Central Standard Time (CST), the pilot was mustering cattle along a creek system. The helicopter was above tree height, at about 50 ft above ground level (AGL), when the pilot slowed the helicopter to an airspeed of about 40 kt. The pilot felt a small vibration, and initially thought it was due to loose tape on the main rotor blade. The pilot looked for a suitable landing site, but the vibration increased significantly.

As the helicopter descended, the pilot manoeuvred the helicopter through a gap between trees, and pushed the cyclic^[1] forward to maintain airspeed. The pilot lowered the collective^[2] and noticed the engine seemed to go very quiet and the low rotor revolutions per minute (RPM) warning horn sounded. The pilot made a radio call to advise another pilot operating nearby that the helicopter was going down. The pilot then flared^[3] the helicopter to try to cushion the landing impact. The right skid touched down first, and the helicopter rolled onto its right side.

The pilot sustained minor injuries and the helicopter was substantially damaged (Figure 1).

Pilot comments

The pilot provided the following comments:

• There was no noise to indicate that the helicopter had hit anything.
• The helicopter did not yaw when it vibrated.
• The wind was from the southeast at less than five knots, and the pilot turned the helicopter towards the east between the trees.
• The weather was fine; there were no visible signs of moisture, and only a few high level clouds.
• The pilot did not select the carburettor heat on at any stage during the flight.
• Both tail rotor blades broke off the helicopter’s tail. This may have occurred prior to striking the ground, but the pilot did not detect the helicopter tail rotor colliding with anything.

Figure 1: Accident site showing damage to VH-VOH

Source: Aircraft owner

Meteorological conditions

Weather observations from the Bureau of Meteorology’s automatic weather station at Alice Springs indicated that at 1300, the temperature was 12.9°C, relative humidity 45%, and the dew point^[4] was 1.2°C. The dew point depression, calculated by subtracting the dew point from the temperature, at that time was 11.7.

According to the Carburettor Icing Probability chart (Figure 2), the conditions indicated a high probability of serious carburettor icing at descent power.

Figure 2: Carburettor icing chart showing prevailing conditions in yellow

Source: Civil Aviation Safety Authority – modified by the ATSB

Safety message

The ATSB advises all pilots of aircraft fitted with a carburettor to check the forecast conditions and know the risk of carburettor icing prior to each flight. The carburettor icing probability chart is available on the CASA website.

The following publications provide additional information on carburettor icing:

• Melting Moments: Understanding Carburettor Icing
• Flight Safety Australia – A chill in the air
• Mornington Sanford Aviation – No ice, thank you
• Helicopter Safety - Carb icing

Aviation Short Investigations Bulletin - Issue 43

About this report Decisions regarding whether to conduct an investigation, and the scope of an investigation, are based on many factors, including the level of safety benefit likely to be obtained from an investigation. For this occurrence, a limited-scope, fact-gathering investigation was conducted in order to produce a short summary report, and allow for greater industry awareness of potential safety issues and possible safety actions. Publishing information  Released in accordance with section 25 of the Transport Safety Investigation Act 2003 Published by: Australian Transport Safety Bureau © Commonwealth of Australia 2015 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this publication is licensed under a Creative Commons Attribution 3.0 Australia licence. Creative Commons Attribution 3.0 Australia Licence is a standard form licence agreement that allows you to copy, distribute, transmit and adapt this publication provided that you attribute the work. The ATSB’s preference is that you attribute this publication (and any material sourced from it) using the following wording: Source: Australian Transport Safety Bureau Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

__________

1. A primary helicopter flight control that is similar to an aircraft control column. Cyclic input tilts the main rotor disc varying the attitude of the helicopter and hence the lateral direction.
2. The collective pitch control, or collective, is a primary flight control used to make changes to the pitch angle of the main rotor blades. Collective input is the main control for vertical velocity.
3. Flare reduces rate of descent before ground impact by increasing collective pitch; this increases lift, trading stored rotor kinetic energy for increased aerodynamic reaction by blades, and should result in a gentle touchdown.
4. Dewpoint is the temperature at which water vapour in the air starts to condense as the air cools. It is used among other things to monitor the risk of aircraft carburettor icing or likelihood of fog at an aerodrome.

Safety summary

What happened

On 11 July 2015, the Melbourne to Albury service 8625 was approaching Wallan crossing loop when the train traversed the points into the loop road at excessive speed. As a result, the passengers and crew experienced a rough ride resulting in some passengers requiring medical attention from the on board service crew. The service crew reported the incident to V/Line operations and the train continued on to Seymour.

On arrival at Seymour, the train was met by a Regional Driver Supervisor (RDS). The RDS questioned the driver about the rough ride at Wallan and after a short discussion, the train departed making scheduled stops along the way.

Shortly after a stop at Euroa station, some of the passengers approached the conductor about persons in their care showing signs of discomfort and stress as a result of the earlier rough ride. The conductor decided to arrange for an ambulance to meet the train at Benalla station. Meanwhile, V/Line operations were arranging to have the driver of 8625 relieved of duty at Benalla.

On arrival at Benalla station, the conductor told the driver of 8625 that he was to be relieved of duty. Ambulance officers treated some of the injured passengers for minor injuries with one passenger and their carer taken to a nearby hospital. The service continued onto Albury with the new driver.

What the ATSB found

The ATSB found that the driver of train 8625 did not demonstrate effective train handling techniques when approaching a signal displaying a low speed aspect. As a result, 8625 traversed the points at a speed significantly greater than the allowable engineering speed.

In addition, the driver did not immediately report the severity of the incident to the ARTC train control or V/Line. As a result, the possibility of infrastructure damage exposed trains travelling on the Standard Gauge at Wallan to a potentially elevated safety risk.

The ATSB also found that V/Line did not have a procedure in place that specifically required other V/Line employees to report incidents in the event that a driver did not. Furthermore, the ATSB found that V/Line’s processes did not consider the potential for rolling stock or infrastructure damage in the event that a train had traversed a turnout at significantly greater speed than designed.

What's been done as a result

V/Line have developed and implemented a procedure, which acknowledges and addresses the risk associated with gross over-speed. They have also updated their Just Culture policy in relation to reporting and have incorporated redundant pathways to ensure reporting should a driver fail to do so.

Safety message

All incidents that could compromise the safety integrity of the network must be reported immediately to Train Control. Operators need to ensure robust reporting procedures are implemented.

For incidents that involved gross over-speed, train operators should implement procedures that identify and manage the risk appropriately.

The occurrence

At 1802^[1], on 11 July 2015, V/Line passenger train 8625, departed Southern Cross Station (Melbourne, Victoria) for a scheduled service to Albury (New South Wales).

Shortly after departing Southern Cross Station, train 8625 stopped on the passing loop at Tottenham and waited for a freight service to pass on the main line. Train 8625 then continued onto Broadmeadows for its first scheduled stop. At about 1840, train 8625 departed Broadmeadows, bound for its next scheduled stop at Seymour (Figure 1).

Figure 1: Location map – Wallan, Seymour, Benalla Victoria

Source: NatMap Railways of Australia

At about 1845, the Australian Rail Track Corporation (ARTC) Network Control Officer (NCO) had set the signals at Wallan (Figure 1) for train 8625 to take the passing loop, allowing an oncoming Melbourne bound train to pass on the main line. This action set number 3 points for the passing loop and then the signal interlocking cleared signal WLN/2 for a low speed entry into the loop track. Signal WLN/2 is located about 50 m from the turnout (number 3 points) and signal ES1475 about 3.5 km from the turnout (Figure 2).

Figure 2: Signalling approaching Wallan

The above schematic shows the position of the signals, level crossings, and points. The signals show the aspects at the time of the train approaching ES1475 and WLN/2. Not to scale. Source: ATSB

At about 1858, train 8625 approached signal ES1475 displaying a yellow aspect and passed it travelling at about 104 km/h. It was just after sunset and it had been raining intermittently. The driver continued on to Wallan, managing the train’s throttle to maintain a speed of about 100 km/h.

About 2 minutes later, train 8625 approached and passed signal WLN/2 signal displaying a low speed aspect (Figure 4). At about this point, the driver placed the train’s throttle into the idle position. At the time, the train was travelling at a speed of about 98 km/h.

Shortly after, train 8625 crossed over number 3 points and onto the loop track, travelling at about 97 km/h. The train swayed violently, severe enough that some of the passengers were thrown out of their seats. About 5 seconds after passing over the points (a distance of about 185 m), the driver made a service brake application. The train stopped about 1.5 km later, at WLN/U6 signal (Figure 3) in readiness for its departure.

Figure 3: Layout of signalling at Wallan loop Standard Gauge.

Figure 3 shows the location of the signalling. Train 8625 arrived at Wallan from Melbourne (from left of figure 3) Once 8625 traversed the number 3 points the train came to a stand at WLN/U6 signal on the loop track (No.2 road) waiting for train 6WP2 to approach from the Seymour (right side of Figure 3) on the main line (No.1 road). Source ARTC

Figure 4: Approaching WLN 2 signal and 3 points into Wallan Loop road.

Figure 4 shows the location at and beyond WLN/2 signal over Wallan Whittlesea Rd and approaching 3 points with the signal and points set for the loop road. Note the low speed aspect (yellow) on the ‘C’ arm. At the time, 8625 was travelling over 90 km/h. The top right call out box shows number 3 points in the reverse position and the loop road diverging to the right hand side. The lower right hand call out box show the track layout beyond 3 points through the cripple road points (right hand side). Train 8625 was traversing the middle road (Figure 3 No.2 road) after traversing 3 points in reverse. Source: ATSB

After stopping at signal WLN/U6, the driver contacted the conductor to apologize for the rough ride into the loop road. The conductor responded on the radio that they were attending to some of the passengers who had been thrown out of their seats. At that time the driver of train 8625 did not contact ARTC NCO to advise of the occurrence.

Post incident

At about 1905, freight service 6WP2 passed through Wallan on the main line and continued over number 3 points towards Melbourne. There was no report from the crew to the ARTC Network Controller regarding any rough ride over number 3 points at Wallan.

Shortly after the passage of 6WP2, signal WLN/U6 (Figure 3) cleared to proceed and train 8625 continued on towards Seymour.

As the train continued towards Seymour, the buffet car attendant contacted the V/Line Network Customer Centre^[2] (NCC) to report the rough ride incident on the 8625 Albury service and that passengers had been shaken up. The buffet car attendant then contacted the driver via the radio and asked him to also talk to the NCC regarding the incident.

The driver contacted the NCC and advised that he went a bit quick over the points at Wallan, but that everything was OK. The operations officer enquired about the passenger injuries and informed the driver that the Regional Driver Supervisor (RDS) would be notified. The driver asked that the incident not be reported to the RDS, but the NCC explained that it was protocol to report incidents causing injuries, via text message, to the RDS as well as other appropriate staff within operations.

Meanwhile, a relief service crew (excluding driver) were waiting to join train 8625 at Seymour station. The buffet car attendant on train 8625 contacted Seymour station to advise the relief service crew that there had been a rough ride event and to keep an eye on the passengers when they took over the service. At about the same time, the NCC contacted the on-duty RDS, who was also located at Seymour, and requested the RDS meet the train to enquire further with the driver regarding the incident.

At about 1940, train 8625 arrived at Seymour and the relief service crew boarded the train. The conductor joining the service noticed that some of the passengers disembarking from the train were in some discomfort. He talked to the outgoing conductor who informed about the condition of the remaining injured passengers, including a passenger in a wheel chair that had also sustained an injury. Passengers also remarked to the conductor about the violent nature of the event.

The RDS also met the driver at Seymour and discussed the reported rough ride at Wallan. The driver informed that the train was going a bit quick over the points at Wallan. The RDS was not aware that the train was in fact traversing the points into the loop road at the time of rough ride. The RDS advised the driver to be careful of his speed and to slow down. As the RDS walked away, an off-duty driver travelling on the train approached the RDS and explained that the service was signalled into the loop road at Wallan. The RDS re-engaged with the driver and verified that the train was signalled into the loop road at Wallan. Upon ascertaining that the train was going into the loop road, the RDS acknowledged that the move into the loop road was a low speed and that driver was going too fast for the move over the points. The RDS then reiterated to the driver to slow down and take more care on the rest of his journey.

At about 1944, train 8625 departed Seymour to continue its journey, with scheduled stops at Avenel, Euroa, and Violet Town.

Meanwhile, the (V/Line) General Manager Train Services contacted the RDS to discuss the situation. A decision was made for the RDS to relieve the driver of train 8625.

The RDS departed Seymour by taxi and travelled towards Violet Town to meet and take over control of the service.

Shortly after at about 1955, Melbourne bound freight service 7MP7 was travelling through Wallan on the mainline. There was no report of 7MP7 experiencing any rough ride as it passed through Wallan and over number 3 points.

While on route to Violet Town, the RDS contacted the ARTC NCO to request that train 8625 be held at Violet Town to relieve the driver. At that time, the NCO had not been advised of the incident that had occurred at Wallan.

After some further consideration, V/Line operations decided to hold the train at Benalla (about 25 km past Violet Town) to avoid disrupting passengers transferring to a bus service and to avoid blocking the town’s main road over the railway tracks. The RDS contacted the ARTC Train Control and advised that the train would be held at Benalla instead of Violet Town. Again, the NCO was not told of the incident at Wallan.

Shortly after, the RDS contacted ARTC again to inform the NCO of the incident at Wallan regarding the over-speed (rough ride) over the points. In response, the NCO advised that they would slow further trains through Wallan until the points (track) at Wallan could be inspected for damage.

As train 8625 continued towards Benalla making its scheduled stops, the service crew attended to the remaining passengers who were in various levels of discomfort. At about 2046, when between Violet Town and Benalla, the conductor was made aware of a high dependency passenger that was in some discomfort caused by a knock to the head during the incident. The passenger’s carer requested that an ambulance attend to the person. The conductor placed a call to the Emergency Services Telecommunications Authority (ESTA commonly known as ‘000’) and requested that an ambulance meet the train at Benalla station to attend to the injured passengers.

Soon after, another passenger approached the conductor concerning a possible injury to their child. The conductor informed the passenger that an ambulance was meeting the train at Benalla where passengers would be assessed.

Not long after placing the call to ‘000’, the conductor then returned a missed call to his manager. The manager informed the conductor that they could not contact the driver via radio or phone, and asked about the driver’s general demeanour. The conductor advised that the driver’s behaviour had appeared normal. The manager asked the conductor to approach the driver at Benalla to advise that the RDS was on the way to take charge of the train.

Train 8625 arrived at Benalla at about 2053, about the same time as the ambulance. Some of the injured passengers left Benalla on a connecting bus service without seeking treatment, noting also that some had left the service prior to Benalla. The remaining injured passengers were treated at the station and it was decided to convey one person to hospital for further observation.

Meanwhile, the conductor radioed the driver and asked to speak to him at the front of the train on the platform. The conductor informed the driver that he was not to continue and that the RDS was coming to relieve him.

Shortly after the RDS arrived at Benalla and performed an alcohol breath test on the driver. The driver then returned to Melbourne by taxi.

At about 2135, train 8625 continued to Albury under the control of the RDS.

At about 2200, ARTC track and signal maintenance staff arrived at Wallan to inspect and test the signalling. About an hour later, maintenance staff reported to ARTC train control that the equipment and track had been tested ok and was clear for normal traffic to resume. The temporary speed restricted on that section of track was removed and services through the area returned to normal.

The rolling stock was not examined for any damage until the train had returned to Melbourne the following day. The inspection did not find any damage as a result of the over-speed entry at Wallan.

__________

1. The 24-hour clock is used in this report. Local time was Eastern Standard Time (EST)
2. The V/Line Network Customer Centre (NCC) provides operational support to V/Line passenger trains on the ARTC rail network, but does not provide train control services on the ARTC rail network.

Context

Location and track information

Wallan is located on the Melbourne to Sydney mainline about 47.5 km^[3] from Southern Cross Station Melbourne.

The track is a bidirectional standard gauge^[4] track consisting of continuously welded rail on concrete sleepers, fastened by resilient clips. Crossing loops^[5] are provided at various locations, including Wallan, to facilitate the passing or crossing of trains. The loop track is often referred to as the loop road or Number 2 road.

A railway station is located at Wallan. However, the station only services an adjacent broad-gauge^[6] track. There are no passenger facilities for standard gauge trains travelling through Wallan.

The Australian Rail Transport Corporation (ARTC) lease and manage the standard gauge track, with signal control from the ARTC Network Control Centre South at Junee (NSW).

The line speed through Wallan is 130 km/h. However, the signalled speed for the diverging route over the turnout into the loop road is 15 km/h (low speed signal).

Track inspection

Post incident track inspection at Wallan loop was undertaken by the ARTC later that evening. The track was undamaged and there was no pre-existing track condition evident that may have contributed to the rough ride.

Signalling

The signalling at Wallan loop on the standard gauge is a CTC^[7] relay based type interlocking.

The three-position signal heads used on signals ES1475 and WLN/2 were Westinghouse K3 searchlights (Figure 5). The K3 searchlight signals are a long-range signal head and have a viewing range of up to 2500 m in clean air.

Figure 5: Signal WLN2 and ES 1475

Signals WLN/2 and ES1475 taken on the 7 July after the incident during testing of the signals. Signal WLN/2 is displaying a red over red over yellow aspect (Low speed). Signal ES1475 is displaying a yellow over red aspect (Normal speed warning). Both signals are displaying the aspects that the driver would have seen approaching Wallan. The ‘A’ arms on both signals are K3 (long-range) Searchlight signals. Source: ARTC.

Signal testing

The ARTC examined the operation of signal ES1475 and WLN/2 and number 3 points at Wallan Standard Gauge Loop. The tests concluded that the signalling and points equipment were working as per design at the time of the incident.

Signal sighting and operation

The ARTC conducted an inspection on the sighting distance of the signals. Both ES1475 and WLN/2 was found to be in good working order with no issue of sighting over the designed viewing distance.

The ATSB site inspection also noted that the sighting distance from ground level for both signals was clear and unobstructed for the respective distances.

Number 3 Points and turnout

Number three points at Wallan are McKenzie and Holland M23A dual operation points.^[8] The points are on a 1 in 10^[9] turnout. The designed speed of the turnout when track conditions are optimum is 40 km/h. At Wallan however, the signalling aspect for the movement over the points is 15 km/h.

At the time train 8625 was traversing the turnout, the train was travelling at approximately 97 km/h. This was about 80 km/h greater than the signalled speed and about 55 km/h greater than the turnout design speed. As a result, there would have been significant lateral forces on the rolling stock and infrastructure as the train traversed the turnout into the loop road.

The over-speed entry into the loop road contributed to the excessive lateral forces (rough ride) experienced by the passengers and crew on-board train 8625.

Train information

The Melbourne to Albury V/Line service is a driver-only^[10] locomotive hauled passenger service consisting of an N class locomotive (N464) and an N class carriage set (SN16). Carriage set SN16 consisted of three economy wagons, a buffet car, and a first class car. There was also a parcel van/cargo carriage at the rear of the train.

In addition to the driver, train 8625 was also serviced by a crew consisting of a conductor and a service attendant for the buffet car. The conductor provided customer service and was responsible for all passenger operations and their welfare.

Rolling stock inspection and maintenance

In April 2015, Locomotive N464 underwent schedule maintenance and was returned to service in May 2015.

In June 2015, carriage set SN16 underwent scheduled maintenance and was returned to service in the same month.

On 12 July (the day following the incident), inspection of carriage set SN16 was undertake by Bombardier to determine if there was any damage as a result of the over-speed. No subsequent damage was found.

During interview, the driver stated that the train felt ok to continue and there were no issues with it as a result of the over-speed.

V/Line’s internal report concluded that there was no pre-existing condition of N464 or SN16 that may have contributed to the occurrence.

The ATSB noted that following the incident at Wallan, the passenger service ran to Albury and back to Southern Cross station the following day before being inspected for damage. While no problems may have arisen, the condition of the rolling stock was not considered until after it having returned to Melbourne.

Train Driver

The Melbourne to Albury service was operated with a driver as the sole person in the locomotive cab (driver-only operation). The driver was qualified in October 2014, to operate the driver-only service but was undergoing Train Driver Safety Audits as a result of a performance management issue.

Training

The driver of 8625 was certified competent to operate diesel hauled trains on the Southern Cross to Albury line. He had completed a written exam on 28 October 2014 and completed a route knowledge assessment on 30-31 October 2014. The assessment included completing four return trips while being supervised and assessed by a trainer driver. However, neither the written nor the practical assessments included a movement into the loop road at Wallan.

Drugs and Alcohol medical

The driver underwent testing for drugs and alcohol at Benalla Station and returned zero readings.

At the time of the incident, the driver was deemed to be medically fit for driver-only train operations.

Fatigue

The driver’s roster, sleep patterns, and general health were analysed. The investigation concluded that fatigue impairment was unlikely to be a contributing factor to this incident.

Train handling

Track speed permitted on the main line through Wallan is 130 km/h. However, V/Line mandates that V/Line trains are limited to 100 km/h.

At the time of the incident signal ES1475 was indicating a yellow aspect informing the driver that the next signal (WLN/2) was at stop. After passing ES1475 a driver should be managing the train to stop at signal WLN/2 (Figure 6).

