Executive summary

What happened

On the afternoon of 3 October 2022, a Pilatus Britten-Norman Islander BN2A-21, registered VH-WQA and operated by Torres Strait Air, was conducting a non-scheduled passenger air transport flight from Saibai Island Airport, Queensland (QLD) to Horn Island Airport, QLD. There was 1 pilot and 6 passengers on board.

About 19 km NE of Moa Island both engines began to surge. The pilot diverted towards Kubin Airport on Moa Island. As the aircraft passed to the south of the township of Saint Pauls, the pilot determined there was insufficient altitude remaining to reach the airport. As a result, the pilot conducted a forced landing on a road 7 km ENE of Kubin Airport. There were no reported injuries to the pilot or the passengers. The aircraft was substantially damaged.

What the ATSB found

The ATSB found that the dual engine speed fluctuations and associated power loss was probably the result of fuel starvation.

The mechanism was not conclusively determined, however it was identified that the pilot did not operate the aircraft's fuel system in accordance with the aircraft flight manual, and that the configuration and location of the aircraft’s fuel controls and tank quantity gauges were probably not conducive to rapid and accurate interpretation. The aircraft manufacturer released a service letter in June 2022 that detailed an optional modification to centralise the fuel system controls and gauges, however this modification was not fitted to VH-WQA. The ATSB considered that these factors increased the risk of inadvertent fuel tank selection.

Safety message

Accidents involving fuel mismanagement are an ongoing aviation safety concern. Pilots are reminded of the importance of understanding an aircraft’s fuel supply system and being familiar and proficient in its use. Adhering to procedures, maintaining an accurate fuel record, and ensuring appropriate tank selections are made for the phase of flight will lessen the likelihood of fuel starvation.

The investigation

The occurrence

On the afternoon of 3 October 2022, a Pilatus Britten-Norman Islander BN2A-21 (Islander), registered VH‑WQA and operated by Torres Strait Air, was conducting a non-scheduled passenger air transport flight from Saibai Island Airport, Queensland (QLD) to Horn Island Airport, QLD. There was 1 pilot and 6 passengers on board.

At about 1333, when the aircraft was at a cruise altitude of 6,000 ft and approximately 19 km NE of Moa Island (Figure 1), the pilot recalled that the right engine began to surge. The pilot observed the right engine speed fluctuate, accompanied by yawing[1] of the aircraft.

Figure 1: VH-WQA Torres Strait flightpath

Source: Google Earth annotated by the ATSB

About 40–50 seconds later, the left engine also began to surge. The pilot recalled disconnecting the autopilot, selecting auxiliary fuel pumps on and placing the left and right engine mixture, pitch and power levers in the full forward position. The pilot then varied the throttle and mixture levers in an attempt to resolve the issue, but the surging continued. The pilot reported that they did not check the fuel contents indicators at the time, as they believed the main tanks were supplying fuel to the engines and there was sufficient fuel within those tanks to complete the flight.

As a result of the engine issues, the pilot was unable to maintain level flight and recalled descending at 800–1,100 ft per minute to maintain an airspeed of about 75–85 knots. When the engine issues commenced, the closest land to the aircraft was Moa Island. However, the pilot initially elected to divert to Badu Island, as there was cloud along the track[2] to Moa Island. The pilot commenced a right turn towards Badu Island and recalled advising Brisbane Centre air traffic control of the diversion.

A short time later, the pilot revised their diversion plans and set a course for Kubin Airport on Moa Island. As the aircraft passed to the south of the township of Saint Pauls, the pilot determined there was insufficient altitude remaining to reach Kubin Airport (Figure 2). The pilot considered the available options and chose to land on a road that ran east–west. At an altitude of about 600–700 ft, the pilot recalled switching the fuel supply from main tanks to wing tip tanks but noticed no improvement in engine performance.

Figure 2: VH-WQA Moa Island flightpath

Source: Google Earth annotated by the ATSB

About 15 seconds prior to touchdown, the pilot recalled the right engine stopped. They then shut down the left engine and recalled setting both mixtures to cut-off, closing the throttles, and selecting the main fuel cocks to off. The pilot positioned the aircraft to land on the road, before noticing a power line in the intended touchdown area, which they manoeuvred to avoid. The pilot continued to fly the aircraft above the road and as the road traversed a hill and made a left turn, the pilot followed with the aircraft. At this point the pilot recalled that the stall warning system[3] sounded and that they called for the passengers to ‘brace’.