On approach to signal WLN/2 (about 200 m prior the signal), a low speed aspect would have been visible, indicating that a route had been set for the loop road (Figure 6). The driver must be prepared to stop clear of any obstruction and not exceed a speed of 15km/h.

Figure 6: Typical speed profile versus actual speed

Figure 5 illustrates the typical speed profile for a train being signalled into the loop compared to the speed data taken from the locomotive data logger. Not to scale. Source: ATSB.

Train data logger

Locomotive N464 was fitted with a Fischer data logger device, which captured various parameters, including:

• Time, speed, distance
• Throttle position
• Brake pipe pressures
• Horn (Country and Town)
• Vigilance
• Dynamic brake

The data logger showed that on the approach to both signals the driver was making slight changes to the throttle to maintain speed of about 100 km/h (track speed). The driver also sounded the horn for the approach to Beveridge Rd and Wallan Whittlesea Rd level crossing. However, the data showed no reduction of the throttle nor brake application, at the point where a response to signals ES1475 & WLN/2 would be expected (Figure 6).

Analysis showed that the driver was managing the train in a manner consistent with being signalled straight through Wallan on the mainline. However, with the route set for the loop road, the train traversed the points at more than 90 km/h, significantly greater than the designed (signalled) speed.

The driver of train 8625 did not demonstrate effective awareness and train handling techniques consistent with approaching a signal displaying a low speed aspect.

Drivers awareness of the signalling system at Wallan

The driver had been trained on the Southern Cross to Albury section (otherwise known as route knowledge). The driver was supervised during his training and made four returned trips before being assessed as competent to drive a train on this network.

During interview, the driver was asked questions regarding the signalling at Wallan. In particular, the driver was asked about signals ES1475 and WLN/2.

The driver explained the type of aspects ES1475 would show in relation to the signal ahead (in this case WLN/2). The driver explained that if ES1475 were showing a yellow aspect, then the signal ahead (WLN/2) would be at stop^[11] (red over red).

When discussing WLN/2 signal, the driver commented that he had never been signalled into the loop track at Wallan before and that his expectation was a yellow on the “B” arm. However, at Wallan the aspect for trains to enter the loop track is in fact a “C” commonly referred to as a low speed. The driver was not aware the WLN/2 signal would show a low speed aspect for the move into the loop road and that the “B” is a fixed red^[12] aspect.

Although there may have been some confusion on what aspect the driver was expecting, when viewing WLN/2 displaying a low speed at distance, the driver would have seen a red signal (stop aspect). As such, the driver should have been preparing the train to stop at the signal. At a distance of around 200 m, the driver should have seen the ‘C’ arm low speed illuminated. Having slowed the train in preparation to stop, a driver would then proceed into the loop road at a speed no greater than 15 km/h.

Although the driver was not aware of the signal indications for the Loop Road at Wallan, the driver should have been handling the train in preparation for stopping at the signal.

Driver history

The driver of 8625 started his career in 1984 as a fireman^[13] with V/Line in Bendigo Victoria. In 1988, the driver transferred to Metro Trains in Melbourne to drive on the electrified network. In 2007, the driver returned to V/Line under the “Right of Return” agreement.

Right of return agreement

Under an agreement between V/Line and the Victorian government, a policy for ‘right of return’ was an historical arrangement dating back to the separation of electric trains and diesel services in Victoria. When the split happened drivers were told in order to complete their training they would have to transfer over to the electric trains. At the time, they were given a commitment from the Government that they could return at any time provided that V/Line was recruiting for drivers.

Under the agreement, drivers accepted back did not have to go through a recruitment process. The right of return was premised on the proviso that the driver had continuous service with electric trains and the driver had not refused re-employment with V/Line if an offer was made.

Drivers accepted under the right of return agreement were treated as a transfer. Medical records would be transferable, but the driver’s performance history was not disclosed.

Training and performance

In this case, the driver of 8625 was accepted back under the right of return agreement to V/Line in July 2007 as a Conversion Driver. V/Line acquired the driver’s medical history. V/Line also attempt to ascertain if there was any disciplinary action with Metro Trains. Due to privacy constraints, it was up to the discretion of Metro Trains or the employee to provide the information. At the time, no information was given to V/Line regarding any driver performance issues.

The driver underwent a conversion driver program, which recognised prior knowledge and experience. The driver then underwent network specific training eventually progressing to the North East network.

During the driver’s progression to operate through various networks, the driver was also undergoing performance management for various indiscretions, which included issues with inattentiveness.

Over the next six years, the driver had a number of incidents recorded against his driving record. Formal performance interviews and V/Line’s demerit point system were used to manage these incidents. Some of the incidents included SPAD’s^[14] and failure to stop at platforms. These types of incidents could be attributed to lack of concentration (inattentiveness).

The most recent incident was a SPAD in February 2014. Following this incident, the driver received a final written warning and was placed on a schedule of Train Driver Safety Audits (TDSA) and check rides.

Driver performance management

As a result of the SPAD in February 2014, (when considering the drivers prior history) V/Line issued the driver a final written notice as per V/Line’s Employee Misconduct and Discipline Procedure HRPR-33.

The driver was also required to undergo the following auditing process:

• Check rides
• 1 TDSA per month for the first six months then,
• 1 TDSA per three months for the following 12 months

Train Driver Safety Audits

V/Line Train Driver Safety Audits Policy states that:

Train Driver Safety Audits (TDSA) are aimed at ensuring that the functions associated with train driving are performed in accordance with the Book of Rules and Operating Procedures 1994, specified standards, Polices and all relevant Work Instructions and Procedures.

The policy also describes the following categories of Safety Audits as:

• Train Driver Safety Audit
• Train Drivers Safety Re-accreditation Audit
• Train Driver Promotion Assessment
• Trainee Train Driver Safety Validation
• Practical Driver Trainer Safety Audit

When considering the above categories of safety audits, the driver underwent a series Train Driver Safety Audits. The procedure explains it as:

Train Driver Safety Audit – Safety Audit conducted on all drivers that assumed the full responsibility of the position.

TDSA’s are used to monitor a drivers performance with regard to the functions associated with driving trains within the relevant rules and operation procedures on an annual basis (Train Driver Safety Audits).

Check ride

The driver underwent two separate check rides in March and April 2015 before undergoing the TDSA’s.

During the drivers first check ride over a return journey to Southern Cross Station, the auditor commented on the report that the driver;

An observation was if the driver is distracted in any way in the sense of talking on a phone or radio or having someone in the cab making conversation etc. It is my opinion that he may struggle to determine the sense of what situation he is in. i.e. speed restrictions, signals ahead at Stop, approaching platforms and how to prioritise tasks to keep the situation safe not only for himself but for others.

The second check ride was conducted over the period of two weeks. The auditor surmised in the report that;

The driver showed good concentration for all trips but I must comment that he did miss two warning boards without good reason and not centring reverser at signals after his previous misdemeanours with SPAD’s.

Train Driver Safety Audits regime

In the first 6 months all but one TSDA’s was supplied to the ATSB (June 2014 is missing).

For the following 3 months, TDSA between the month of October 2014 and June 2015 there was only one TDSA documented (June 2015). In that time, there should have been another TDSA’s sometime around February/March 2015.

The driver was currently undergoing the second phase of the TDSA regime in undertaking an audit every 3 months for the next 12 months (audits scheduled to finish around October 2015) at the time of the occurrence.

After reviewing the TDSA’s that were supplied to the ATSB none of the RDS’s undertaking the audits commented on any issues of performance such as losing concentration. In general, the driver performed without incident during the audits with the exception of a small non-conformance.

At the time of the occurrence at Wallan, the driver was still undergoing the TDSA/check ride process.

Reporting an incident

The driver of 8625 did not report the incident to (V/Line) Centrol^[15] or the ARTC Network Control. The initial report to V/Line operations was made by the buffet attendant as a result of the minor injuries suffered to passengers.

Once V/Line operations became aware of the injuries, an internal notification was sent out via text to the General Manager for Train Services and relevant staff. However, even once the extent of the over speed was known, no one within V/Line Operations immediately contacted the ARTC Network Control to notify them of the over-speed at Wallan number three Points.

V/Line reporting procedure

V/Line procedure SAPR-33 Incident and Hazard Reporting defines a Rail Safety Incident as;

A circumstance, act or omission relating to rail infrastructure or operations that had the potential to result in the death or serious injury to any person, or significant damage to property. Includes:

- Any defect in, or failure of, any part of the rail infrastructure or rolling stock and/or

- Any failure or breach of any rail operations practice, procedure, or rule.

The over-speed at Wallan constituted a Rail Safety Incident as it breached the rule of entering into the loop on a low speed at 15 km/h.

For reporting Rail Safety Incidents SAPR-33 requires that;

To report a Rail Safety Incident on V/Line or Metro Infrastructure, contact Centrol immediately (at first safe opportunity) using train to base radio or telephone:

(Note: If an incident occurs in a network not controlled by V/Line. The incident should be reported to the infrastructure managers train control via train-to-base radio. The incident must also be reported to V/Line – either via Centrol or the Report a HSE incident online form).

ARTC TA20 Section 1 General Rules state;

Rule 6b; Conditions that can or do affect the safety of rail operations in the Network must be reported promptly to the Network Controller responsible for the affected portions of the track.

Train operators are to immediately notify infrastructure owners of any incident that can compromise the safety integrity of the network to the infrastructure owner. ARTC Emergency Management procedure, TA44 requires that once an operator becomes aware of an actual incident, they are to take all necessary steps to ensure that the incident site is protected and immediately contact ARTC Network Control to ensure the protection of the network.

Incident management

At the time of the occurrence the driver attempted to ‘down play’ the incident with the V/Line NCC for fear of reprisals and asked that the RDS not be notified. However, due to the excessive speed, the occurrence increased the safety risk in relation to the network (damaged track), potential damage to rolling stock, as well as possible passenger injuries. At the time, the driver was more focussed on trying to avoid any escalation of the incident and was less focussed on the issues such as damaged track rolling stock and passenger injuries.

The driver was spoken to on the platform at Seymour by the RDS to establish the cause of the rough ride. At the time during the conversations the RDS was determining the drivers demeanour and if the driver was affected by drugs and alcohol.

The RDS understood the complexities of removing the driver from service, locating a relief driver at short notice and the possibility of cancelling the service. The RDS also understood the track speed at Wallan Loop over the points.

The RDS, while anecdotally knowing the drivers past history, did not have access to the driver’s performance history with V/Line. While this RDS does not manage this driver, to be able to make an informed decision on the driver’s capacity to continue, it would have been advantageous to have had access to, or been provided with information about the driver’s history and ongoing performance issues at that time.

In hindsight, due to the severity (speed) of the incident, the effort in trying to conceal and ‘down play’ the incident, and the driver’s prior history, the RDS should have relieved the driver at Seymour regardless of the operational issue surrounding the cancellation of the service. However, the RDS was not comfortable in respect to the organisational pressures in making such a decision without a clear policy on relieving a driver for a gross over speed.

Driver fit to continue

There is no V/Line policy for RDS’s when dealing with performance issues such as a significant over-speed to relieve the driver. The Locomotive Driver Demerit OPPR-33 policy deals with the duty of an RDS to stand drivers down on the advice of an irregularity. However, in regards to standing down a driver as a result of an over-speed, it is not specific about how to deal with an over-speed greater than 21 km/h in relation to relieving a driver.

Passenger welfare

At the time of the occurrence, passengers were thrown around the carriage with some sustaining some bumps and scalding from hot drinks. Also traveling on the service was two high dependency passengers one of which was ejected from their wheelchair. The service crew administered first aid to those passengers that requested some assistance.

When the train was at Seymour, there was no communication between conductors (those joining and leaving the service), the RDS and V/Line NCC regarding passenger welfare or any further medical assessment or assistance.

Once the service had departed Seymour, the new conductor (now on the service) was approached by concerned passengers that were caring for high dependency person(s) with concerns.

Rolling stock fit to continue

The over-speed at Wallan over the points would have placed excessive lateral loads on the track and rolling stock. The excessive lateral load could have damaged the rolling stock or track (points).

At the time, the driver considered the train safe to continue because in the driver’s opinion the train felt normal after the event. Similarly, once the RDS and V/Line operations became aware of the excessive speed that the train traversed the points there was no consideration given to the state of the rolling stock to continue. It was not until the next day that the rolling stock was examined.

__________

3. Distance in kilometres from a track reference point at Southern Cross Station.
4. The name given to the gauge of track 1435mm wide between running rails.
5. A length of track connected to the main line by switches at both ends to provide a facility that permits trains to both cross and pass each other.
6. The name given to the gauge of track of 1600mm wide between running rails
7. CTC – Centralised Traffic Control. A system of remotely controlling the points and signals at a number of interlocked stations, junctions and crossing loops in automatic signalling areas, from a centralised control or signal box.
8. Dual control points - A power operated point machine also equipped for hand operations.
9. 1 in 10 refers to the crossing rate of the turnout. The crossing rate is a measure of the angle made by the rail gauge faces at the theoretical point. The larger the crossing rate, the smaller the angle the faster the speed through the crossing.
10. Driver-only operations do not have a second person in the cab.
11. WLN/2 signal was showing a Red over Red over Yellow indication. The yellow is a low speed signal. At approach distance, the driver would only see the Red over Red signal as the yellow (low speed) is focused for a short range viewing of around 200 meters.
12. Fixed red is a signal that can only show a red aspect (single position)
13. Fireman is a term used for the second person/observer.
14. Signals Passed At Danger
15. Centrol is the V/Line train control centre.

Safety analysis

The driver of 8625 tried to under report the severity of the occurrence. V/Line operations then contributed to the delay in notification as a result of operational impact to services. This also contributed to the delayed reporting to the network owner.

In the first instance, a timely notification to the network owner was delayed by the driver’s attempts to cover up and under report the occurrence. Once V/Line understood the extent (speed) of the over speed, there was still a delay in notifying ARTC train control as V/Line operations was more concerned with managing the relief of the driver, ensuring minimal impact to passengers travel, and containing the train at Benalla.

At the time, there was no consideration of the fit state of the rolling stock at the time. It was not until the return of 8625 the following day to Southern Cross Station was there an inspection of the rolling stock.

Reporting a Rail Safety Incident

ARTC Emergency Management procedure, TA44 requires that any responsible rail employee must immediately report any Rail Safety Incidents that can compromise the safety integrity of the network to the network owner.

V/Line policy SAPR-33 Incident and Hazard Reporting procedure requires that the driver communicate all safety matters that can compromise the safety of the network immediately to the effected network owner or V/Line operations. However, on this occasion the driver did not report it and instead tried to downplay the incident.

The driver not immediately report the severity of the over-speed to ARTC Train Control as required under procedure SAPR-33. As a result of not immediately reporting the incident, other services running on that section of track were expose to a potentially elevated risk.

V/Line reporting procedure

During the process of information coming from the field, the RDS, conductor, and train driver were reporting back to different people within V/Line operations. However, there was no coordinated communications to a central point within V/Line operations.

When a Rail Safety Incident occurs in V/Line territory, it is reported to V/Line train control (Centrol). V/Line train control assumes not only the train control function but also (in this case) coordinates any notifications process such as emergency services.

When a Rail Safety Incident occurs in non-V/Line territory, the assumption is that the driver has made the report to the affect network owner. Once the information has been passed on to Centrol, they then notify internal stakeholders, rail regulator, and the Australian Transport Safety Bureau.

Throughout the process of a notification of a Rail Safety Incident being reporting on a network other than V/Line’s, there is no step to ensure that the affected network owner has been notified.

V/Line does not have a procedure that ensures that the affected network owner is immediately notified of an incident (Rail Safety Incident) if the driver has not reported it.

Reporting by employee

V/Line SAPR-33 HSE Incident and Hazard Reporting Procedure states that any employee or sub-contractor must immediately report any Rail Safety Incident. The policy dictates that any Rail Safety Incident must be communicated through the train driver.

Conductor Emergency Procedures also states that any communications should go through the driver. Any delays or disruptions to services can be reported directly Network Customer Centre.

There is reference to an incapacitated driver whereby the conductor would contact NCC directly. However, there is no provision to make any contact with NCC if there is an immediately notifiable rail safety incident that the driver has not reported.

V/Line had no provision in place whereby service crew may take action in the event that a driver does not (or cannot) respond appropriately to a rail safety incident.

Reporting by Network Customer Centre (NCC)

V/Line have an NCC Incident Escalation Procedure OPPR-57 on the reporting and escalation of a safety incident. The procedure defines a serious incident and the communication protocol.

There is no specific requirement to ensure that the relevant network owner (ARTC) has been notified of a serious incident. By definition of the procedure OPPR-57, a safety issue that can affect the network is required by SAPR-33 and ARTC’s TA 44 to be immediately reported.

V/Line’s policy SAPR-33 requires that the driver is to inform the infrastructure owner regarding any incident as well as V/Line NCC. As a result of the non-reporting of the occurrence by the driver, ARTC Network Control was not immediately notified.

The ARTC Network Control was contacted on two separate occasions by V/Line operations and on the third occasion (75 minutes after the incident) was formally notified of the over speed event at Wallan.

Any incident the compromises the safety integrity of the network must be report directly to the infrastructure owners Train Control. As a result of the delay, a freight and inter-rail service had passed over the section of track (points) of the over-speed exposing the services to a potentially elevated risk.

Even after NCC was aware of the incident, there were no steps in place to ensure that ARTC was aware of the incident.

V/Line’s procedure OPPR-57 Cars (NCC) Office Escalation Procedure did not include a requirement to contact the infrastructure manager in the event that the assessed severity of an incident had escalated.

Rolling stock fit to continue

V/Line consideration of a gross over speed is in relation to the driver demerit points system (driver performance). As a result, V/Line did not understand the risk associated with a gross over-speed (greater than 21 km/h) when considering the rolling stock.

V/Line did not have an effective procedure in place to reduce the risk when dealing with the management of a gross over-speed incident. With no procedure in place the train was allow to continue on as the operational staff directly involved with the service could make a determination on how to best manage the incident.

As a direct result of having no procedure (guidance) and not considering the risk, V/Line allowed the train to continue. The result of allowing the train to continue exposed passengers and other network users to risk as the rolling stock had not been considered for its fitness to continue.

V/Line’s processes did not consider the potential for rolling stock damage in the event that a train had traversed a turnout at significantly greater speed than designed.

Findings

From the evidence available, the following findings are made with respect to the over-speed that occurred at Wallan Crossing Loop on the standard gauge 47 km north of Southern Cross Station by rail in Victoria, on 11 July 2015. These findings should not be read as apportioning blame or liability to any particular organisation or individual.

Safety issues, or system problems, are highlighted in bold to emphasise their importance. A safety issue is an event or condition that increases safety risk and (a) can reasonably be regarded as having the potential to adversely affect the safety of future operations, and (b) is a characteristic of an organisation or a system, rather than a characteristic of a specific individual, or characteristic of an operating environment at a specific point in time.

Contributing factors

• The driver of train 8625 did not demonstrate effective awareness and train handling techniques consistent with approaching a signal displaying a low speed aspect.
• Train 8625 traversed the points at a speed significantly greater than the engineering speed, contributing to the excessive lateral forces (rough ride) experienced by the passengers and crew.

Other factors that increased risk

• The driver did not immediately report the severity of the over-speed to ARTC Train Control as required under procedure SAPR-33. As a result of not immediately reporting the incident, other services running on that section of track were expose to a potentially elevated risk.
• V/Line does not have a procedure that ensures that the affected network owner is immediately notified of an incident (Rail Safety Incident) if the driver has not reported it.
• V/Line had no systems in place whereby service crew may take action in the event that a driver does not (or cannot) respond appropriately to a rail safety incident.
• V/Line’s OPPR-57 Cars (NCC) Office Escalation Procedure did not include a requirement to contact the infrastructure manager in the event that the assessed severity of an incident had escalated.
• V/Line’s processes did not consider the potential for rolling stock damage in the event that a train had traversed a turnout at significantly greater speed than designed.

Other findings

• All signalling associated with incident at Wallan Loop was tested after the incident and found to be working as per design with no signal sighting issues.
  • The V/Line internal report determined that there was no pre-existing defect with the rolling stock that may have contributed to the rough ride.
  • There was no pre-existing track condition evident that may have contributed to the rough ride.

Additional safety action

Whether or not the ATSB identifies safety issues in the course of an investigation, relevant organisations may proactively initiate safety action in order to reduce their safety risk. The ATSB has been advised of the following proactive safety action in response to this occurrence

V/Line

V/Line have investigated and implemented the following short-term actions:

• The HSE Incident and Hazard Reporting Procedure (SAPR-33) has been revised and updated, and the online form for reporting incidents has also been revised to facilitate easy reporting. The Just Culture Policy (SAPO-10), which emphasises reporting responsibilities, was updated in July 2015 and Just Culture awareness and training will be conducted as part of the upcoming People Manager Induction and other Awareness training.
• An Incident Response Procedure providing a clear, documented process to assist in decisions about driver relief has been revised and is in place and an interim briefing process for interim response has been delivered and implemented into daily operations.
• HSE discussed this potential issue with People & Capability staff on 10th December 2015 to ensure that the principles of Just Culture are understood and that final warning communications are delivered clearly in accordance with these principles.

V/Line have investigated and are implementing the following longer term actions:

• The operational history of existing right of return drivers will be reviewed to identify any potential 'at risk' drivers (who will be subject to safety profiling and retraining as needed). NB. Review of drivers will be prioritised according to known incident history.
• Consideration of audible warnings for approaching signals and passing lanes in the business case for the Driver Advisory System. The V/Line Rollingstock department, has taken responsibility for this Project and a Project Nomination is being developed and funding is yet to be confirmed.
• V/Line and ARTC have held discussions regarding TPWS fitment and have agreed to work together on a train enforcement solution including TPWS. The Installation of TPWS on the North Eastern Line has been included in the Network Safety funding submissions that PTV has made to government.
• A review is currently underway to consider redundant reporting pathway(s) in the event that a driver fails to report an incident.
• V/Line’s Just Culture Implementation Procedure has been updated to reflect the criticality of appropriate incident reporting.
• A Rail Resource Management training program (named ‘Operational Awareness Training’) has been developed and is currently being rolled out to all staff in operational safety roles at V/Line to promote non-technical skills development.
• A formal on call process has been introduced to ensure the correct driver relief processes are followed after confirmed or alleged incidents and relevant staff have been briefed on their requirements when responding to such incidents.
• An internal review of potentially ‘at risk’ drivers has been undertaken and these are being monitored on a case by case basis as relevant.
• In relation to tracking driver development plans, a lead has been implemented for monitoring the current database to improve date tracking.
• A new Competence Management System for V/Line drivers is under development to support appropriate management of driver competence, including the management of non-technical skill deficiencies.

Sources and submissions

Sources of information

The sources of information during the investigation included the:

• V/Line
• ARTC
• Emergency Service Telecommunications Authority (ESTA (000))

References

RISSB Glossary of Railway Terminology – Guidelines Vol 1 2010

Westinghouse Brake and Signal product information for Searchlight Signals Style K2 & K3

V/Line Just Culture Policy SAPO-10 14/1/2013

V/Line Locomotive Driver Demerit System Procedure OPPR-33

V/Line Train Driver Safety Audits procedure OPPR-15

V/Line Train Driver Safety Audits checklist OPFO-13 Rev10

V/Line Position Description for Conductors

V/Line internal report into the over-speed at Wallan 11 July 2015

V/Line Conductor Emergency Procedures Conductor Training May 2015

V/Line Employee Misconduct and Discipline Procedure HRPR-33 Rev 8

V/Line HSE Incident and Hazard Reporting SAPR-33

V/Line NCC Incident Escalation Procedure OPPR-57 Rev 2

V/Line memo; VPO386-Speed Restrictions on North East Standard Gauge

Turnout design and components. Robin Stevens Queensland Rail

TA20 Section 2 Fixed Signals 4 October 2015 Rev 2

TA20 Section 1 General Rules 4 October 2015 Rev 2

Submissions

Under Part 4, Division 2 (Investigation Reports), Section 26 of the Transport Safety Investigation Act 2003 (the Act), the Australian Transport Safety Bureau (ATSB) may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. Section 26 (1) (a) of the Act allows a person receiving a draft report to make submissions to the ATSB about the draft report.

A draft of this report was provided to the Australian Rail Track Corporation, V/Line, the driver of 8625, and the Office of the National Rail Safety Regulator.

Submissions were received from Australian Rail Track Corporation, V/Line, the driver of 8625, and the Office of the National Rail Safety Regulator. The submissions were reviewed and where considered appropriate, the text of the report was amended accordingly.

Purpose of safety investigations The objective of a safety investigation is to enhance transport safety. This is done through: identifying safety issues and facilitating safety action to address those issues providing information about occurrences and their associated safety factors to facilitate learning within the transport industry. It is not a function of the ATSB to apportion blame or provide a means for determining liability. At the same time, an investigation report must include factual material of sufficient weight to support the analysis and findings. At all times the ATSB endeavours to balance the use of material that could imply adverse comment with the need to properly explain what happened, and why, in a fair and unbiased manner. The ATSB does not investigate for the purpose of taking administrative, regulatory or criminal action. Terminology An explanation of terminology used in ATSB investigation reports is available here. This includes terms such as occurrence, contributing factor, other factor that increased risk, and safety issue. Publishing information  Released in accordance with section 25 of the Transport Safety Investigation Act 2003 Published by: Australian Transport Safety Bureau © Commonwealth of Australia 2017 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this publication is licensed under a Creative Commons Attribution 3.0 Australia licence. Creative Commons Attribution 3.0 Australia Licence is a standard form licence agreement that allows you to copy, distribute, transmit and adapt this publication provided that you attribute the work. The ATSB’s preference is that you attribute this publication (and any material sourced from it) using the following wording: Source: Australian Transport Safety Bureau Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

Safety summary

What happened

In the early hours of 14 July 2015, the offshore support vessel (OSV) Skandi Pacific was loading cargo containers from the semi-submersible oil rig Atwood Osprey at its offshore location, about 90 miles north-west off Dampier. Shortly after 0505, cargo transfer was stopped due to worsening weather conditions. Skandi Pacific was moved 30 m away from the rig with the rough seas still on its port quarter. Two crewmembers then began securing cargo on the vessel’s aft deck.

While securing the cargo, the crewmembers slackened the securing chain they had used to secure the containers on the starboard side to better secure the entire stow. At about 0523, two large waves came over Skandi Pacific’s open stern, shifting the unsecured containers forward. One of the crewmembers was trapped between the moving containers, chains and a skip and suffered fatal crush injuries.

What the ATSB found

The ATSB investigation found that the risks associated with securing the cargo in the prevailing weather conditions on 14 July had not been adequately assessed. The fatally injured man was standing in a dangerous location near the unsecured cargo containers when they shifted.

The investigation identified that Skandi Pacific’s safety management system (SMS) procedures for working/securing cargo on deck in poor weather were inadequate with no clearly defined weather limits. Further, there were no clearly defined limits for excessive water on deck that necessitated stopping operations, leaving individuals to make difficult, and necessarily subjective, decisions about whether or not to stop work.

The ATSB also found that Skandi Pacific’s managers had not adequately assessed the inherent high risks associated with seas coming over the vessel’s open stern when work, including cargo handling operations, was being undertaken on its aft deck.

What's been done as a result

Proactive safety action by Skandi Pacific’s managers to avoid a similar accident includes improved cargo handling practices across its OSV fleet. Amongst these measures are updated procedures for working in adverse weather and cargo loading, including specific weather condition limits. In addition, existing risk assessments for offloading deck cargo at installations have been updated to include a section on risks associated with securing cargo.

The safety action taken by the vessel’s managers has adequately addressed the safety issues related to cargo handling/securing in adverse weather. The action taken has partially addressed the safety issue with regard to open stern vessels.

Therefore, the ATSB has issued a safety recommendation to the vessel’s managers to undertake further work to better address the risks associated with the use of vessels with open sterns. The ATSB has also issued a safety advisory notice to shipmasters, owners, and operators of OSV’s to highlight the risks posed by the open stern vessels to the industry more broadly.

Safety message

Offshore support vessel operations are inherently high risk because they often occur in exposed locations in a particularly dynamic environment. Multiple factors, including the weather conditions, schedule requirements, time of day, limited crew numbers, restrictions due to vessel design and systems, amongst others, add complexity to operations. Therefore, risk assessments are critical, with the weather and its impact on factors, such as an open stern, invariably a vital consideration.

The occurrence

At 1515^[1] on 7 July, Skandi Pacific (Figure 1) sailed from the Port of Dampier with a cargo for the semi-submersible oil rig, Atwood Osprey (Figure 2), about 90 miles^[2] north-west of Dampier.

Figure 1: Skandi Pacific and Figure 2: Atwood Osprey

Source: DOF Management Source: Atwood Oceanics

At about 0600 on 8 July, Skandi Pacific arrived on location at the rig. The vessel was scheduled to carry out cargo handling operations over the following days. Throughout that time, the vessel’s master and mates maintained a two person, 6-on/6-off schedule for navigational watches.

Over the next few days, Skandi Pacific cargo handling operations were conducted with the vessel in Dynamic Positioning^[3] (DP) mode.

The wind throughout this time was from the south-southeast at force^[4] 5 to 6 (17 to 27 knots).^[5]

On July 10, the weather deteriorated and winds increased to force 8 to 9 (34 to 47 knots) preventing Skandi Pacific from carrying out cargo handling operations. The adverse weather conditions continued over the following days and the vessel remained on standby off the rig.

On 13 July, Skandi Pacific’s master received Bureau of Meteorology (BoM) reports, including its commercial weather services report, forecasting that conditions would ease temporarily during the early morning of 14 July.

During the 1800-2400 watch on 13 July, Skandi Pacific was standing off the rig. The master noted in his night orders that cargo handling operations were due to commence after another offshore support vessel (OSV) had departed. In his orders, he also stated that ‘the weather was due to deteriorate again so keep this in mind’.

At 0010 on 14 July, the rig’s controller called Skandi Pacific’s master and instructed him to move into the 500 m exclusion zone off the rig and prepare to backload cargo (cargo transfer from the rig to the vessel). Shortly after, the master handed over the watch to the chief mate. The master then remained on the navigation bridge (bridge) and conducted a toolbox talk^[6] for the cargo work with the chief mate, second mate, and the two integrated ratings^[7] (IR) who were to be on deck during cargo work.^[8] This talk included instruction for crush hazard awareness, backloading cargo in block stows, escape routes and for the IRs to stop work if seas were shipped on deck.

Shortly thereafter, the master left the bridge. At that time, the wind was easterly at 15 to 25 knots with 2 to 3 m seas, consistent with the weather forecast.

Figure 3: Dampier and position of the Atwood Osprey

Source: Australian Hydrographic Service with ATSB annotations

At about 0025, Skandi Pacific’s chief mate moved the vessel closer to the rig, as planned, and placed it in DP mode. At 0045, after completing the DP checklist, he began moving it under the rig’s loading platform, which was on its leeward side. At about this time, the master briefly returned to the bridge. He noted the weather as acceptable for the operations, with occasional sea spray on deck.

By 0105, the chief mate completed the move and the two IRs were already on deck, releasing lashings in preparation for backloading empty mini-containers, open top containers and 10-foot sea containers from the rig.

At 0140, backloading of containers from the rig started and from time to time, there was spray across Skandi Pacific’s deck. The vessel’s motion in the seas meant that it moved around its position under the loading platform, but the DP system kept it within the set range (2 m) and on a true heading^[9] of between 237° and 238°.

Shortly before 0200, the IR starting his watch came to the bridge for the toolbox talk and signed the briefing form. He then proceeded to the aft deck and the IR completing the 2000-0200 watch handed over to him and left the deck.

At 0400, the wind was recorded in the vessel’s deck log book as northeast force 5 to 6 (17 to 27 knots). The sea state was recorded as 5 (that is, wave height 2.5 to 4 m or rough seas). In those weather conditions, waves were occasionally shipped on deck over the vessel’s open stern. This became more frequent over the next hour and, in the minutes before 0500, some bigger waves washed across a large part of the aft deck.

At 0501, the second of two mini-containers was landed at the forward part of the aft deck on the starboard side. The IRs released the crane hook from the container and moved clear. Between 0502 and 0503, two waves were shipped on deck in succession as the vessel pitched with the water washing forward about 40 m across the port side of the aft deck.

At 0503, the DP status alarm triggered indicating that Skandi Pacific had moved outside the set limit of 4 m. The chief mate informed the rig that it was ‘too rough’ with the vessel getting pushed ‘further out of position’ and ‘more and more water’ coming on deck, making it ‘dangerous’ for the operations and the men on deck. He asked the rig to stop backloading and then advised the IRs that the operations had been suspended. The deck log book entry stated ‘0505 Stop Job, weather increasing in strength’. Shortly after 0505, another large wave came over the stern.

By 0507, the chief mate had stepped Skandi Pacific about 30 m to leeward of the rig. The vessel remained in DP mode and on the same heading. He then instructed the IRs to lash the containers on deck.

Shortly thereafter, the two IRs moved aft to start lashing the containers. From the chief mate’s seated position at the DP console, facing aft, he could see the men as they moved around the aft end of the deck. They first lashed the cargo platforms on the port side with chains, run from the crash barrier at the stern of the vessel around the platforms. The chains were attached at the forward end of the deck to a tugger wire^[10] and tensioned with the port tugger winch.

By 0511, the IRs had completed lashing on the port side. During this time, a couple of smaller waves had been shipped on deck.

At 0512, the IRs moved to the starboard side cargo stow, which included open top, mini and 10-foot sea containers. They moved between the vessel’s stern and the forward part of the stow. A few smaller waves came over the stern during this time.

By about 0518, the IRs had lashed the cargo stow using the starboard winch to heave the tugger wire (in a similar manner to the port side) to tension the chains rigged around the containers. They then began checking the lashings starting from aft. At 0519, while they were aft at the starboard quarter, a wave came over the port quarter and washed across the port side.

At about 0520, when the IRs checked the forward part of the stow, they found the two forward mini-containers were not properly secured by the primary chain. They decided to secure them using a secondary chain secured to the crash barrier between the mini-containers and the skip, and then to the primary chain (Figure 4). When tightened, the secondary chain would tighten the primary chain against the mini-containers. That would require the primary chain to be slackened off (tensioned down – the term used on board the vessel).

Shortly thereafter, the tugger wire was payed out, which again unsecured the entire starboard cargo stow. No one informed the mates on the bridge that the primary chain had been tensioned down.

The IRs began rigging the secondary chain forward of the mini-containers. They decided to use a shackle looped around the primary chain and secured to the end of the secondary chain. This arrangement would allow the shackle to run along the primary chain and keep the tension in it.

At 0522, both IRs were preparing the secondary chain. Thirty seconds later, one IR went forward to the nearby store to get the shackle, leaving the other to connect the chain to the crash barrier.

Figure 4: Skandi Pacific’s aft deck securing chain arrangement

Source: Skandi Pacific’s CCTV at 0521 on 14 July 2015 (annotated by ATSB)

Moments before 0523, a large wave came over the stern and washed across the deck. The water on deck had not yet receded, when a larger wave came over the stern. From his position on the bridge, the chief mate had seen the seas coming over the stern and called out a warning to the men on deck over the UHF radio. The IR getting the shackle had the UHF radio while the one rigging the secondary chain had the VHF radio.

At 0523,^[11] as the large quantity of water washed forward along the deck, it shifted some of the starboard side cargo containers (Figure 5) and continued as far forward as the Skandi Pacific’s superstructure. The IR rigging the secondary chain just forward of the mini-containers tried to move clear of the advancing containers but did not have enough time to get clear. He was trapped in between the forward mini-container and the primary chain, and was crushed against a skip.

Figure 5: Aft deck plan showing stowage and other key positions

Source: DOF Management (annotated by ATSB)

The IR returning with the shackle heard the warning over his radio. When he reached the mini-containers, he saw the injured man. He immediately called the bridge and informed the chief mate that the other IR had been injured. The chief mate told the second mate on watch with him to call the master, and to take over the DP console. He then went to the accident area.

Shortly thereafter, the master arrived on the bridge, and the second mate informed him that an IR was pinned between the deck cargo. The master called the rig, advised that there had been an accident and requested medical assistance.

At 0527, the master sounded the general alarm followed by a public address to muster all crewmembers at their stations. He also asked the rig to arrange a helicopter for a medical evacuation

A short while later, the chief mate and other crewmembers freed the IR, who was unresponsive. The crewmembers then started cardiopulmonary resuscitation (CPR) on the seriously injured man and the master manoeuvred Skandi Pacific under the rig’s loading platform.

At 0605, three crew from Atwood Osprey, including two medics boarded Skandi Pacific. They continued CPR and moved the injured IR onto a stretcher in preparation for transfer to the rig.

At 0613, the injured IR was transferred to Atwood Osprey. He was then taken to the rig’s hospital, where, at 0630, the medics declared him deceased.

__________

1. All times referred to in this report are local time, Coordinated Universal Time (UTC) + 8 hours.
2. A nautical mile of 1852 m.
3. Dynamic Positioning is a computer-controlled system to automatically maintain a vessel's position and heading by adjusting its propulsion units to suit the weather and sea conditions.
4. The Beaufort scale of wind force, developed in 1805 by Admiral Sir Francis Beaufort, enables sailors to estimate wind speeds through visual observations of sea states.
5. One knot, or one nautical mile per hour equals 1.852 kilometres per hour.
6. The toolbox talk communicates the risk assessment and specific controls to the work party.
7. Integrated ratings are qualified to perform the duties of both an able seaman and an engine rating.
8. The IRs’ working hours consisted of sea watches, plus 2 hours either before or after their watch during cargo operations. On 13/14 July, the three IRs work routines were 2000-0200, 0000-0600 and 0200-0800. Additionally, a day work IR was available from 0600-1800.
9. All ship’s headings in this report are in degrees by gyro compass with negligible error.
10. Relatively small (15-20 mm) diameter wires, stowed on their own winches, one each side at the forward end of the working deck. The direction of pull is facilitated by passing the wires through snatch blocks, which can be secured in various places along the working deck.
11. The accident was recorded in the vessel’s deck log book at 0525, some 2 minutes later than the CCTV footage.

Context

Skandi Pacific

At the time of the accident, the anchor handling tug supply (AHTS) vessel, Skandi Pacific, was registered in the Bahamas, classed with Det Norske Veritas (DNV) and managed by DOF Management, Norway. The vessel was under charter to Chevron Australia to facilitate support operations for the semi-submersible oil rig, Atwood Osprey.

As an offshore support vessel (OSV) Skandi Pacific was designed for offshore operations including anchor handling and towing, and had a multi-national crew of 12, including the master. The master had 20 years of seagoing experience, including 5 years as master on OSVs. He had been assigned to the vessel for the past 2½ years and had joined it about 2 weeks before the accident.

The chief mate had 14 years of seagoing experience, including 7 years on offshore support vessels. He had been chief mate on Skandi Pacific for 1½ years as chief mate and had joined about 2 weeks before the accident.

The Integrated Ratings (IR) on deck at the time of the accident were suitably qualified in their respective ranks. Both IRs had experience at this rank on board OSVs and had joined Skandi Pacific about 2 weeks before the accident.

DP system

A computerised Dynamic Positioning (DP) system uses the vessel’s thrusters and propellers to control its position and heading. The DP system automatically adjusts the propulsion units to counteract the environmental forces (Figure 6) to keep the vessel on location at, or very near to a specified position.

Figure 6: Vesselpropulsion, thrusters, movement and external forces

Source: DOF Management (annotated by ATSB)

Accurate measurement of the vessel’s position at any point in time is necessary for precise DP. Additionally, as the vessel is subject to wind, wave and current forces, these forces also need to be accurately measured by the DP system. The system then controls the vessel’s motion in the three horizontal degrees of freedom:

• surge^[12]
• sway^[13]
• yaw.^[14]

The position reference systems consist of DGPS^[15], RADius^[16] and Fanbeam,^[17] of which all three reference systems are required for DP class 2^[18] operations.

Skandi Pacific was equipped with two stern thrusters, two bow thrusters and two controllable pitch propellers, designed to comply with a DP equipment class two system. The vessel’s position could also be maintained manually by using a single joystick to control all thrusters.

When the vessel was operating in DP mode, its status could be one of the following:

• Green status: normal operation indicating the position and heading were within predetermined limits.
• Yellow status: degraded DP control indicating position keeping was deteriorating and/or unstable (moving 2 m from the set position) and weather conditions becoming unsuitable for DP operation (the DP operator or DPO,^[19] master and rig supervisor should then discuss continued operations).
• Red status: indicating a DP emergency triggered by either by a system failure or external condition, preventing the vessel maintaining its position within 4 m from the set position.

Annual DP trial programme

Annual DP trials are a thorough test of a DP system to verify its capability and performance and to confirm it has been maintained in accordance with international requirements. Further, all major DP systems and sub-systems are tested to determine the ability of the vessel to maintain position after single system failures,^[20] associated with the assigned equipment class.

These tests confirm that the ‘failure modes and effects analysis’^[21] conducted during the design process of the vessel is still valid. Any new equipment or system upgrades are incorporated in all trials to ensure they react as intended.

DP capability plot

DP capability defines a vessel’s ability to remain in position under given environmental and operational conditions. The capability plots (Figure 7) are calculated for normal operation and for DP system failures, such as a loss of thruster, including worst case failures.

These plots are not of actual performance, but are based on calculations and provided by the manufacturer of the DP system. The calculated DP capabilities of the vessel can be used to determine if the vessel will remain in position during DP operations in the prevailing or forecast weather conditions.

Skandi Pacific’s DP capability plots (Figure 7) show the ‘normal operations’ and ‘single bow thruster failure’. The plots show the limiting wind speed 360 degree envelopes (blue line), where each point on the envelope represents the wind speed at which it is calculated that the vessel will be unable to maintain position in DP mode. With a single bow thruster failure, the vessel had reduced capability when winds exceeded 60 knots.