At about 1339, the aircraft impacted the ground heavily, with the rear fuselage and tail breaking away from the aircraft. The aircraft came to rest in an area of vegetation on the northern side of the road. The pilot reported instructing the passengers to evacuate, which they did through the emergency window hatches. The pilot then exited through the pilot door.

There were no reported injuries to the pilot or the passengers. The aircraft sustained substantial damage.

Context

Pilot information

The pilot held a commercial pilot licence (aeroplane), issued in January 2018 with single and multi-engine aircraft class ratings, multi-engine instrument rating, and a valid Class 1 medical.   The pilot had accrued a total flight time of about 2,400 flying hours, including about 200–250 hours flying the Islander, and commenced employment with Torres Strait Air in April 2022.

The pilot reported they were well rested and the ATSB found no risk indicators of the pilot experiencing a level of fatigue known to affect performance.

Aircraft information

VH-WQA was a Pilatus Britten-Norman Islander BN2A-21; a twin-engine, high-wing, unpressurised aircraft, with fixed landing gear and seating for up to 9 passengers. It was fitted with Textron Lycoming IO-540-K1B5 piston engines and Hartzell HC-C2YK-2CUF propellers. The aircraft was manufactured in 1975 and first registered in Australia on 5 October 1978. Torres Strait Air became the registered operator of the aircraft on 14 August 2018.

The last periodic inspection was carried out on 21 September 2022. There were no defects recorded on the aircraft’s maintenance release. At the time of the accident, the aircraft had accumulated 14,081.92 hours total time in service, and the left and right engines had accumulated 294.15 and 2,185.36 hours since overhaul respectively.

Fuel system

The fuel system in VH-WQA was specification Mod NB/M/364 that consisted of a tip tank and main fuel tank within each wing. The usable capacity of each main tank was about 65 US gallons (246.1 l), and 27.5 US gallons (104.1 l) for each wing tip tank.

Fuel was fed from either the main or wing tip tanks (but not simultaneously) to the auxiliary fuel pumps, and then supplied to either engine via the main fuel cocks (Figure 3).

Figure 3: Fuel system

Source: Britten-Norman

The electrically actuated tank cocks were controlled via an overhead panel, located above the windscreen central pillar. The panel consisted of port and starboard switches that selected either the wing tip tank or main tank. There was a corresponding indicator light for each selection. The panel also contained a switch that controlled the brightness of the tip tank indicator light when that tank was selected (dim or bright). The same switch also determined the main tank indicator light when that tank was selected, either on or off (for the tip tank light positions of dim or bright, respectively) (Figure 4, Label A).

Figure 4: VH-WQA cockpit fuel system configuration

Source: Queensland Police annotated by the ATSB

The main tank contents indicators were located above the fuel tank section panel, with corresponding auxiliary fuel pump switches located either side (Label B). The wing tip tank contents indicators were located on the right side of the cockpit above the side window (Label C). The main fuel cock rotary selectors were located above the main tank contents indicators (Label D).

Fuel system placards,[4] highlighting the configuration and usage of the fuel tanks, were located between the main fuel contents indicators and to the left of the wing tip tank contents indicators (Figure 5). The placards indicated that wing tip tanks were to be used last (after main tanks), and that main tanks were not to be used for landings when the gauge reads less than 3 gallons.

Figure 5: VH-WQA fuel system placards

Source: Queensland Police annotated by the ATSB

Service letter SL 145

On 1 June 2022, Britten-Norman released service letter (SL) 145 applicable to all BN2 series Islander aircraft with Mod NB/M/364 wing tip tanks. The service letter detailed an optional modification to provide an alternative pilot interface that centralised the fuel system controls, specifically relocating the wing tip tank indicators to the overhead panel adjacent to the main tank fuel indicators, with the fuel selection switches mounted between the indicators. VH-WQA did not have this modification fitted.

Aircraft flight manual

The aircraft flight manual[5] for VH-WQA contained a supplement that detailed procedures, limitations, and information for the operation of the modified fuel system. This supplement stated that 13.5 US gallons (51 l) of fuel was to be retained in each wing tip tank at all times for structural reasons, except that this fuel could be used as reserve for holding or diversion to an alternate airfield. The wing tip tanks were required to be refuelled before the main tanks and used in flight after the main tanks were exhausted. The supplement stated:

Fuel

The flight time records indicated that VH-WQA flew 6 sectors on the day of the occurrence, prior to the accident flight. The occurrence pilot flew all of those sectors (Table 1). 