Figure 7: Skandi Pacific's DP capability plots

Source: Skandi Pacific (annotations by ATSB)

DP system check

Offshore facilities are protected by the establishment of an exclusion safety zone^[22] around the structure and entry into the exclusion safety zone is controlled by the facility.

Prior to entering the exclusion zone at any facility, all vessels are required to complete a pre-entry checklist. The purpose is to ensure satisfactory operation of the vessel’s DP system, including full functional checks of the operation of the thrusters, power generation, automatic DP and joystick/manual control.

Additionally, when departing the facility, the vessel is to be manoeuvred well clear of the facility before changing operating modes. The Guidelines for Offshore Marine Operations (G-OMO)^[23] recommend that this distance is usually between 1½ to 2½ vessel lengths, depending on the drift conditions.

Operational risk management

The DOF Management operational risk management procedures are mainly based on best practice and the G-OMO. These procedures included the following guidance.

Risk assessment

Risk Assessments (RA) are used to identify and then mitigate risks to an acceptable level. If the risks cannot be mitigated to an acceptable level, the work should not proceed.

The RAs on board Skandi Pacific included details of the task, the team members, activity description, identified hazards, the effects of those hazards, existing controls and additional control measures.

Toolbox talks

A toolbox talk (TBT) is a meeting of the personnel involved in an imminent task to review the task, individual responsibilities, equipment required, competency of the individuals, hazards, and any RAs in place.

On board Skandi Pacific, all personnel involved in cargo handling operations attended TBTs. The objective was to communicate the RA and to capture any additional controls not already identified. This included (but was not limited to):

• individual roles
• tools, methods and procedures to be used
• a review of the RA
• promote the ‘stop the job’^[24] culture.

Management of change

The company had a management of change (MOC) process in place for all tasks. This process was used when unexpected changes in circumstance occurred before or during the course of the task. It was then used to identify the change type, details, criticality and what action was required now that the change had occurred. The task would not continue until the implications of that change were reviewed. If appropriate, the RA would be reviewed before resumption of the task and a TBT carried out. For example, the loss of the use of a bow thruster on board Skandi Pacific changed the operational capability of the vessel. Therefore the MOC process was used to define and assess subsequent operational limitations.

Cargo stowage and securing

There are many sources of relevant guidance for cargo handling on board OSVs. The International Maritime Organization’s (IMO) Resolution A.863(20)^[25] states the following for OSVs with an open stern:

vessels are provided with instructions to counter dangerous situations if cargoes with a tendency to float and/or with low friction coefficients are stowed on the exposed deck

the number of cargo handlers should be sufficient for safe and effective cargo operations

the crew of OSVs should be adequately trained

safe havens and escape routes for personnel from the cargo deck should be properly marked and kept clear at all times. A crash barrier, fitted along each side of the deck, could be one method of achieving a safe haven.

Further guidance is contained in Regulation 5, Chapter VI, SOLAS^[26] (Stowage and securing) which specifically references all cargoes, cargo units and cargo transport units shall be loaded, stowed and secured in accordance with the Cargo Securing Manual (CSM).

The Cargo Securing Manual^[27] (CSM) is to include information on a number of pre-identified items:

details of fixed securing arrangements and their locations

examples of correct application of portable securing gear on various cargo units, carried on the vessel.

Skandi Pacific’s CSM contained the above procedures for cargo handling alongside installations and included the following guidance for open stern AHTS vessels:

Open stern anchor handling vessels require special care, especially with regards to freeboard. Consideration should be given to the open stern being physically barriered. Use RA and TBT to minimise crew or cargo exposure to elements, particularly when working with the stern towards the weather.

Additionally, the IMO Resolution A.717(17), Code of Safe Practice for Cargo Stowage and Securing, states:

The proper stowage and securing of cargoes is of the utmost importance for the safety of life at sea. Improper stowage and securing of cargoes has resulted in numerous serious vessel casualties and caused injury and loss of life, not only at sea but also during loading and discharge.

Maritime boundaries

In 1994, the Australia government ratified the United Nations Convention on the Law of the Sea (UNCLOS) and became legally bound to its provisions. As such, Australia is bound to follow UNCLOS rules for maritime boundaries, access to the various marine zones and managing the resources and activities within those boundaries.

Figure 8: Maritime boundaries off the North West Shelf, Western Australia

Source: Geoscience Australia (annotated by ATSB)

To determine the maritime boundaries (Figure 8) requires the delineation of the Territorial Sea Baseline^[28] (TSB). The TSB is the line from which the outer limits of a number of maritime zones are calculated and include the:

• 3 nautical mile limit of the coastal waters^[29]
• 12 nautical mile limit of the territorial sea^[30]
• 24 nautical mile limit of the contiguous zone^[31]
• 200 nautical mile limit of the Australian Exclusive Economic Zone.^[32]

Jurisdiction

The Offshore Constitutional Settlement (OCS) is an agreement between the Commonwealth and the States, which provides the basis for an agreed division of powers in relation to coastal waters and other matters. These include the regulation of shipping and navigation, offshore^[33] petroleum exploration, crimes at sea, and fisheries.

National Offshore Petroleum Safety and Environmental Management Authority

The National Offshore Petroleum Safety and Environmental Management Authority’s (NOPSEMA) functions and powers are derived from the Offshore Petroleum and Greenhouse Gas Storage Act 2006 (OPGGS Act) and associated regulations. The OPGGS Act provides that NOPSEMA is charged with regulating health and safety, well integrity and environmental management for all offshore petroleum facilities and activities in Commonwealth waters.^[34] Additionally, it has similar functions where State and Territory powers to regulate have been transferred, covering the area within the first three nautical miles for coastal waters.

The Australian Maritime Safety Authority

The Australian Maritime Safety Authority (AMSA) is charged with the responsibility for ensuring the safety of Australian flagged vessels and foreign flagged vessels in Australian ports and waters, and the seafarers on board. International and national conventions and regulatory frameworks provide the standards enforced by AMSA to achieve safety outcomes. 

According to AMSA, it delivers necessary safety outcomes through inspections such as those under Port State Control, flag State requirements, Cargo, Maritime Labour Convention, and Occupational Health and Safety (OHS) when the ship is in port.

Trained AMSA OHS inspectors perform investigations of accidents and dangerous occurrences on ‘prescribed ships’^[35]:

• in an Australian port
• entering or leaving an Australian port
• in the internal and territorial waters of Australia.

Skandi Pacific was not covered under the above requirements and, hence, Australian OHS legislation did not apply to the vessel.

Memorandum of Understanding between AMSA and NOPSEMA

The Memorandum of Understanding (MoU) between the two authorities provides for the cooperation of both parties in the administration of their respective commitments including audits, inspections and accident investigations of offshore facilities. It delivers a consistent and comprehensive regulatory regime in offshore waters and avoids duplication in respect of vessels and facilities where the two organisations have regulatory obligations. The MoU was last revised in 2013, when AMSA and NOPSEMA agreed to further joint inspections of floating facilities.

At the time of the accident, Skandi Pacific was not engaged in cargo handling operations with Atwood Osprey. The vessel was also outside the territorial waters of Australia. Its location and status meant that it was outside the jurisdiction of both NOPSEMA and AMSA.

Safety investigations

The IMO’s Casualty Investigation Code is mandatory under SOLAS and requires the flag State^[36] of a vessel to conduct a marine safety investigation into every ‘very serious marine casualty’.^[37] The Bahamas, as the flag State of Skandi Pacific was obliged to investigate the fatal accident on 14 July. The Bahamas Maritime Authority commenced a safety investigation under their national legislation.

The ATSB, as a substantially interested State,^[38] also commenced a safety investigation.

Similar past accidents

The ATSB’s predecessor, the Marine Incident Investigation Unit (MIIU), investigated a similar fatal accident on board an OSV in 1995.

Shelf Supporter was a 1985-built OSV equipped with a DP system. The vessel had a large, clear aft deck that enabled it to carry stores and equipment to/from offshore facilities. It operated off the north-west coast of Australia.

At about 0550 on 29 December 1995, Shelf Supporter’s master manoeuvred the OSV to approach an offshore platform’s crane position with its bow in (that is, stern to sea). It had cargo on the aft deck, which had been secured by the vessel’s crewmembers using tugger wires and winches.

Shortly after 0600, two crewmembers went out to the aft deck to prepare for unloading cargo. They released the tugger wires and re-spooled the starboard wire onto the winch drum. However, before they could re-spool the port wire, the vessel arrived under the crane and its lifting hook was lowered just above their heads. They left the port wire flaked on deck, and hooked on the first lift.

On the bridge, the master was maintaining the vessel’s position and saw a wave breaking over the stern. He broadcast a warning to the two men on deck. As the water ran up the deck, he saw one of the men just forward of a skip. He then saw the skip move forward and crush the man against another skip.

The following summary of some of the MIIU’s findings is particularly relevant:

• OSVs frequently work in moderate and even rough seas. Their design with a long low aft deck, generally with no protective stern bulwark makes seas breaking over the stern common place.
• Sea conditions at the time of the accident were not such as to warrant cancelling operations.
• Although a warning may be called out to crewmembers on deck, they have little time to react. If a sea breaks over the stern and they do not see it approaching, their reactions must be instinctive.
• The securing of cargo in block stows with the tugger wires is common practice in the offshore industry. It is simple and easy to set up and quick to release. However, it has a big disadvantage because once removed, all of the deck cargo is unsecured until discharged.

In June 2016, the International Marine Contractors Association (IMCA)^[39] promulgated details of a recent incident involving cargo shift on an OSV’s aft deck in heavy weather while alongside a platform.

The OSV was engaged in cargo handling operations with its starboard quarter to the weather when it experienced a sudden and unexpected squall. A large wave was shipped and flooded the aft deck. The water shifted one container and turned another one over. The backloaded containers had not yet been secured due to on-going backloading. Fortunately, no one was injured.

The lessons learned included:

• The risk of abnormal waves should be taken into consideration in risk assessments and tool box talks for work, particularly when the vessel is positioned stern to the weather
• Greater emphasis should be placed on the ‘stop work policy’ – anyone should be able to stop the job, any time, when in doubt
• A new and higher bulwark to be fitted to the vessel and other similar vessels.

__________

12. The bodily motion of a ship in a seaway forward and back along the longitudinal axis.
13. The side-to-side bodily motion of a ship in a seaway, independently of rolling, caused by a uniform pressure being exerted all along one side of the hull.
14. The horizontal oscillation of a ship about a vertical axis approximately through its centre of gravity.
15. Differential Global Positioning System.
16. RADius is a short distance reference system consisting of transponders installed as targets, in this case on the platform and sensors on the OSV vessel.
17. Fanbeam is a short range laser scanning radar system.
18. DP equipment Class 2 has redundancy so that no single fault in an active system will cause the system to fail. Loss of position should not occur from a single fault of an active component or system such as generators, thruster, switchboards, remote controlled valves etc.
19. The designated watchkeeping officer responsible for managing the dynamic positioning of the vessel.
20. Single failure criteria include any active component or system (such as generators, thrusters, switchboards or remote controlled valves), together with any normally static component (such as, cables, pipelines or manual valves) that cannot be shown to have adequate protection from damage or have proven reliability.
21. A systematic analysis of the systems to demonstrate that no single system failure will cause an undesired event.
22. Established within a radius extending to a distance determined by the relevant legislations beyond the outline of any installation, excluding submarine pipelines (NOPSEMA).
23. The G-OMO provides guidance in the best practices which should be adopted to ensure the safety of personnel on board all vessels servicing and supporting offshore facilities, and to reduce the risks associated with such operations.
24. Stopping The Job is one in which everyone has the right to stop the job, the duty to stop the job and the moral responsibility to stop the job.
25. The Code of Safe Practice for the Carriage of Cargoes and Persons by Offshore Support Vessels (OSV Code).
26. The International Convention for the Safety of Life at Sea, 1974, as amended.
27. IMO Resolution A.489(XII), adopted on 19 November 1981, Safe Stowage And Securing Of Cargo Units And Other Entities In Ships Other Than Cellular Container Ships Cargo Securing Manual.
28. The TSB corresponds with the low water line along the coast, including the coasts of islands. The baseline can be drawn around low tide elevations which are defined as naturally formed areas of land surrounded by and above water at low tide but submerged at high tide, provided they are wholly or partly within 12 nautical miles of the coast. For Australian purposes, the baseline corresponds to the level of Lowest Astronomical Tide (LAT)
29. Coastal Waters is a belt of water between the limits of the Australian States and the Northern Territory and a line 3 nm seaward of the territorial sea baseline. Jurisdiction over the water column and the subjacent seabed is vested in the adjacent State or Territory as if the area formed part of that State or Territory.
30. The Territorial Sea is a belt of water not exceeding 12 nm in width measured from the territorial sea baseline. The major limitation on Australia's exercise of sovereignty in the territorial sea is the right of innocent passage for foreign ships.
31. The Contiguous Zone is a belt of water contiguous to the territorial sea, the outer limit of which does not exceed 24 nm from the territorial sea baseline.
32. The Exclusive Economic Zone (EEZ) is an area beyond and adjacent to the territorial sea. The outer limit of the EEZ cannot exceed 200 nm from the baseline.
33. An offshore area starts 3 nm from the TSB from which the breadth of the territorial sea is measured and extends seaward to the outer limits of the continental shelf.
34. The Commonwealth marine area is any part of the sea, including the waters, seabed, and airspace, within Australia’s exclusive economic zone and/or over the continental shelf of Australia, that is not State or Northern Territory waters.
35. A prescribed ship is a ship registered in Australia, a ship engaged in coastal trading, a ship on which the majority of crewmembers are residents of Australia and which are operated by persons or firms which have their principal place of business in Australia or a ship declared by the Minister to be a prescribed ship.
36. Flag State means a State whose flag a ship is entitled to fly.
37. A very serious marine casualty means a marine casualty involving the total loss of the ship or a death or severe damage to the environment.
38. A substantially interested State means a State where, as a result of a marine casualty, nationals of that State lost their lives or received serious injuries.
39. The IMCA is a trade association representing companies and organisations engaged in delivering offshore, marine and underwater solutions.

Safety analysis

The accident

At 0505 on 14 July, Skandi Pacific’s chief mate stopped backloading after the vessel’s DP system was no longer able to maintain its positon within defined parameters in the deteriorating weather conditions. In the rough seas, waves were regularly coming on deck over the vessel’s open stern.

By 0507, the chief mate had stepped Skandi Pacific out, to about 30 m off the rig on its leeward side, in DP mode. He then instructed the IRs to lash the containers on the aft deck.

At about 0520, after the IRs had lashed the cargo they found two mini-containers on the starboard side forward, not properly secured by the primary chain. In order to better secure the containers, the IRs decided to use a secondary chain secured to the crash barrier and then to the primary chain. Tightening the secondary chain would tighten the primary chain against the mini-containers. Their plan required tensioning down the tugger wire (and hence the primary chain). When they slackened the chain, the block of cargo on the starboard side was unsecured.

At 0523, two large waves came over the vessel’s stern in quick succession, shifting the unsecured containers on the starboard side. Their movement caused the primary chain to tension (Figure 9). As a result, the slack chain was lifted up, blocking the escape route of the IR who had stepped inside the primary chain to secure the secondary chain to the crash barrier. Although he tried to move clear of the advancing containers, in the few moments he had (to get clear) his escape (route) was impeded by the chain and he could not get clear.

Figure 9: Diagrams to illustrate the event sequence leading to the accident

Source: Cargo manifest (annotated by ATSB)

Prevailing weather conditions

At 1500 on 13 July, a BoM weather forecast (Figure 10) was issued for the general area where the Atwood Osprey platform is located:

A high is stalling over the southwest coast of WA. The ridging associated with the high will keep a relatively tight E/SE’ly pressure gradient over the Pilbara and adjacent waters. Expect fresh to strong winds through the forecast period.

Monday 13 July Confidence High, Partly Cloudy

• Mean winds 25/30 knots (possibly up to 32 knots)
• Total wave 3.1 m to 3.5 m (possibly up to 4.0 m)

Tuesday 14 July Confidence High, Partly Cloudy, Mostly sunny

• Mean winds 23/28 knots (possibly up to 32 knots)
• Total wave 2.6 m to 2.9 m (possibly up to 3.5 m)

The forecast showed a temporary lull in the wind speed and sea/swell conditions during the early hours of the following day, 14 July.

Figure 10: BOM’s commercial weather services forecast for 13/14 July 2015

Source: Skandi Pacific

Later in the afternoon on 13 July, another OSV started cargo handling operations on the lee side of Atwood Osprey. Shortly after 1800, the rig’s controller instructed Skandi Pacific’s master that his vessel would be called in next, about midnight, after the other vessel had departed.

The master continued monitoring the weather conditions throughout his 1800-2400 watch. The DP capability plot (Figure 7) also indicated that the vessel could maintain position in the conditions. Therefore, he noted in his night orders to the watchkeeping officers:

Make sure you are happy with position keeping and vessel movement before starting any ops.

There is quite some backload.

Stow as well as possible i.e. good blocks to secure for transit back to port. Weather is due to deteriorate again so keep this in mind.

Call me if in doubt.

Standing orders apply.

Skandi Pacific’s CSM provided guidance based on G-OMO, which is considered to be the best practice for offloading and backloading cargo alongside offshore installations, including:

Weather Conditions

• Weather conditions must be appropriate for the operation
• Always stop work in adverse weather conditions
• Comply with relevant guidelines on weather criteria.

However, consideration of the weather conditions and what could be deemed ‘adverse weather’ is subjective. Further, the company’s safety management system (SMS) did not provide any clearly defined guidance on weather limits for undertaking deck cargo handling operations or triggers to stop operations. The only such guidance was contained in Skandi Pacific’s charterer’s marine operating guidelines for bulk cargo operations:^[40]

Weather Limit Guidelines

Provided proper preparations have been made, the appropriate bulk transfer checklist has been completed and all other factors considered and found favourable, the following weather limits are recommended for conducting bulk transfers between vessels and an offshore facility:

• Wind < 30 knots / 15 m/s
• Sea < 3 metres (significant wave height)^[41]

The master used these guidelines to determine that the cargo handling operations could proceed. At midnight on 13 July, the forecast weather conditions (wind speed and significant wave height) were within the above weather limits.

The company’s SMS did not contain any clearly defined limits or parameters for cargo handling operations in adverse weather conditions. To determine if the conditions were appropriate for operations was at the discretion of the master. On 14 July, one of the conditions that he stipulated was to stop work if there was water on deck.

Clearly defined limits in the SMS would have removed subjectivity from the decision making and established a common understanding for everyone involved in cargo handling operations.

Cargo handling operations on 14 July

Toolbox Talk

On July 14, Skandi Pacific’s master outlined the main points of the toolbox talk (TBT) for cargo operations to the persons involved in the cargo handling operations, that is, the IRs and 0000-0600 mates. The briefing outlined a number of pre-determined safety topics, including:

Type of operation, work equipment, methods/procedures to be adopted, communication, and human factor assessment, ‘Stop the Job’ policy, crane/lifting requirements, individual responsibilities for controls, access, manual handling, working environmental conditions, potential hazardous..., …‘good blocks stows’ and ‘lash cargo well.’

The weather conditions were also discussed and the master instructed the IRs and OOWs that if water was shipped on deck, then the work was to be stopped. All of them verbally acknowledged this instruction and then signed the TBT form. Further, before starting his duties, the IR on the 0200-0800 watch received a briefing from the chief mate and signed the TBT form to acknowledge he had been shown all the required provisions of the earlier TBT.

Water on deck

Between 0140 and 0502, seas were occasionally shipped on deck, increasing in regularity and size before 0500. Between 0502 and 0503, two large waves were shipped over the aft deck in quick succession, followed by another one at 0505 (Figure 11).

Figure 11: Seas washing across the aft deck at 0502 and 0505

Source: Skandi Pacific’s CCTV (annotations by ATSB)

Between 0507 and the time of the accident, after Skandi Pacific had stepped out from the rig, waves continued to be shipped on the aft deck from time to time. However, the IRs continued working, moving around the deck. At 0519, while they were at the aft end of the deck on the starboard side, a wave washed over the port side (Figure 12).

Figure 12: The CCTV frame at 0519

Source: Skandi Pacific’s CCTV (annotations by ATSB)

The design of the AHTS vessel with long, low aft decks without protective stern bulwarks meant that spray on deck and seas breaking over the stern were common. However, neither the mates nor the IRs on deck considered the water shipped on deck from 0502 onwards prevented the crew from lashing the cargo in those circumstances and conditions.

After the accident, the 0200-0800 IR who was also on deck stated that ‘at no time did he feel that we were working in unsafe weather conditions during any of the operations’. Hence, he did not consider stopping the job. This indicates that the definition of conditions when work was to be stopped, such as excessive water on deck, was subject to a person’s judgment and decision. The IRs statement suggests that there was a sense of security despite the worsening conditions.

Had the SMS contained clearly defined limits for excessive water on deck that necessitated stopping operations, the crewmembers would have been in a better position to make safe decisions. For example, the chief mate did stop backloading shortly after the DP status limit was exceeded. He had informed the rig that it was too rough with more water coming on deck making it dangerous for those operations and his crew. While the chief mate probably felt the need to ensure the cargo on deck was lashed, the risks to the men working on deck did not decrease by moving the vessel 30 m away from the rig. If anything, those risks increased as they moved around the deck lashing cargo in deteriorating weather conditions.

In submission to the draft investigation report, DOF management stated:

… There was no forewarning of the 2 unexpected waves, which occurred in close succession, and caused the incident.

However, between 0450 and the time of the accident, progressively larger waves were shipped on the vessel’s deck. The CCTV footage shows that during this period of about 33 minutes, at least 15 waves large enough to wash across the deck were shipped. When the vessel’s stern pitched into the wave coming over the stern, more water came on deck.

Backloading

From the commencement of cargo handling operations, six items were transferred to Atwood Osprey and nine items backloaded to Skandi Pacific. Figure 13 details the positions, relative sizes and types of cargo on deck after the cargo handling operations were suspended.

The shipboard CSM included guidance for crewmembers when backloading cargo:

• be aware that as soon as sea fastenings are released, there is a chance for the cargo to shift
• be aware of the potential of trapped hands between the tugger wire and cargo
• be aware that cargo can move when tightened by the tugger.

Securing mini-containers

It follows that if unsecured cargo can shift then safe practice is to minimise the amount of time the cargo is unsecured. The fact that cargo will be unsecured at times and be subject to external forces possibly leading to its movement should be identified through an appropriate RA.

In submission to the draft investigation report, DOF management stated:

Cargo is always unsecured for a period when backloading occurs. The Risk Assessment specifically dealt with the hazards posed by shifting cargo and stipulated additional control measures. The Toolbox Talk specifically provided for the cargo to be well lashed before departure.

However, the RA for loading deck cargo was completed before the cargo handling operations started. The identified hazard of shifting containers was to be mitigated by ‘Staying away from containers. Avoid enclosed areas’. After cargo handling operations were stopped at 0505, the RA was not reviewed nor were the risks associated with securing cargo in the prevailing conditions adequately assessed.

By about 0518, the IRs had hauled in the starboard tugger wire and tensioned the primary chain around the cargo. When checking the lashing, however, they found that forward two mini-containers were not secured and decided to use a secondary chain to secure them.

Figure 13: Stowage arrangement after cargo operations were stopped on 13 July

Source: Skandi Pacific’s master (annotated by ATSB)

Their plan involved tensioning down the primary chain and securing a secondary chain to it. They decided to shackle the end of the secondary chain to the primary, to allow it to run along without snagging (Figure 14).

Figure 14: Utilising a secondary chain to secure cargo deck

Source: DOF management (annotations by ATSB)

Shortly afterwards, one of the IRs left the aft deck to get the shackle. Meanwhile, at the container, the other IR remained in the crush zone on the starboard forward end of the deck. He was in between the primary chain and the forward mini-container, inboard and outside of the crash barrier (Figure 9).

While the intent of the plan was to secure the cargo properly, the risks associated with the sequence of intended actions were not properly assessed. Importantly, the risk of working near unrestrained cargo that would be more susceptible to shifting as the weather conditions deteriorated, was not assessed. Furthermore, they did not comply with the master’s instruction to stop work if water was shipped on deck.

In submission to the draft investigation report, DOF management stated:

No prior assessment of the risk or any policy or safety procedure would have prevented the incident.

However, the need to secure cargo was a foreseeable risk. Securing the cargo needed to be undertaken after making conditions safe for the men to work on deck. At about 0520, the risk increased when the tugger wire was tensioned down, leaving containers 14, 10, 9 and 11 unrestrained. The risk further increased when the IR stepped into the crush zone forward of the cargo stow. At 0523, when the two large waves were shipped, the entire stow moved forward.

Essentially, the dangerous conditions while the IRs were securing the cargo were compounded when the primary chain was tensioned down and the IR stepped into a position from where he did not have a clear and unobstructed escape route.

Suspending cargo handling operations

Skandi Pacific was alongside the rig for about 4 hours on 14 July. The vessel’s station keeping status deteriorated to yellow several times, as it shifted more than 2 m from the set position.

At 0503, an alarm activated indicating the DP system status was red (that is, the vessel was more than 4 m from its set position). The chief mate stopped operations as required by the procedures and informed the rig. However, no one called the master to inform him of the suspended cargo handling operations nor discuss the current situation and changed activities on deck.

The chief mate moved Skandi Pacific, in DP mode, about 30 m away from the rig, stayed in the lee of Atwood Osprey, and maintained a heading of between 237° and 238°. The sea conditions continued to deteriorate and waves continued to be shipped over the vessel’s open stern.

The chief mate then instructed the IRs to secure the cargo with the vessel in this position. However, the risks of securing cargo with rough seas on the quarter and the vessel pitching heavily at times and shipping seas on deck were not adequately assessed or reviewed. Further, the master’s instruction to stop work if water was shipped on deck was not complied with.

Amongst the options the chief mate indicated he had considered were running with the weather until the cargo was secured or stepping out from the rig and staying in the lee until the cargo was secured. He had taken the latter option as in his opinion staying in the sheltered position off the rig was the safest option.

However, with rough seas on the quarter, the vessel pitching heavily at times and shipping seas on deck, it would have been prudent (and necessary) for the chief mate to call the master and discuss cargo securing options with him and the IRs. In other words, a review of the previous risk assessment or a reassessment. Calling the master would also have provided him the opportunity to reiterate or clarify his instruction that work be stopped if there was water on deck.

Had the chief mate called the master, other options to secure the cargo could have been considered. For example, in this situation the master indicated that he would have also considered running with the weather or tensioning the tugger wire around the cargo, then turning the vessel to protect the crewmembers on deck. As a result, the opportunity for the master to consider available options to prevent or reduce the likelihood of waves being shipped over the stern was lost.

Supervision and communication

Skandi Pacific’s risk assessment for cargo transfer stated ‘Minimum of two competent persons to be on the bridge, one solely in charge of the DP, one monitoring communications and watching around the vessel for dangers. However, at about 0520 on 14 July, the chief mate, who was solely in charge of the DP system, was performing both tasks. At that time, the second mate on duty was in the forward part of the bridge entering the passage plan to Dampier into the electronic chart display and information system.

Communications between the bridge and the IRs on deck was maintained via UHF and VHF radio. At the time of the accident, the IR securing the secondary chain had the VHF radio for communications with the bridge and the rig’s crane. The IR who went to get the shackle had the UHF for direct communications with the bridge.

The risk assessment also required two men to be on deck when carrying out cargo operations. At interview, the master stated this was partly to ensure that no individual went into a crush zone without another watching for water coming over the stern and also to watch out for the other. However, this situation occurred shortly before 0523 when the IRs were separated as one IR went to the deck locker (located about 10 m forward) to get the shackle. It was during this short period of time (30 seconds) that seas broke over the stern and shifted the containers.

While the chief mate warned the deck crewmembers over the UHF radio ‘watch out, water on deck’, the IR working on the secondary chain did not have a UHF radio. In any case, even if he had heard the warning, he would have had little time to react.

Skandi Pacific’s procedures for cargo handling alongside installations included the following for backloading operations:

The OOW in charge of backloading should always have full sight of all cargo operations and personnel on deck including the crane wire and hook.

Figure 15: Line of sight of Skandi Pacific’s aft deck from the Bridge

Source: DOF Management (annotated by ATSB)

After the IRs had initially tensioned the tugger wire at about 0518, the chief mate observing from the DP console (Figure 15) on the bridge, believed that the cargo handling operations had been completed. After the accident, the chief mate stated ‘I thought they were almost finished as they were both out of sight and I could see the chains tightening’.

The aft deck had CCTV coverage and was visible on the monitors above the DP console. In this way it was possible to monitor the status of the work as far forward as the tugger winch including the area in which the IRs were working. However, the chief mate was the DPO and his responsibilities operating the DP system did not allow him to continuously monitor the CCTV screen. Furthermore, he was not aware of the IRs plans and intentions to secure the mini-containers as they did not inform the mates on the bridge.

In submission to the draft report, DOF management stated:

The IRs are not required to report that information to the mates. That was work which was performed by the IRs in the ordinary course of operations.

However, lashing cargo in the prevailing conditions with rough seas on the quarter with the vessel pitching heavily at times and shipping water on deck, was not the ‘ordinary course of operations’. Working in those high risk conditions was also contrary to the master’s instructions. The risks further increased when the IRs tensioned down the primary chain at about 0520 and then started rigging the secondary chain. At 0522, both men were working forward of the mini-container. Had one of them not left to get a shackle, the consequences of the accident could have been worse.

In any case, the purpose of the IRs carrying the two radios clearly was two-way communication to mitigate any risks by having a common understanding of the work/situation and for important messages, including warnings to be issued.

Exposed aft deck

Skandi Pacific’s aft deck had an open stern, which permitted seas to be shipped. The shipboard operating procedures for cargo handling alongside installations identified the exposed deck as follows:

Open stern anchor handling vessels require special care and consideration should be given to the open stern being physically barriered.

The procedures also stated the crewmembers and cargo were exposed to the elements when working with the stern towards the weather. The guidance stated that a RA and TBT were to be used to minimise their exposure to the weather.

However, Skandi Pacific’s managers had not adequately assessed the risks associated with working on the aft deck of vessels with open sterns, including the consideration of engineering controls to minimise shipping water on the vessel’s aft deck.

DP status

On 8 July, Skandi Pacific’s master submitted a management of change (MOC) form for a change of situation because number two bow thruster was out of service. This meant that the vessel no longer met DP class 2 status requirements.

The MOC form noted:

• Conditions of working alongside installation to be assessed by master and rig’s OIM prior to entry of 500 m zone and deemed acceptable.
• Manual manoeuvring if weather is acceptable but there is a force pushing on, however, resultant force must be drift off. No hose work in this instant.
• DP class 1^[42] criteria for leeside working.
• RA to be done if hose work is intended stern should be into weather to reduce effect of loss of bow thruster.

The change in operational conditions necessitated a new or revised RA to be completed before work could be started.

The MOC was signed by the master, vessel manager and marine advisor. An additional RA was then undertaken, for working on an installation with DP class 1 status, when on the lee side of the rig. The activities assessed were setting up DP, DP work on lee side and deck cargo handling operations.

The risk level was assessed as ‘medium’^[43] which required additional control measures be implemented to reduce the residual risk level to ‘low’.^[44]

The measures put in place following the failure of the bow thruster were appropriate in the prevailing circumstances. Skandi Pacific was manoeuvred into position off the oil rig, stern in, on a heading of between 237° and 238°, similar to the previous OSV. The vessel maintained position on station at the rig until 0505, when operations were stopped by the chief mate.

__________

40. Bulk cargo operations involves the transfer of potentially hazardous liquids by use of hoses between vessel and rig.
41. The average of the highest one-third of a set of measured waves. The maximum wave height can be up to twice the significant wave height.
42. DP equipment class 1 has no redundancy. Loss of position may occur in the event of a single fault.
43. Task should only proceed with appropriate management authorisation after consultation with special personnel and assessment team. Where possible the task should be redefined to take account of the hazards involved or the risk should be recorded further prior to task commencement.
44. The task may be acceptable, however, it should be reviewed to determine if the risk can be reduced further.

Findings

From the evidence available, the following findings are made with respect to the crewmember fatality on board Skandi Pacific, at sea on 14 July, 2015. These findings should not be read as apportioning blame or liability to any particular organisation or individual.

Safety issues, or system problems, are highlighted in bold to emphasise their importance. A safety issue is an event or condition that increases safety risk and (a) can reasonably be regarded as having the potential to adversely affect the safety of future operations, and (b) is a characteristic of an organisation or a system, rather than a characteristic of a specific individual, or characteristic of an operating environment at a specific point in time.

Contributing factors

• Skandi Pacific’s officer of the watch suspended cargo handling operations due to adverse weather conditions, including water on deck, and moved the vessel off the oil rig to secure cargo on its aft deck.
• The task of securing the cargo after operations were stopped and the options to safely complete it were not adequately assessed.
• The vessel remained on the same heading in dynamic positioning (DP) mode because the officer of the watch considered this the safest option instead of others, such as running with the weather.
• The master was not called and remained unaware of the situation and, hence, did not get the opportunity to consider safer options for securing cargo, such as running with the weather or changing the vessel’s heading.
• Working on deck when water was being shipped was contrary to the master’s instructions.
• In an attempt to secure cargo effectively, the vessel’s crewmembers tensioned down the primary securing chain, which left the entire block of cargo unsecured.
• The crewmembers on deck carried out their plans in isolation without the active involvement of the officers on watch.
• A crewmember was standing in a position of danger when attempting to fasten a securing chain forward of a block of cargo.
• A large wave came over the vessel’s stern and shifted the block of cargo, which crushed the crewmember, fatally injuring him.
• While it was required to have two men on the working deck, at the time of the accident, the men on deck were separated for about 30 seconds, after one of them went forward to the deck locker to get a shackle to resecure the cargo.
• Skandi Pacific’s safety management system (SMS) procedures for cargo handling in adverse weather conditions were inadequate.Clearly defined weather limits when cargo handling operations could be undertaken and trigger points for suspending operations were not defined, including limits for excessive water on deck. [Safety issue]
• Skandi Pacific’s SMS procedures for cargo securing were inadequate. There was no guidance for methods of securing cargo in adverse weather conditions. [Safety issue]

Other factors that increased risk

• Skandi Pacific’s managers had not adequately assessed the risks associated with working on the aft deck of vessels with open sterns, including consideration of engineering controls to minimise water being shipped on the aft deck. [Safety issue]
• At the time of the accident, the officer who was solely in charge of the DP system was performing the tasks of both the DP officer and the communications/monitoring officer, contrary to the risk assessment for cargo transfers. The other officer on the bridge was occupied entering the vessel’s next passage plan into the electronic chart display and information system.
• During the operations leading up to the accident, the officers on the bridge did not keep the men on the aft deck in sight at all times as required by the vessel’s cargo handling procedures. Further, the radio communication between the men on deck and the officers was inadequate.

Other findings

• While one of Skandi Pacific’s bow thrusters was not operational at the time of the accident, the ‘management of change’ (MoC) process implemented as a result of its failure was appropriate. The vessel was capable of conducting operations within the limitations/conditions defined by the MoC process, and did so while on station at and off the rig.

Safety issues and actions

The safety issues identified during this investigation are listed in the Findings and Safety issues and actions sections of this report. The Australian Transport Safety Bureau (ATSB) expects that all safety issues identified by the investigation should be addressed by the relevant organisation(s). In addressing those issues, the ATSB prefers to encourage relevant organisation(s) to proactively initiate safety action, rather than to issue formal safety recommendations or safety advisory notices.

All of the directly involved parties were provided with a draft report and invited to provide submissions. As part of that process, each organisation was asked to communicate what safety actions, if any, they had carried out or were planning to carry out in relation to each safety issue relevant to their organisation.

Where relevant, safety issues and actions will be updated on the ATSB website as information comes to hand. The initial public version of these safety issues and actions are in PDF on the ATSB website.

Cargo handling procedures

Safety issue description:

Skandi Pacific’s safety management system (SMS) procedures for cargo handling in adverse weather conditions were inadequate. Clearly defined weather limits when cargo handling operations could be undertaken and trigger points for suspending operations were not defined, including limits for excessive water on deck.

Proactive safety action taken by DOF Management

Action number: MO-2015-005-NSA-003

Following the accident, DOF management submitted a ‘safety flash’ to the offshore industry organisation, International Marine Contractors Association. The safety flash summarised key safety matters and the accident, allowing wider dissemination of lessons learned to the industry. The company also undertook to review the adverse weather working guideline and the cargo loading procedure to include cargo lashing arrangements, and risk assessment for securing cargo.

Subsequently, DOF Management advised the ATSB that the existing working in adverse weather procedure had been reviewed and updated. The procedure now includes weather limits related to specific wind speeds, sea state, tidal streams, visibility and vessel movement. The precautions to be taken are directly related to each of the trigger points. In addition, working parameters for vessels with open sterns include a trigger point in the event that water is shipped on deck over the stern. The master is to be informed, who is then required to have a risk assessment conducted.

Current status of the safety issue

Issue status: Adequately addressed

Justification: The revised procedures for working in adverse weather and cargo handling, and the additional safety action taken has adequately addressed the safety issue.

Cargo securing procedures

Safety issue description:

Skandi Pacific’s safety management system (SMS) procedures for cargo securing were inadequate. There was no guidance for methods of securing cargo in adverse weather conditions.

Proactive safety action taken by DOF Management

Action number: MO-2015-005-NSA-002

Following the accident, DOF management submitted a ‘safety flash’ to the offshore industry organisation, International Marine Contractors Association. The safety flash summarised key safety matters and the accident, allowing wider dissemination of lessons learned to the industry. The company also undertook to review and update the cargo securing procedure and risk assessment to include the requirement for a lashing plan and load sequence planning with rig prior to entry of 500 m zone.

Subsequently, DOF Management advised the ATSB that its existing cargo loading procedure and risk assessment had been revised to include further clarification surrounding cargo lashing and hazards of shifting cargo. The revised procedure contains trigger points for assessing further operation when there is:

• water on deck over the stern
• an unexpected deterioration in weather
• an unfavourable change of the vessel position/heading.

Additionally, when offloading and backloading cargo alongside offshore installations, the risk assessment for securing cargo is to be reviewed and updated to ensure all hazards of shifting cargo are identified and addressed.

Further, a simultaneous operations (SIMOPS) procedure for combined rig and vessel cargo loading operations has been developed and implemented. The SIMOPS procedure is in addition to the procedures for working in adverse weather and cargo loading.

Current status of the safety issue

Issue status: Adequately addressed

Justification: The revised procedures and risk assessments for cargo handling and securing, and the additional safety action taken has adequately addressed the safety issue.

Open stern offshore support vessels

Safety issue description:

Skandi Pacific’s managers had not adequately assessed the risks associated with working on the aft deck of vessels with open sterns, including consideration of engineering controls to minimise water being shipped on the aft deck.

Response to safety issue by DOF Management

Action number: MO-2015-005-NSA-004

Following the accident, DOF management submitted a ‘safety flash’ to the offshore industry organisation, International Marine Contractors Association. The safety flash summarised key safety matters and the accident, allowing wider dissemination of lessons learned to the industry. The company also undertook to conduct of a risk assessment for anchor handling tug supply (AHTS) vessels conducting cargo operations. The risk assessment was to include considering engineering controls to minimise excessive water on the vessel back deck (for example, stern door or other vessel type).

Subsequently, DOF Management advised the ATSB that the existing risk assessment for offloading deck cargo at installation had been reviewed and updated to include a section regarding risks associated with securing cargo. The company indicated that it had assessed and mitigated risks associated with vessels with open stern.

ATSB comment/action in response

The ATSB acknowledges the proactive safety taken by DOF Management. However, the ATSB considers further action is necessary to adequately address the risks associated with the use of vessels with open sterns, including the engineering controls to minimise shipping seas on the aft deck proposed in the company’s response. Therefore, the ATSB has issued the following recommendation and safety advisory notice.

ATSB safety recommendation to DOF Management

Action number: MO-2015-005-SR-006

Action status: Released

The Australian Transport Safety Bureau recommends that DOF Management take further action to adequately address the safety issue concerning the use of vessels with open sterns.

Current status of the safety issue

Issue status: Partially addressed

Justification: Further action by DOF Management to adequately address the safety issue, including the company’s proposal to consider engineering controls to minimise shipping seas on the aft deck of open stern vessels, is considered necessary.

ATSB safety advisory notice to masters, owners and operators of offshore support vessels

Action number: MO-2015-005-SAN-005

Action status: Closed

The Australian Transport Safety Bureau advises the masters, owners and operators of all offshore support vessels to ensure that the risks associated with working on the aft deck of vessels with open sterns are adequately assessed.

Additional safety action

Whether or not the ATSB identifies safety issues in the course of an investigation, relevant organisations may proactively initiate safety action in order to reduce their safety risk. The ATSB has been advised of the following proactive safety action in response to this occurrence.

DOF management advised the ATSB that a marine risk awareness presentation titled SAFE the RITE WAY and a safety handbook have been developed and implemented. A 'lessons learned' communication plan from this accident are included in the SAFE the RIGHT WAY induction and presentation. Presentations have been delivered both externally to industry partners and forums and internally via mandatory DOF induction training.

Sources and submissions

Sources of information

The sources of information during the investigation included:

• Skandi Pacific’s master and directly involved crewmembers
• DOF Management
• Geoscience Australia
• National Offshore Petroleum Safety and Environmental Management Authority (NOPSEMA)
• Australian Maritime Safety Authority (AMSA).

References

International Maritime Organisation, 2015, The International Convention for the Safety of Life at Sea (SOLAS) 1974, as amended, IMO, London.

International Maritime Organisation Resolution A.489(XII), Adopted on 19 November 1981, Safe Stowage And Securing Of Cargo Units And Other Entities In Ships Other Than Cellular Container Ships Cargo Securing Manual.

International Maritime Organisation Resolution A.863(20), Code of Safe Practice for the Carriage of Cargoes and Persons by Offshore Support Vessels (OSV Code), as Amended.

Guidelines for Offshore Marine Operations (G-OMO), Guidelines for Offshore Marine Operations Revision: 0611-1401, 2013. Visit www.g-omo.info and follow the link on the G-OMO tab to Guidelines for Offshore Marine Operations.

National Offshore Petroleum Titles Administrator (NOPTA), Offshore Petroleum and Greenhouse Gas Storage Act 2006. Visit www.nopta.gov.au and follow the link on the Legislation, determinations, guidelines & fact sheets tab to offshore petroleum acts.

Ritchie, G 2008, Offshore Support Vessels, A Practical Guide, The Nautical Institute, UK. Visit www.nautinst.org and follow the link on the Publications to the publications list.

Submissions

Under Part 4, Division 2 (Investigation Reports), Section 26 of the Transport Safety Investigation Act 2003 (the Act), the Australian Transport Safety Bureau (ATSB) may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. Section 26 (1) (a) of the Act allows a person receiving a draft report to make submissions to the ATSB about the draft report.

A draft of this report was provided to the Australian Maritime Safety Authority, The Bahamas Maritime Authority, DOF Management, National Offshore Petroleum Safety and Environmental Management Authority and the master, chief mate, second mates and integrated ratings on board Skandi Pacific and the integrated rating’s next of kin.

Submissions were received from the Australian Maritime Safety Authority, The Bahamas Maritime Authority, DOF Management, National Offshore Petroleum Safety and Environmental Management Authority and the master on board Skandi Pacific and the integrated rating’s next of kin. The submissions were reviewed and where considered appropriate, the text of the report was amended accordingly.

Purpose of safety investigations The objective of a safety investigation is to enhance transport safety. This is done through: identifying safety issues and facilitating safety action to address those issues providing information about occurrences and their associated safety factors to facilitate learning within the transport industry. It is not a function of the ATSB to apportion blame or provide a means for determining liability. At the same time, an investigation report must include factual material of sufficient weight to support the analysis and findings. At all times the ATSB endeavours to balance the use of material that could imply adverse comment with the need to properly explain what happened, and why, in a fair and unbiased manner. The ATSB does not investigate for the purpose of taking administrative, regulatory or criminal action. Terminology An explanation of terminology used in ATSB investigation reports is available here. This includes terms such as occurrence, contributing factor, other factor that increased risk, and safety issue. Publishing information  Released in accordance with section 25 of the Transport Safety Investigation Act 2003 Published by: Australian Transport Safety Bureau © Commonwealth of Australia 2016 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this publication is licensed under a Creative Commons Attribution 3.0 Australia licence. Creative Commons Attribution 3.0 Australia Licence is a standard form licence agreement that allows you to copy, distribute, transmit and adapt this publication provided that you attribute the work. The ATSB’s preference is that you attribute this publication (and any material sourced from it) using the following wording: Source: Australian Transport Safety Bureau Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

The occurrence

A limited-scope, fact-gathering investigation into this occurrence was conducted in order to produce this short summary report and allow for greater industry awareness of potential safety issues and possible safety actions.

What happened

In the early hours of 10 July 2015, Hyundai Dangjin (cover) was in the final stages of loading its cargo of iron ore at Port Walcott, Western Australia. The ship was starboard side alongside the wharf and the chief mate and draught surveyor were on the wharf to check the ship’s draught. They could see the forward and aft draught marks but not the midships marks.

At 0450,^[1] the chief mate asked the second mate, via UHF radio, to read the midships draught on the ship’s port (outboard) side. The ship’s crew had already rigged a rope ladder (Figure 1) adjacent to the draught marks there.

Figure 1: Rope ladder, as rigged at the time of the accident

Source: Australian Maritime Safety Authority

In preparation to climb down the rope ladder, the second mate donned a life vest (non-inflatable flotation aid). The able seaman (AB) on duty offered to go down the ladder instead of the second mate, who was a large and heavy man. The second mate declined the AB’s offer (mates are trained to read draught marks).

Just after 0455, the chief mate and draught surveyor returned from the wharf to the ship’s office. The chief mate then called the second mate and asked for the midships draught. The second mate did not reply.

At that time, the second mate was near the bottom of the ladder, about 7 m below the ship’s deck. He called out to the AB for help and said he was having difficulty. When the AB checked, he saw the second mate struggling to hold on to the ladder. As the AB looked around for a rope to throw down, the second mate fell into the water. The AB threw a nearby lifebuoy to the second mate and it landed a few metres away.

The second mate tried to swim to the lifebuoy, but was not able to reach it. The sea was rough (1.4 m sea on a 0.4 m swell) and the water temperature was about 22 °C.

Rescue attempts

At about 0458, the AB called the third mate on the radio and told him that the second mate had fallen into the water. The AB then climbed down the ladder and entered the water. At this time, the second mate was about 4 m from the ladder and drifting further away. The AB had difficulty breathing and swimming in the rough, cold seawater. He was unable to reach the second mate and returned to the ladder.

The third mate reported the man overboard to the chief mate, before hurrying to the rope ladder. When he arrived there, he saw the second mate about 20 m from the ladder. His arms were moving slowly and he was not getting any closer to the lifebuoy about 3 m away. The third mate then went aft to get a lifebuoy with a lifeline attached.

At about 0510, the chief mate informed the master that the second mate had fallen from the rope ladder. The master left his cabin and went to the ship’s office, where he ordered that the port accommodation ladder and pilot ladder be lowered to the water.

The AB was at the bottom of the rope ladder when an ordinary seaman (OS) arrived. The AB was already suffering from the effects of the cold water and was having trouble holding on to the ladder. The OS threw a rope to the AB and he tied it around his waist.

Meanwhile, the third mate returned with a lifebuoy with a lifeline and threw it towards the second mate. The lifebuoy landed close to him and he was able to get an arm through it. The third mate then started to pull him towards the ship’s side near the rope ladder.

The AB had remained at the bottom of the ladder, to assist the second mate when he was close enough. When the second mate was about 5 m from the ship’s side, crew at the scene saw that he was no longer holding onto the lifebuoy.

Raising the alarm

At about this time, the chief mate arrived on deck near the rope ladder. He saw the AB holding on to the bottom of the ladder, up to his waist in the water. The chief mate took the lifebuoy line from the third mate and instructed him to go to the bridge and raise the alarm.

At about 0512, the third mate rang the general alarm. He then announced over the ship’s public address system that there was a man in the water, and for the crew to go to their muster stations.

The AB could not assist the second mate who was drifting further away, so he climbed the ladder to the deck. He was suffering from the effects of the cold water and exhaustion and was escorted to the ship’s hospital by two other crewmembers.

As the second mate drifted further aft, the chief mate continued to call out to him and shine a torch in his face. He did not receive any response.

When the third mate returned to the main deck, the accommodation ladder had been lowered to the water level. He climbed down and was able to drag the second mate onto the ladder’s lower platform. He then commenced cardiopulmonary resuscitation (CPR) on the second mate while the accommodation ladder was being raised to deck level with both men on its lower platform.

By 0520, two members of the terminal’s emergency response team and the port’s emergency management officer had boarded Hyundai Dangjin. Shortly after, when the accommodation ladder was at deck level, they visually assessed the second mate and detected no signs of life.

The master was concerned due to the gap between the ship’s side and the accommodation ladder platform, and the second mate’s weight. In order to avoid a further incident, he instructed his crew to transfer the second mate to the deck using the stores crane and a suitable sling. The master then went to inform the ship’s managers of the accident.

At 0540, the AB, suffering from hypothermia symptoms, was taken ashore for assessment at a local hospital.

At about 0555, a St John Ambulance paramedic boarded the ship. By this time, the second mate had been moved to the deck, CPR was continued and the paramedic assessed him. At 0605, after finding no sign of life, the paramedic informed the master that the second mate had died.

ATSB comment

The rope ladder had been rigged upside down (Figure 1). With their wrong side up, the ladder steps (folded aluminium) did not provide a flat surface to stand on comfortably. Further, the steps were not good handholds.

The sole precaution taken by the second mate while reading the draught marks was his life vest. No fall prevention measures were put in place or used. The life vest’s specifications could not be determined but similar types provide around 7 to 10 kg of buoyancy.

While the AB was standing by on deck, man overboard response measures (such as a lifebuoy with light and line near the ladder) were not in place. Fortunately, his well-intentioned but impulsive descent of the ladder in an attempt to rescue the second mate did not result in another casualty.

The second mate’s initial postmortem examination report stated the cause of death as ‘undetermined (pending further investigation)’. However, the report noted that some findings of the examination ‘could be seen with drowning’. The report stated that the body was of a man of large build and included his height and weight. This is consistent with the attending police officer’s report, which noted that the second mate ‘was a man of large overweight build’.

Safety action

Whether or not the ATSB identifies safety issues in the course of an investigation, relevant organisations may proactively initiate safety action in order to reduce their safety risk. The ATSB has been advised of the following proactive safety action in response to this occurrence.

Toyo Sangyo

Toyo Sangyo Company (Toyo Sangyo), Hyundai Dangjin’s managers, have taken the following safety actions to avoid a similar accident.

Review of shipboard safety management system

Toyo Sangyo’s review of its shipboard safety management system (SMS) resulted in a ‘Safe Draft Check Instruction’ being included in the procedures related to cargo operations for bulk carriers. The instruction details the procedures (including permits to work) when checking the ship’s draught from a rope ladder.

The ship’s managers issued a circular to its managed fleet to raise awareness of the accident, lessons learned and inform crew of the changes to SMS procedures.

Rio Tinto Iron Ore

Review of draught survey methods

Rio Tinto Iron Ore, the Port Walcott terminal managers, reviewed the draught survey methods at its terminals. As a result of the review, the reading draught marks from rope ladders was prohibited. An alternate method, using a manometer, was put in place.

Rio Tinto Iron Ore has promulgated the revised draught survey requirements and methods through a circular to all of its terminals. The revised requirements and methods have also been provided to ship’s agents for inclusion in pre-arrival information for shipmasters.

Safety message

In many cases, little attention is paid to planning apparently straightforward tasks, such as using a rope ladder. This can lead to important factors and relevant considerations not being taken into account, including the experience and physical ability of persons undertaking the task.

Purpose of safety investigations The objective of a safety investigation is to enhance transport safety. This is done through: identifying safety issues and facilitating safety action to address those issues providing information about occurrences and their associated safety factors to facilitate learning within the transport industry. It is not a function of the ATSB to apportion blame or provide a means for determining liability. At the same time, an investigation report must include factual material of sufficient weight to support the analysis and findings. At all times the ATSB endeavours to balance the use of material that could imply adverse comment with the need to properly explain what happened, and why, in a fair and unbiased manner. The ATSB does not investigate for the purpose of taking administrative, regulatory or criminal action. Terminology An explanation of terminology used in ATSB investigation reports is available here. This includes terms such as occurrence, contributing factor, other factor that increased risk, and safety issue. Publishing information  Released in accordance with section 25 of the Transport Safety Investigation Act 2003 Published by: Australian Transport Safety Bureau © Commonwealth of Australia 2016 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this publication is licensed under a Creative Commons Attribution 3.0 Australia licence. Creative Commons Attribution 3.0 Australia Licence is a standard form licence agreement that allows you to copy, distribute, transmit and adapt this publication provided that you attribute the work. The ATSB’s preference is that you attribute this publication (and any material sourced from it) using the following wording: Source: Australian Transport Safety Bureau Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

__________

1. All times referred to in this report are local time (WST), Coordinated Universal Time (UTC) + 8 hours.

Safety summary

What happened

At about 1618 Eastern Standard Time on 8 July 2015, the pilot/owner of an amateur-built Pitts Model 12, registered VH‑JDZ, took off from Maitland Airport, New South Wales.

Witnesses reported hearing a loud engine noise at about 1630 that caught their attention. They then observed the aircraft at the top of what appeared to be a vertical climb. The aircraft slid backwards, tail first, before entering a horizontal spin. Shortly after, the witnesses lost sight of the aircraft below the tree line and some reported hearing a loud bang.

The aircraft was located by search aircraft and ground personnel who arrived at the site of the accident at about 1715. The aircraft had collided with terrain in thick bushland, fatally injuring the pilot. The aircraft was destroyed by impact forces and an intense post-impact fire.

What the ATSB found

Radar data and witness reports were consistent with the aircraft being used for aerobatic manoeuvres in the minutes prior to the accident.

The ATSB considered the results of the pilot’s post-mortem examination, which indicated the pilot had coronary artery disease that may have resulted in permanent incapacitation. However, while that remained a possibility, there was insufficient evidence to conclude that it influenced the development of the accident. The ATSB found that for reasons that could not be determined, VH‑JDZ entered a vertical manoeuvre from which the pilot did not regain control before colliding with terrain. Additionally, the aircraft was being flown at a height which reduced the time available to effect a recovery, if required.

The ATSB also identified instances of misinterpretation of a number of the regulations concerning the maintenance of amateur-built experimental aircraft. This has the potential to affect the safety of these aircraft and those on board.

Safety message

Aerobatic flying requires extensive training and ongoing commitment to maintain the skills necessary for safe operations. Unauthorised aerobatic manoeuvres increase the risk to the pilot, any passengers and third parties in the vicinity of the aerobatics.

Flying at low level reduces the safety margin available should something unexpected happen. The ATSB has issued a series of ‘Avoidable Accidents’ publications, the first of which details the risks involved in low-level flying and includes the following statement:

Low-level flying also presents fewer opportunities to recover from a loss of control compared to flight at higher altitudes. It takes time to react and to regain control of an aircraft, and the closer to the ground you are, the less time and distance you have. Flying at low altitudes is not only risky when things are going right; it becomes downright perilous when things are going wrong.

More information is available from the ATSB’s avoidable accident web page.

Finally, ongoing safety requires aircraft owners and maintainers to operate and maintain the aircraft in accordance with relevant regulations, including those specific to experimental aircraft. Aircraft operation and maintenance outside the regulatory requirements increases safety risk.

Amateur-built Pitts Model 12 registered VH-JDZ

Source: Supplied

The occurrence

At about 1618 Eastern Standard Time (EST)^[1] on 8 July 2015, the pilot/owner of an amateur-built Pitts Model 12, registered VH-JDZ (JDZ), departed Maitland Airport, New South Wales on a local flight. The aircraft was being operated in the experimental category and the pilot was the only person on board.

JDZ was fitted with a transponder^[2] however, the aircraft was not detected by secondary radar. The lack of a secondary surveillance radar return was consistent with either the transponder not being activated or, if activated, the equipment not operating correctly (see the section titled Radar and airspace information). However, primary surveillance radar detected an aircraft at about 1622, to the south of Maitland Airport. The location and movements of this aircraft were consistent with the area in which the pilot of JDZ was reported to regularly conduct aerobatics (see the following discussion).

The primary radar data indicated that the aircraft initially flew in a south-easterly direction, before heading west toward an area near Lovedale Road and Keinbah Quarry. The aircraft then tracked variously east and west, consistent with the conduct of aerobatic manoeuvres. The aircraft disappeared from radar before reappearing on two occasions during the flight. At about 1630 the aircraft disappeared from radar a third time and was not detected again.

Witnesses reported that the area in which they saw the aircraft leading up to the accident is regularly used for aerobatics. Several witnesses reported hearing a particularly loud engine sound at about 1630 that caught their attention. Witnesses described:

• the aircraft at the top of what appeared to be a vertical climb
• the aircraft sliding backwards, tail first, before entering a flat spin
• that the flat spin descent was slow and almost straight down, with little forward speed
• that the aircraft completed several rotations during the spin before sight was lost behind trees.

Three of the witnesses believed the aircraft had impacted terrain and contacted the police. At about 1650 emergency personnel arrived in the area and commenced a ground search. At the same time, a number of aircraft from Maitland Airport commenced an aerial search. A rescue helicopter was also dispatched.

The aircraft was located from the air at about 1712 and ground crews arrived at the site shortly after. The aircraft had collided with terrain in thick bushland in the area in which aircraft regularly conducted aerobatics and close to where the previously-discussed aircraft disappeared from primary surveillance radar. The pilot was fatally injured and there was an intense post‑impact fire. The fire could not be extinguished with the available resources and was monitored until it self‑extinguished (Figure 1).

Figure 1: Accident site viewed from the front of the aircraft and showing the surrounding thick vegetation and minimal structural damage to the wings and airframe (consistent with the reported vertical descent)

Source: ATSB

__________

1. Eastern Standard Time (EST) was Coordinated Universal Time (UTC) + 10 hours.
2. A radio device that, when triggered by the correct radio signal (called interrogation), transmits a pre-coded reply. Air traffic control ground equipment interrogates the aircraft’s transponder, identifies the aircraft by its reply code and displays the aircraft’s position on the controller’s radar screen.

Context

Pilot information

The pilot commenced flying in December 2006 and was issued with a Private Pilot (Aeroplane) Licence in January 2008. In addition, the pilot was endorsed for single-engine aeroplanes not exceeding 5,700 kg maximum take‑off weight and category design features including retractable undercarriage and manual propeller pitch control.

The pilot had accrued about 750 hours flying experience, mostly in Piper PA32 and Eagle Aircraft 150-type aircraft. The pilot completed an emergency manoeuvre training course in a Pitts S2A in June 2010 and commenced aerobatic training in a similar aircraft in August 2012. In February 2014, the pilot was issued an endorsement to fly ‘tail wheel undercarriage aircraft’ and completed a flight review at the same time.

The pilot commenced dual familiarisation flying in VH-JDZ (JDZ) in December 2014. Basic aerobatic training was also conducted in JDZ from March 2015. At the time of the accident the pilot had accrued about 22 hours in JDZ, of which about 7 hours was solo. The pilot’s aerobatic flying experience, under instruction and in a Pitts S2A, a Yakovlev Yak 52 and JDZ, totalled about 21.5 hours.

The pilot had not yet achieved the required competency to be authorised for spinning or an aerobatic endorsement. In addition, the pilot was advised by his instructors to not conduct solo aerobatics.

The pilot held a current Class 2 Medical Certificate that was issued on 22 January 2015.

Aircraft information

General

JDZ was an amateur-built Pitts Model 12 two-seat tandem biplane, serial number 42. It was powered by a Vedeneyev M14P radial engine and had a two-blade composite constant speed propeller. The only special requirements to operate the Pitts Model 12 were an endorsement for tail wheel undercarriage and for manual propeller pitch control.

The pilot purchased JDZ from the builder on 11 November 2014 and became the registered operator on 11 December 2014.

Construction and certification

JDZ was purchased as a kit from the aircraft manufacturer in the United States and assembly commenced in Australia in 1999. The aircraft was issued with a Special Certificate of Airworthiness (SCOA) on 24 October 2006. An annex to the SCOA outlined a number of Phase I operational limitations and specific maintenance requirements. This included the requirement for at least 25 hours of flight testing within a 35 NM (65 km) radius of Wedderburn Airport, New South Wales. Unusually, the SCOA was not limited by an expiry date.

An entry in the aircraft logbook on 12 November 2014 stated that the required 25 flight test hours were complete. Manoeuvres recorded as tested included 'loops, rolls, spins, half‑cuban eights, roll off the top of a loop, wingover and snaps [snap rolls]’. The SCOA flight test area was amended on 12 December 2014 to include Maitland Airport.

On 12 June 2015, SCOA Phase II replaced Phase I, authorising operation of an amateur-built aircraft in accordance with the Phase II limitations. These limitations restricted the conduct of aerobatic manoeuvres to those that had been flight tested.

Airworthiness and maintenance

The aircraft logbook statement required JDZ to be maintained under Civil Aviation Regulation 1988 (CAR) Schedule 5 and in accordance with the additional requirements of the annex to the SCOA. An annual inspection was to be carried out every 100 hours or 12 months, whichever came first. The last annual inspection was completed on 16 March 2015, and the associated maintenance release was valid at the time of the accident.

Uncertified maintenance

Maintenance carried out on an aircraft is required to be certified complete in accordance with CAR 42ZE and CAR Schedule 6. However, there were multiple maintenance actions carried out on JDZ that were not certified complete in the aircraft logbook. These included:

• a weld repair to the tail wheel post structure, including fabric replacement and rudder system removal and re‑fitment
• replacement of the original wooden propeller blades with composite blades^[3]
• replacement of engine oil and fuel delivery ‘flop tubes’^[4]
• replacement of the fuel quantity indicating system
• replacement of the tail wheel
• adjustment of the flying wires during the aircraft ‘flight test phase’.

Pilot maintenance

CASA authorised the holder of a private pilot licence to conduct certain maintenance as specified in CAR Schedule 8 (Schedule 8). In addition, CASA Instrument 33/13 Authorisation of persons to carry out maintenance on certain amateur‑built, kit-built and light sport aircraft with a special certificate of airworthiness detailed that, in some circumstances, a person who builds or has previously built an amateur‑built aircraft of a similar type, may be authorised to conduct certain maintenance and issue a maintenance release for that aircraft.

All maintenance conducted by a pilot under CASA Instrument 33/13 and/or Schedule 8 was required to be certified as per CAR 42ZE.

The occurrence pilot was authorised to conduct maintenance on JDZ as specified in Schedule 8. However, the pilot was not authorised to conduct maintenance on JDZ under Instrument 33/13 as they did not build the aircraft.

It was reported that the pilot replaced the tail wheel of JDZ due to excessive wear and shimmy. This maintenance action was permitted if carried out in accordance with Schedule 8. In addition, it was reported that adjustment of the aircraft flying wires, which was not included in Schedule 8, was carried out by another pilot during the Phase I flight testing period. Neither the tail wheel replacement nor the flying wire adjustments were certified in the aircraft logbook.

Maintenance of amateur-built aircraft

This and other ATSB investigations have identified a degree of uncertainty amongst a number of pilots and engineers regarding the certification of maintenance performed on amateur-built experimental aircraft. When asked, CASA advised that, in accordance with CASA Instrument 33/13, a licenced aircraft maintenance engineer (LAME) is authorised to certify for maintenance of certain amateur‑built, kit‑built and light sport aircraft with a SCOA. This authorisation extends to the scope and privileges of the LAME’s licence. Additionally, a LAME may certify for maintenance ‘for and on behalf of’ a maintenance organisation, if the aircraft undergoing maintenance is within the scope of activities for that organisation.

CASA also advised that other than a number of specific exemptions, amateur-built aircraft are to be maintained in accordance with the regulations and any additional instructions detailed in a SCOA annex. Ultimately, as per CAR 42CB Experimental aircraft, the certificate of registration holder is responsible for ensuring the aircraft is maintained in accordance with the regulations. In addition, CAR 133 Conditions of flight requires the pilot in command to ensure that all the required maintenance has been completed and is appropriately certified.

Weight and balance

An initial weight and balance was carried out on JDZ on 19 October 2006.

The ATSB estimated the aircraft’s weight and balance for the accident flight based on the available data. This indicated that the aircraft was likely operating within limitations at the time and the ATSB concluded that weight and balance was unlikely to be a factor in the loss of control.

Meteorological information

Witnesses described the weather at the time they observed JDZ as being calm with a clear sky.

Weather observations were recorded by an automatic station at the Maitland Visitor’s Centre. Observations for 1500 on 8 July 2015 recorded a temperature of about 15 °C, relative humidity of 60 per cent and a southerly wind at 6 km/h (about 3 kt). This was consistent with the witness observations.

According to the Geoscience Australia website (see www.ga.gov.au/), at about 1630 the sun was at an elevation of 5° above the horizon and an azimuth of 300°.^[5] The aircraft that was observed on primary radar, and the ATSB concluded to have been JDZ, was flying in an easterly direction just prior to disappearing from radar at about 1631 (see the section titled Radar and airspace information). The ATSB concluded that it was unlikely sun glare was a factor in the loss of control.

Communications

A review of the recorded Maitland common traffic advisory frequency^[6] broadcast data determined that the pilot transmitted a departure call on that frequency at 1609 local time. No further transmissions from JDZ were recorded.

Radar and airspace information

Aircraft transponder equipment

The aircraft logbook indicated that during the assembly, the aircraft was fitted with a Mode C transponder that could be activated by the pilot. There was no subsequent logbook entry in respect of the transponder.

Aeronautical Information Publication Australia^[7] EN ROUTE (ENR) 1.1 Section 56 OPERATING REQUIREMENTS FOR TRANSPONDERS, paragraph 56.1 stated that:

Pilots of aircraft fitted with a serviceable Mode 3A or Mode S transponder must activate the transponder at all times during flight, and if the Mode 3A transponder is Mode C capable, that mode must also be operated continuously.

In addition to the potential benefits to air traffic control of the activation of an aircraft’s transponder, aircraft traffic collision avoidance systems^[8] rely on transponder information for their traffic alerting and collision avoidance functions.^[9] These systems have the potential to enhance crews’ self separation, in particular in uncontrolled airspace.

Analysis of recorded radar data

The primary radar data for the aircraft that was manoeuvring in the area in which the pilot of JDZ was reported to regularly conduct aerobatics showed:

• an unidentified aircraft immediately to the south of Maitland Airport at about 1622
• from immediately south of the airport the aircraft tracked to the vicinity of the Keinbah Quarry
• the aircraft then tracked variously east and west for about 4 minutes
• the primary radar return for the aircraft was lost on two occasions and was reacquired after 23 and 10 seconds respectively
• the primary radar return for the aircraft was lost a third time at about 1631 and was not detected again.

Expert analysis of the aircraft’s radar data as it tracked variously east and west for 4 minutes indicated that the aircraft was manoeuvring consistent with the conduct of aerobatic manoeuvres.

In addition, the recorded flight path of a search aircraft with an operational transponder was also reviewed. The radar data from this aircraft showed that the:

• unidentified aircraft’s last detected location was consistent with the accident site
• primary surveillance radar could detect aircraft in that area down to an altitude of about 200 ft.

The ATSB concluded that, based on the analysis of the primary radar data, and its consistency with the witness observations, the unidentified aircraft was highly likely to have been JDZ. The ATSB could not determine whether the lack of a secondary surveillance radar return for JDZ was a result of the transponder not being activated or the equipment not operating correctly. Of note, secondary surveillance radar returns were observed for other aircraft in the area, indicating that terrain shielding was not a factor.

Airspace

• The Visual Terminal Chart^[10] for the area showed that JDZ was operating in uncontrolled, Class G airspace. This airspace, which extended up to an altitude of 8,500 ft, provided ample altitude in which to conduct aerobatic manoeuvres.

Site and wreckage information

Witness information

Witnesses reported that a loud engine noise caught their attention and caused them to look up to see JDZ near the top of what appeared to be a vertical climb. The aircraft was then observed to slide backwards tail first, before entering a flat spin. The flat spin descent was described as slow. Sight of the aircraft was lost behind trees and, shortly after, an explosion was heard by some witnesses.

Table 1 provides a summary of the observations that were unique to each witness. Figure 2 shows the location of each witness with respect to the accident site. Of the witnesses, witness D had flying experience.

Table 1: Summary of specific witness observations

Figure 2: Witness location with reference to the accident site

Source: Google earth, modified by the ATSB

The area was reported to be popular with aerobatic pilots. The witnesses advised that they regularly observed various aerobatic manoeuvres by these aircraft.^[11] It was reported that the manoeuvres by the pilot of JDZ were, up to the point where sight was lost, consistent with the regularly-observed aerobatic manoeuvres.

The height of JDZ at the top of the vertical climb was estimated from the witness observations. While it could be expected that the accuracy of this estimation was affected by the witness’ limited visual references, observations from the three closest witnesses to the accident site suggested that the height of the aircraft at the top of the vertical climb was less than 2,000 ft above ground level.

Wreckage information

The accident site was about 8 km south-west of Maitland Airport and was located in thick bushland near the Hunter Expressway. Examination of the site and wreckage determined that immediately before colliding with terrain, the aircraft was inverted, slightly nose-down and in an almost vertical descent. An initial fuel-fed post-impact fire was followed by an intense oil-fed fire, which in combination destroyed most of the aircraft.

The aircraft structure was severely damaged by impact forces and the post-impact fire, limiting the evidence available and conclusions that could be drawn from an examination of the wreckage. However, that examination determined that:

• all components of the aircraft were located at the impact site
• there were no pre-accident defects with the aircraft and engine
• the cockpit canopy frame was consistent with it being in the closed position at the time of impact
• damage to the aircraft’s tubular structure was consistent with impact with trees and/or terrain
• what remained of the wooden wing spars showed little damage, consistent with little forward speed just prior to the impact with terrain
• the engine was partly submerged in the soft earth, at an angle of about 23° to the horizon
• continuity of engine controls was confirmed; however, the rest of the engine and its accessories were extensively damaged, or destroyed by the fire, preventing further assessment
• one propeller blade was torn from the hub and was wedged between the engine and the ground. The other blade was undamaged and had stopped in a vertical position. The positions of the propeller blades indicated that the engine rotated no more than a quarter turn following the impact with terrain.

Flight controls

The flight control system and cables were examined as far as possible. With the exception of the elevator control tube and hinge bolt, continuity of the flight controls was established on-site. A number of elevator and tail structure components were recovered for subsequent technical examination at the ATSB’s technical facilities in Canberra, Australian Capital Territory.

Technical examination of the recovered components determined that the fractures of the control tube and hinge bolt were consistent with overstress. No pre‑existing defects were identified and the failures were likely the result of impact with trees and/or terrain.

Seat harness and buckles

The webbing of both seat harnesses was almost entirely destroyed in the post-impact fire. Despite this level of damage, various buckles and harness adjust mechanisms were identified. Of these, one five-point harness buckle and one lap-belt harness buckle were found in the secured, or closed position. The other buckles were found unsecured (Figure 3). Refer to the section Test and research for more detail.

Figure 3: As-found five-point and lap-belt harness buckles. Of these four buckle sets, one of each type was secured (see the lower left and right images below)

Source: ATSB

Medical and pathological information

Post-mortem examination

The post-mortem examination identified that the pilot was fatally injured as a result of multiple injuries, consistent with an aircraft accident, and was deceased at the time of the post‑impact fire. The examination also identified significant coronary artery disease. Toxicology results were negative for alcohol and commonly-tested drugs.

Medical information

CASA records showed that the pilot had been monitored for borderline high blood pressure (hypertension) by the same Designated Aviation Medical Examiner (DAME) since January 2007. In this regard, the pilot had been taking anti-hypertensive medication as prescribed by his general practitioner (GP) since August 2012.This medication was permissible in accordance with the International Civil Aviation Organization Manual of Civil Aviation Medicine and was reportedly effective in maintaining the pilot’s blood pressure within prescribed limits.

The DAME and the GP advised that the pilot had not reported any side effects to the medication.

It was reported that the pilot also regularly took various herbal tonics that were dispensed by a naturopath. Additionally, the pilot was reported to use over-the-counter herbal supplements for relief from allergies, sinusitis and colds. The pilot did not report the use of these herbal supplements to their GP or DAME.

Civil Aviation Safety Regulations 1998 (CASR) 67.155 stated the criteria for medical standards for the holder of a private pilot licence. In particular, item 2.3 of the regulation required a pilot to not use:

…any over-the-counter or prescribed medication or drug (including medication or a drug used to treat a disease or medical disorder) that causes the person to experience any side effects likely to affect the person to an extent that is safety-relevant

As part of the initial issue and/or renewal of a medical certificate, any prescribed or over‑the‑counter medications taken for greater than 2 weeks (including herbal or alternative therapies) are to be documented. A review of the pilot’s CASA medical file showed that the anti‑hypertensive medication was the only medication noted.

The pilot was reported to be active and healthy. The DAME and GP advised they had no indication to suspect the pilot had significant coronary artery disease, as was identified by the post‑mortem examination.

It was reported that the pilot had been suffering from a head cold for at least a week prior to the accident. However, the severity of the symptoms could not be determined. In addition, the pilot had been heavily involved in the Hunter Valley Air Show over the period 4–5 July 2015. Reportedly this included preparations over several weeks, activities over the weekend and assisting with the post-event clean-up.

The pilot was reported to be well rested and in good spirits in the 2 days prior to the accident. There was insufficient evidence to indicate fatigue was a factor in the loss of control.

Nausea during aerobatic flying

It was reported that the pilot was affected by nausea during certain flying activities. Low g Load (g)^[12] tolerance and associated airsickness resulted in the pilot’s aerobatic training flights being kept brief or cut short on a number of occasions and with a number of instructors. Certain manoeuvres, such as spinning and negative g, reportedly affected the pilot more significantly than others. In addition, on more than one occasion during an aerobatic training flight, the nausea overwhelmed the pilot to the point where the instructor had to take control of the aircraft.

Civil Aviation Advisory Publication (CAAP) 155-1(0) Aerobatics provided detailed information on many aspects of aerobatics, including the physiological effects of g forces. Section 4.1 stated:

Aerobatic manoeuvres involve rapid changes in speed and direction which impose significant accelerative forces on the aircraft and pilot. The physiological effects of these G forces can range from minor discomfort to loss of consciousness.

Pilots beginning aerobatics may be adversely affected by airsickness, disorientation and discomfort but continued practice, and the use of appropriate methods of mitigating the physiological effects, will allow most pilots to adapt fairly quickly to standard aerobatic manoeuvres.

The CAAP also provided further and more-detailed information on the various g forces and their effects on the body. It described that pilots can build up their tolerance to g forces with practice but need to be aware that established tolerance levels can be significantly reduced by various factors that affect their physical condition. These factors included, but were not limited to, fatigue, illness, medication, low blood pressure and dehydration. In addition, lack of recent aerobatic practice will also reduce a pilot’s g tolerance and pilots returning to aerobatics after some time need to check and then gradually re-establish their tolerance level.

Section 4.20 Disorientation, included the following statement:

Sustained rapid rotation, such is in flick manoeuvres or spins, can also lead to disorientation because visual reference is made difficult due to the rapid rotation and there is no visual correction to the confused signals from the balance mechanism.

Recent ear, nose, throat infections may cause injury or pain, as well as disorientation, during aerobatics.

The pilot’s last recorded aerobatic training flight was 28 May 2015. Training notes for that flight indicated the pilot was unable to recover from a stable inverted spin. The pilot’s reported disorientation and nausea that day led to the decision to end the flight and return to the airfield.

Survival aspects

The pilot was not required to and did not notify a SARTIME,^[13] or leave a Flight Note with a responsible person. In addition, consistent with there being no need for the carriage in the aircraft of an ELT,^[14] the aircraft was not equipped with a permanent emergency locator transmitter (ELT). It was reported the pilot did carry a portable ELT but, only on extended cross‑country flights.

The pilot’s decision to not nominate a SARTIME or leave a Flight Note and not carry an ELT increased the risk of a delayed emergency response in the event of an accident or incident. However, in this case the accident was not considered survivable due to the magnitude of the impact forces and intensity of the post-impact fire.

Operational information

Low-level flying

The pilot in command must not fly the aircraft over non-built-up or –populous areas at a height lower than 500 ft unless it is within the exemptions detailed in CAR 157 (4) Low flying. Radar data showed JDZ below 200 ft on two occasions during the flight that day. There was no evidence to indicate any requirement, or authorisation for going below 500 ft during the flight.

Flying endorsements

The CASR Dictionary, Part 1 Definitions defined aerobatic manoeuvres as those that involve:

• bank angles that are greater than 60˚; or
• pitch angles that are greater than 45˚, or are otherwise abnormal to the aircraft type; or
• abrupt changes of speed, direction, angle of bank or angle of pitch.

A pilot was required to hold the appropriate aeroplane category and a spinning endorsement before an aerobatic endorsement was issued. In addition, the pilot was required to have received training and shown competency in all the course units mentioned in Part 61 Manual of Standards.^[15] That training included performance criteria covering loops, rolls, stall turns, recovery from unusual attitudes and spins. The knowledge requirements included, but were not limited to an understanding of the definitions of negative and positive g and the associated effects on the pilot and the aircraft.

CASR 61.065 prohibited the conduct of any activity for which the licence holder was not authorised. In addition, CASR subpart 61.S stated the requirements for aerobatic endorsements. These included that:

• an initial aerobatic endorsement would authorise the pilot to conduct aerobatic manoeuvres in an aeroplane above 3,000 ft above ground level (AGL)
• subsequent endorsements were necessary for aerobatic activities at lower altitudes.

Aerobatic flying

It was reported by a number of Pitts Model 12 pilots and instructors that:

• loops in a Pitts Model 12 aircraft usually required 1,000 to 1,500 ft to complete
• loops with a vertical consistency of around 1,500 ft would be required by the student to achieve aerobatic training competency
• aerobatic manoeuvres during training were normally commenced at 5,000 ft
• in all cases the loop should be entered and exited at the same altitude.

The height required to recover from a spin varies with pilot competency and experience. Spin recovery within the number of turns normally required for the aircraft type is essential to achieve basic competency. Early identification of the spin and correct control inputs should result in a prompt recovery.

The Pitts Model 12 pilot operating handbook Aerobatic Flight included, in part, that the pilot should:

…

-Do be certain that you have ample altitude for the maneuvers [sic] that you want to perform.

-Do be sure that you are familiar enough with the maneuvers [sic], and the airplane, that a bad recovery will produce no worse result than embarrassment.

-Know the limitation on your PITTS Model 12 and yourself.

In addition, the pilot’s operating handbook identified the procedures for emergency spin recovery as:

- POWER OFF

- REMOVE YOUR HAND FROM THE STICK

- DETERMINE DIRECTION OF SPIN ROTATION

- APPLY FULL OPPOSITE RUDDER

- WHEN ROTATION STOP [sic], RECOVER

Finally, CAAP 155-1(0) Aerobatics provided pilots with:

• information and guidance on safety issues related to aerobatic flight, including in respect of the aircraft, pilot and regulations
• an explanation of spin recovery techniques
• advice on the importance of ensuring sufficient height to recover from an aerobatic manoeuvre by 3,000 ft AGL (or the lower limit of the pilot’s approval). In particular, section 7.3.2 of the CAAP stated:

It is highly probable that the consequence of an error or failure during low-level aerobatics will be fatal to the participants.

Test and research

Unsecured seat harness

CAR 155 (5) Aerobatic manoeuvres required an unoccupied seat harness to be secured by the pilot in command prior to conducting an aerobatic manoeuvre. In addition, CAAP 155‑1 (0) Aerobatics discussed the importance of securing unoccupied seat harnesses to prevent them fouling the controls. The ATSB was advised that this procedure had been demonstrated to the pilot and that the pilot was aware of the associated regulatory requirement.

The ATSB determined that the pilot was secured in the rear seat harness and that the unsecured seat buckles were associated with the forward, unoccupied seat (see the previous discussion titled Site and wreckage information). Testing in a Pitts Model 12 aircraft determined that an unsecured forward seat harness had the potential to interfere with the aircraft’s flight controls. Figure 4 shows the orientation of the flight controls with reference to an unsecured forward seat harness.

Refer to appendix A Unsecured seat harness for more information.

Figure 4: Unsecured forward seat harness, showing the potential for the harness to interfere with the flight controls, in particular during aerobatic manoeuvres

Source: ATSB

ATSB research

ATSB research report AR-2007-043(2) Amateur-built aircraft, published 26 March 2013 and available at www.atsb.gov.au concluded the following in relation to amateur-built aircraft accidents:

• Between 1988 and 2010, amateur-built aircraft on the Australian VH-register had an accident rate three times higher than comparable VH-registered factory-built aircraft conducting similar flight operations.
• The fatal and serious injury accident rate was more than five-times higher in amateur-built aircraft than in similar factory-built aircraft.
• Loss of aircraft control led to 25 per cent of all amateur-built accidents, slightly more than for factory-built aircraft accidents. However, the loss of control accident rate was over four times higher. As compared to factory-built aircraft, serious injury was three times more likely after loss of control in amateur-built aircraft accidents.
• Loss of control accidents were more likely to arise from aircraft handling issues where pilots had comparatively lower levels of experience on the aircraft type.
• Loss of control was more likely to occur in the initial climb phase of flight.
• Amateur-built aircraft pilots were significantly more experienced overall than factory-built aircraft accident pilots. However, they were significantly less experienced on the amateur‑aircraft type being flown at the time of the accident when compared to pilots of factory-built aircraft accidents and amateur-built aircraft owners in general. Twenty per cent of amateur-built aircraft accident pilots had less than 10 hours experience on the accident aircraft type.

ATSB research report AR-2008-045 Improving the odds: Trends in fatal and non-fatal accidents in private flying operations, published in June 2010 and also available at www.atsb.gov.au identified that 44 per cent of all accidents and over half of fatal accidents between 1999 and 2008 were attributed to private operations. These figures far surpassed the proportions for any other flying category, even though private operations contributed to less than 15 per cent of the hours flown in that decade. The report also identified that:

• the three most common occurrence types in fatal accidents were collision with terrain, loss of control and wirestrike
• aircraft handling was a significant contributor to loss of control and collision with terrain fatal accidents
• compared with non-fatal accidents, fatal accidents were more likely to be associated with violations of rules and regulations. Violations of rules and regulations remove safety defences and, when coupled with an error, increase the likelihood of an accident.

Previous occurrences

ATSB investigations

The ATSB has investigated a number of fatal accidents involving the conduct of aerobatics. All are available via the ATSB website at www.atsb.gov.au.

199501051 – Collision with terrain involving Pitts S-2A, registered VH-IXY

On 8 April 1995 the pilot was observed practising aerobatic manoeuvres in a Pitts S-2A, registered VH-IXY. The early afternoon flight was observed from the ground by a number of other pilots, including a highly-experienced pilot and flying instructor. This pilot/instructor was in radio contact with the pilot.

The instructor reported a number of unusual manoeuvres by the pilot of VH-IXY and unsuccessful attempts by the instructor to contact the pilot. The aircraft then appeared to stabilise and the pilot responded saying he thought he might have blacked out. The aircraft landed normally and the pilot and instructor discussed g-induced loss of consciousness.

During a later aerobatic flight by the pilot of VH-IXY, after a period of standard aerobatic manoeuvres, the instructor on the ground observed the aircraft pitch up before commencing a continuous roll to the left. The instructor made several unsuccessful attempts to contact the pilot by radio. The nose of the aircraft dropped and the aircraft dived almost vertically into the ground.

The pilot was fatally injured and the aircraft destroyed.

AO-2014-114 - Collision with terrain involving DHC-1 Chipmunk, registered VH-UPD

The ATSB also investigated the collision with terrain involving a DHC-1 Chipmunk, registered VH‑UPD, which occurred near Coffs Harbour, New South Wales on 29 June 2014.The passenger and witness reports indicated that, during an attempted aerobatic manoeuvre, the aircraft entered a spin. The pilot had reportedly received some aerobatic training but had yet to receive an endorsement. Video footage taken by witnesses showed the aircraft established in a slow, upright spin. The on‑site evidence was consistent with the spin continuing until the impact with terrain. The ATSB found it was likely the pilot did not possess the necessary skills and judgement to conduct the manoeuvre safely and consistently.

AO-2014-163 - Collision with terrain involving amateur-built aircraft, registered VH-EGT

On 10 October 2014, the pilot of an amateur-built One Design DR-107 aircraft, registered VH‑EGT, was observed performing a series of low-level aerobatic manoeuvres. Each involved a vertical climb and tumbling manoeuvre followed by a vertical dive and a low altitude recovery. Witnesses reported that, during recovery from the last vertical dive, the aircraft collided with terrain. The aircraft was destroyed by the impact and the pilot was fatally injured.

There was insufficient evidence to determine why the recovery was not accomplished above the pilot’s minimum‑authorised aerobatics height. The accident highlighted the risks inherent in performing low-level aerobatics and the ATSB encouraged pilots to always maintain minimum approved heights above the ground when performing aerobatics.

United States National Transport Safety Bureau investigations

A review of the United States National Transportation Safety Board (NTSB) Aviation Accident Database identified two accidents involving Pitts Model 12 aircraft losses of control and one regarding possible pilot impairment that showed similar traits to the ATSB’s investigation of the accident involving JDZ. Below is a brief summary of the NTSB’s investigation reports into these accidents (available at NTSB website).

Loss of control accidents

NTSB investigation report MIA00LA149 involved a collision with terrain of a Pitts Model 12 on 9 May 2000, with both occupants fatally injured. It was reported the pilot had been flying in formation, but discontinued the return flight to practice aerobatics. Examination of the wreckage showed extensive damage consistent with the aircraft impacting the hard ground inverted, in about a 20°–25°nose‑low attitude. The NTSB determined the ‘probable cause’ of the accident as ‘the pilot’s loss of control in flight for undetermined reasons’.

NTSB investigation report ANC05LA033 found that a Pitts Model 12 collided with terrain while the pilot was performing low‑altitude aerobatic manoeuvres on 12 February 2005. Site and wreckage examination concluded that all damage was consistent with the aircraft impacting the ground in a near-vertical descent. The NTSB determined the ‘probable cause’ of the accident as:

The pilot’s failure to maintain control of the airplane while performing a low altitude, aerobatic maneuver [sic], which resulted in an uncontrolled descent, and an in-flight collision with terrain. A factor associated with the accident was the initiation of a low altitude aerobatic manoeuvre.

Pilot impairment

NTSB investigation report CEN11LA582 found that ‘the pilot’s impairment during an aerobatic airshow performance for reasons that could not be determined…resulted in an in-flight loss of airplane control’. The report identified the pilot had previously sought treatment for vertigo and nausea suffered while conducting aerobatics.

United Kingdom Air Accidents Investigation Branch investigation

The United Kingdom Air Accidents Investigation Branch^[16] published investigation report EW/G2009/05/11 regarding a collision with terrain involving a Pitts S-12 that occurred on 14 May 2009. The pilot was the sole occupant and received serious injuries. The report ‘emphasised the importance of entering a manoeuvre at a height from which recovery was possible in the event of failure to complete it as planned’. Additionally the report referred to the summary of British Aerobatic Association Safety Sense Leaflet 19 Aerobatics, which stated:

…start with sufficient height to give plenty of margin if things go wrong

__________

3. Replacement of the blades was entered in the aircraft logbook, but the entry did not include the associated release and traceability documentation required by CAR Schedule 6. In addition, the registration details and logbook statement had yet to be updated with the new propeller information.
4. A flop tube is a flexible hose with a weighted end. The weight ensures the hose draw point remains submerged in the respective fluid whether the aircraft is in normal or inverted flight.
5. The clockwise horizontal component of the sun’s or moon’s position from true north, measured in degrees.
6. Common Traffic Advisory Frequency is the frequency on which pilots operating at a non-towered aerodrome should make positional radio broadcasts.
7. A package of documents that provides the operational information necessary for the safe and efficient conduct of national (civil) and international air navigation throughout Australia and its Territories.
8. An aircraft collision avoidance system that monitors the airspace around an aircraft for other aircraft equipped with a corresponding active transponder and gives warning of possible collision risks.
9. AIP ENR 1.6 Section 7.1 Operation of SSR Transponders, paragraph 7.1.2.
10. Visual Terminal Charts provide aeronautical and topographical information for operations under the Visual Flight Rules in the vicinity of major aerodromes. They also show controlled airspace.
11. Witness simulations of their observations with an aircraft model were consistent, even when unsure of the correct manoeuvre terminology.
12. g Load is the nominal value for acceleration. In flight, g load values represent the combined effects of flight manoeuvring loads and turbulence. This can be a positive or negative value.
13. The time nominated by a pilot for the initiation of search and rescue action if a report has not been received by the nominated unit.
14. Crash-activated radio beacon that transmits an emergency signal that may include the position of a crashed aircraft. Also able to be manually activated.
15. The purpose of Part 61 Manual of Standards Instrument 2014 is to set out the standards relating to flight crew licensing.
16. AAIB reports can be viewed via https://www.gov.uk/government/organisations/air-accidents-investigation-branch.

Safety analysis

Background

The private flight in a Pitts Model 12, registered VH-JDZ (JDZ), was conducted in favourable weather conditions but ended in a collision with terrain, fatally injuring the pilot. The witness observations were consistent with the aircraft exiting a vertical manoeuvre in an uncontrolled state, with insufficient height for recovery before impacting terrain. Wreckage and accident site examination indicated that the aircraft collided with terrain inverted, slightly nose-down and with little or no forward speed.

Given the extent of the impact damage and intensity and effect of the post-impact fire, there was no evidence that any mechanical failure or aircraft unserviceability contributed to the development of the accident. In addition, the surrounding area contained cleared and open paddocks, which could have been used for a precautionary or emergency landing had that been required.

This analysis will consider the circumstances that preceded the loss of control.

Unauthorised flight manoeuvres

The pilot was appropriately licenced to fly JDZ solo and had been encouraged to fly the aircraft solo and with an instructor to increase familiarity and gain experience on type. The pilot did not hold the required authorisation to conduct aerobatic manoeuvres or fly at low altitude. The pilot’s instructors advised the pilot to not conduct solo aerobatic manoeuvres. However, radar surveillance data and witness reports were consistent with the pilot conducting aerobatic manoeuvres in the minutes preceding the impact with terrain. This included flight below 200 ft above ground level.

The pilot’s decision to conduct the aerobatic manoeuvres solo, and without authorisation, increased the risk of unintended departure from controlled flight. In addition, the pilot’s conduct of those manoeuvres at low altitude reduced the height, and therefore the time available in which to recover control of the aircraft should it depart from controlled flight.

The observed vertical manoeuvre, followed by the reported horizontal flat spin is consistent with a loss of control. The reason for the loss of control could not be determined.

Flight control fouling

The forward seat harness buckles were found unsecured during the wreckage examination. The post-impact fire hindered the ATSB’s ability to determine if the harness had been secured in a non-standard manner prior to the flight, increasing the risk that they might unbuckle. In addition, although unlikely, the possibility that if buckled correctly the buckles may have come undone as a consequence of the impact sequence could not be discounted. Despite these possibilities, the ATSB concluded that the as-found insecurity of the front seat harness buckles suggested that they were not secured correctly as part of the standard pre-flight preparation for solo flight.

Unsecured harnesses have the potential to interfere with the flight controls, in particular during aerobatic flight. The cockpit layout would have inhibited the pilot’s ability to clear such an interference in flight. Any interference could result in an unintended manoeuvre or inhibit the pilot’s ability to recover from a manoeuvre, whether the manoeuvre was intended or unintended.

While flight control fouling remains a possibility, there was insufficient evidence to determine if it was a contributor to the accident.

Pilot incapacitation and non-reporting of medication

Pilot incapacitation can impair a pilot’s performance to the extent that safe operation of the aircraft is adversely affected. It can be due to the effects of a medical condition or a physiological impairment and represents a potential threat to flight safety. It can have either a long- or short‑term effect on the pilot.

The pilot’s post-mortem examination indicated that the pilot had coronary artery disease that could have resulted in a permanent incapacitation event. In addition, reports of other flights indicated the pilot suffered from temporary incapacitation as a result of nausea caused by the onset of g forces. Further, the use of non-prescribed medical or herbal supplements or the effects of a cold or virus can increase the risk of pilot incapacitation. The conduct of aerobatic manoeuvres can exacerbate the effects of a number of these conditions and non‑prescribed ‘treatments’.

All of these factors were present during this occurrence and pilot incapacitation could have preceded the loss of control or influenced any recovery. However, while an incapacitating event remained a possibility, there was insufficient evidence to conclude probable contribution to this accident.

Many documents highlight the risks involved with various medications and their possible effects on pilot performance and flight safety. The Designated Aviation Medical Examiner reported being unaware of all medications and supplements reported taken by the pilot. Therefore, the medical examiner was unable to determine any possible interactions between the prescribed anti‑hypertensives and/or the non‑prescribed medication and herbal supplements. This removed a potential protection against the risk of pilot incapacitation as a result of those medicines and supplements.

Maintenance

The ATSB found that uncertified maintenance was conducted on JDZ by authorised and unauthorised persons. The registered operator is responsible for the airworthiness and maintenance control of the aircraft to ensure its safe operation. Additionally, the pilot in command must not commence a flight unless all required maintenance has been completed and certified.

This and other ATSB investigations have reiterated the importance of adhering to the regulations as they apply to aircraft maintenance and operation. However, there is evidence that some in the industry are unsure of, or are misinterpreting a number of the regulations concerning the operation and maintenance of amateur-built experimental aircraft. Authorised aircraft maintenance is mandated to assure a level of safety for aircraft operations. Unauthorised maintenance increases the risk of mechanical failure, in turn reducing the level of safety and increasing the risk of injury or death.

Findings

From the evidence available, the following findings are made with respect to the collision with terrain involving a Pitts Model 12, registered VH-JDZ, which occurred about 8 km south-west of Maitland Airport, New South Wales on 8 July 2015. These findings should not be read as apportioning blame or liability to any particular organisation or individual.

Contributing factors

• Control of the aircraft was lost during a vertical manoeuvre that was likely entered at an altitude lower than that required and the aircraft was not recovered before impact with terrain.

Other factors that increased risk

• The pilot conducted low-level flight and aerobatic manoeuvres despite not holding the appropriate authorisations, significantly increasing the risk of an accident.
• The pilot used non-prescribed medication and herbal supplements without informing their doctor and Designated Aviation Medical Examiner, removing a protection against pilot incapacitation.
• Uncertified and unauthorised maintenance was carried out on VH-JDZ, which increased the risk that a technical issue would affect the safety of the aircraft and those on board.
• Industry interpretation of the regulations regarding the maintenance of amateur-built experimental aircraft was varied, increasing risk associated with the maintenance of those aircraft.

Other findings

• The front seat buckles were found unsecured and, although unable to be determined if they were not secured correctly as part of the standard pre-flight preparation for solo flight, or they may have undone as a consequence of the impact sequence, any insecurity during aerobatic manoeuvres has implications for flight control fouling.

Sources and submissions

Sources of information

The sources of information during the investigation included the:

• Civil Aviation Safety Authority
• New South Wales Police and Coroner
• Department of Defence
• Airservices Australia
• Bureau of Meteorology
• a number of flight instructors and Pitts Model 12 pilots
• aircraft kit manufacturer and builder
• maintainers of VH-JDZ.

Submissions

Under Part 4, Division 2 (Investigation Reports), Section 26 of the Transport Safety Investigation Act 2003 (the Act), the ATSB may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. Section 26 (1) (a) of the Act allows a person receiving a draft report to make submissions to the ATSB about the draft report.

A draft of this report was provided to the Civil Aviation Safety Authority, the United States National Transportation Safety Board, the aircraft kit manufacturer, a number of flight instructors, the Department of Defence and Airservices Australia.

No submissions were received from those parties.

Appendices

Appendix A – Unsecured seat harness

Examination of seat harness components

The aircraft was fitted with two seats to accommodate a passenger forward and the pilot aft. Both seats had a five‑point harness and a lap-belt harness. The dual harness arrangement is typical for aerobatic aircraft. The five-point and lap-belt harnesses are secured to separate airframe attachment points. In addition, the buckles of each harness type operate in opposite directions to minimise the risk of accidental release. It was reported that VH-JDZ had the same type of harness in the forward and rear seats.

The various secured and unsecured harness buckles located on site were retained for further examination.

Forward five-point harness

The lever mechanism of the unsecured forward five-point buckle was observed to be in the closed position; however, it could not be determined if this lever was in the closed position during flight or as a result of the impact with terrain. It was noted, however, that all five buckle components were separate from each other, consistent with the buckle assembly not being secured or closed. Visual examination of the unsecured five-point buckle components did not identify any deformation or damage. In addition, when tested, the buckle mechanism was capable of operation and security.

Rear harness

The rear pilot shoulder harness restraint cable that connected it to the airframe structure had fractured in overload. The tubular structure to which the rear pilot crotch strap was secured had fractured at each end of the tube, where it is welded to the airframe structure. In addition, components of the ‘secured’ rear five-point harness buckle showed signs of deformation (Figure A1).

Figure A1: Deformation to rear seat buckle in comparison with the forward buckle, and fracture of rear seat crotch strap hard point

Source: ATSB

Conclusion

Damage to the airframe and the severed harness cable at the rear pilot position was consistent with the pilot being secured in the seat at the point of impact. Therefore, the open buckles of the five point harness and lap belt were associated with the unoccupied, forward passenger seat.

Unsecured harness testing

Civil Aviation Regulation 155 (5) Aerobatic manoeuvres required an unoccupied seat harness to be secured by the pilot in command prior to conducting an aerobatic manoeuvre. In addition, Civil Aviation Advisory Publication 155‑1 (0) Aerobatics indicates that unoccupied seats should have their harnesses secured to prevent them fouling the controls.

In the Pitts Model 12 the harnesses can be secured by connecting the buckles, pulling the adjustment straps tight, wrapping the adjusting lengths round the bundle and tucking in the loose ends. The ATSB was advised that this procedure was demonstrated to the pilot and that the pilot was aware of the need to secure the seat harnesses.

The ATSB conducted testing in a Pitts Model 12 aircraft to ascertain if unsecured forward seat harness components had the potential to interfere with the aircraft’s flight controls.

The rear pilot’s rudder pedals are positioned on either side of the forward seat base. A loose lap strap and associated buckle, from either the forward seat lap belt or five-point harness, were found to have the potential to interfere with rudder pedal operation. In addition, a buckle component could lodge between the rudder pedal and the diagonal tubular bracing of the airframe (Figure A2).

Figure A2: Possible rudder pedal interference from an unsecured forward seat harness

Source: ATSB

Access to the rudder pedals from the rear seat was limited and it would be very difficult to release a misplaced front seatbelt by hand. Also, due to the design of the rudder control system, rearward movement of one rudder pedal does not naturally impart a forward movement of the opposite pedal.^[17] Therefore, the cockpit layout and rudder design may inhibit the pilot’s ability to free an obstructed rudder pedal in flight.

It was also noted that it was possible for the forward seat crotch strap and associated buckle to interfere with the forward seat elevator control stick (Figure A3). It would not be possible for a rear‑seat pilot to remove this obstruction by hand during flight operations.

Figure A3: Possible elevator control interference from an unsecured forward seat harness

Source: ATSB

__________

17. Also known as an ‘open loop control’.

Purpose of safety investigations The objective of a safety investigation is to enhance transport safety. This is done through: identifying safety issues and facilitating safety action to address those issues providing information about occurrences and their associated safety factors to facilitate learning within the transport industry. It is not a function of the ATSB to apportion blame or provide a means for determining liability. At the same time, an investigation report must include factual material of sufficient weight to support the analysis and findings. At all times the ATSB endeavours to balance the use of material that could imply adverse comment with the need to properly explain what happened, and why, in a fair and unbiased manner. The ATSB does not investigate for the purpose of taking administrative, regulatory or criminal action. Terminology An explanation of terminology used in ATSB investigation reports is available here. This includes terms such as occurrence, contributing factor, other factor that increased risk, and safety issue. Publishing information  Released in accordance with section 25 of the Transport Safety Investigation Act 2003 Published by: Australian Transport Safety Bureau © Commonwealth of Australia 2016 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this publication is licensed under a Creative Commons Attribution 3.0 Australia licence. Creative Commons Attribution 3.0 Australia Licence is a standard form licence agreement that allows you to copy, distribute, transmit and adapt this publication provided that you attribute the work. The ATSB’s preference is that you attribute this publication (and any material sourced from it) using the following wording: Source: Australian Transport Safety Bureau Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

The aircraft departed Newcastle Airfield (EINC), County Wicklow, Ireland, with two pilots on board. The intention was to perform a short flight before returning to Newcastle Airfield. Approximately 20 minutes into the flight, a small amount of blue smoke was noticed in the cockpit, followed by a loss of engine oil pressure. The pilot decided to carry out a forced landing and a suitable field was selected. To prevent engine damage, the pilot shut down the engine during the approach to the chosen field. The aircraft landed and came to rest without further incident. There were no injuries and the aircraft was undamaged. Subsequent examination revealed that a hose supplying engine oil to the engine oil cooler had disconnected in flight, resulting in the loss of all engine oil. The aircraft was fitted with a Jabiru 2200A engine.

As the accident occurred in Ireland, the Air Accident Investigation Unit (AAIU) Ireland is responsible for investigating this occurrence. As part of its investigation, the AAIU requested assistance from the Australian Transport Safety Bureau (ATSB) as the State of Manufacture of the engine. In accordance with chapter 5.18 of Annex 13 to the Convention on International Civil Aviation, the ATSB appointed an accredited representative to assist the AAIU and initiated an investigation (AE-2015-051) under the Australian Transport Safety Investigation Act 2003.

The AAIU investigation has been completed and the final investigation report is available from the AAIU website at www.aaiu.ie.

The report contains two recommendations for Jabiru Aircraft Pty Ltd:

1. Jabiru Aircraft Pty Ltd should include a warning in the Instruction and Maintenance Manuals for the Jabiru 2200 engine and similar Jabiru engine types, highlighting that if oil cooler hoses are installed with oil on the hose fittings or, with oil on the inside of the hose in the vicinity of the fittings, hose security may be adversely affected.
2. Jabiru Aircraft Pty Ltd should, with the aid of a suitable publication, promulgate de-identified details of the subject event to users of its engines.

In accordance with Chapter 6.12 of Annex 13 to the Convention on International Civil Aviation, Jabiru Aircraft Pty Ltd have advised the ATSB that they have communicated the following actions to the AAIU:

Jabiru Aircraft Pty Ltd released Service Letter (JSL018) on 1 October 2015 as an informational letter on the subject of oil cooler hoses applicable to all Jabiru engines. Additionally, the following text has been added to Installation Manuals JEM3302-6 and JEM2202-8, and Maintenance Manual JEM0002-6:

‘Note: If oil cooler hoses are installed with oil on the hose fittings or, with oil on the inside of the hose in the vicinity of the fittings, hose security may be adversely affected.’

Arrivals radar was being operated by a trainee controller under the supervision of a rated controller. VH-ANJ and VH-FKI were both tracking to Perth via Clackline. The trainee controller had given both heading and altitude instructions to the aircraft in an attempt to manage their separation. A breakdown in separation standards occurred when the aircraft approached to within 3 nautical miles and 500 feet vertically before the instructions took effect. The supervising controller took over and provided VH-FKI with traffic information on VH-ANJ which allowed visual separation to be maintained.

Although the trainee controller recognised that there would be a separation problem the instructions given to the aircraft were not given in time nor were they sufficient to separate the aircraft under all possible conditions.

The supervising controller was slow to take control of the situation because he wished to give the trainee, who was at an advanced stage of training, the opportunity of recognising the situation and making the necessary corrections.

The trainee's task was made more difficult when VH-FKI did not descend as early as anticipated and drifted to the right of the runway 24 localiser, and towards VH-ANJ, after the pilot had reported established.

The following factors led to the development of this occurrence.

1. The trainee controller did not have sufficient expertise to be able to recognise and allow for all possible variables in the traffic conflict situation.

2. The supervising controller made an error of judgement when he delayed his intervention and was unable to prevent the breakdown in separation standards.

3. Whilst not a direct factor, the flight path of VH-FKI complicated the situation.