Table 1: VH-WQA flight time records 3 October 2023

 Source: Torres Strait Air

The aircraft was refuelled with avgas[7] at Horn Island Airport at 0626 (300.3 l) and again at 1150 (200.3 l). The pilot recalled using a fuel dipstick[8] to check the contents of the 2 main tanks and 2 wing tip tanks following refuelling, as well as checking drained fuel for water contamination. The flight time records contained a fuel discrepancy following the 1150 refuelling (the aircraft had an additional 20 l) and recorded tank contents on the Yorke Island – Horn Island sector (which is 19 l more than the recorded fuel remaining). Despite these discrepancies, the aircraft departed Horn Island Airport with a recorded 350 l of fuel on board, which was sufficient for the round trip to Saibai Island Airport. This was composed of 120 l in each of the main tanks and 55 l in each of the wing tip tanks.

The aircraft was recorded to have used 83 l on the flight to Sabai Island Airport, but the pilot did not record the individual levels of each tank in the flight time records on arrival at Saibai Island Airport. The ATSB calculated that 81 l of fuel was required to complete the intended flight from Saibai Island Airport to Horn Island Airport. About 44 l fuel would likely have been used for the flight from Saibai Island Airport to the vicinity where the pilot recalled that the engine surging commenced.

Pilot fuel management

The pilot was familiar with VH-WQA’s modified fuel system and during interview, recalled the correct use of the tip tanks. Despite this, flight time records indicated that the pilot had, on occasion, operated VH-WQA without the required 51 l of fuel in the wing tip tanks and had also used these tanks when fuel remained in the main tanks. The pilot did not give a reason for this usage of the tip tanks.

Site and wreckage

The ATSB did not examine the accident site or wreckage. The site was attended on 6 October 2022 by representatives from the aircraft operator, insurance provider, independent maintenance provider, and Queensland Police. The site inspection was recorded by Queensland Police and the video footage was provided to the ATSB along with forensic photographs taken on the day of the accident. The aircraft wreckage was not guarded during the period between the accident occurring and the on-site examination.

The wreckage was located a short distance from the road in an area of low foliage approximately 6.5 km ENE of Kubin Airport. The undercarriage and empennage were separated from the aircraft and the fuselage was resting right wing low (Figure 6 and Figure 7). No obvious airframe or engine defects were reported by those that attended the site.

Figure 6: Accident site from road

Source: Queensland Police

Figure 7: Accident site

Source: Queensland Police

The right main tank and wing tip tank were ruptured during the accident sequence. The left main fuel tank and wing tip tank were reported as being intact and found to contain approximately 60 l, and less than 500 ml of fuel respectively. No fuel contamination was apparent, and the fuel selectors were found to work correctly when electrical power was applied to the aircraft. Table 1 depicts the positions of the fuel controls as documented in forensic photographs taken shortly after the accident. The wing tip tanks were selected for use, the auxiliary pumps were on, and the main fuel cocks were selected ‘off’, which was consistent with the pilot’s recollection of the occurrence.

Table 2: Fuel system cockpit control positions

Source: Queensland Police

Related occurrences

Collision with terrain, Pilatus Britten–Norman aircraft BN2B-27 near Marcel Marchant Aerodrome, Chile on 16 April 2019, 1895SP (Directorate General of Civil Aviation Chile)

On 16 April 2019, a Pilatus Britten–Norman aircraft BN2B-27 with modified fuel system (Mod NB/M/364) collided with terrain shortly after take-off from Marcel Marchant Aerodrome, Chile. The pilot and 5 passengers on board were fatally injured and the aircraft was destroyed. The final investigation report concluded the near-empty wing tip tanks were likely selected to supply fuel to the engines, resulting in fuel starvation.

Collision with terrain, Pilatus Britten–Norman aircraft BN2B-21 near Devil’s Hole, approximately 2.5 nm north of Jersey Airport, Channel Islands on 3 November 2013, Air Accidents Investigation Branch United Kingdome Bulletin: 10/2014

During a search and rescue flight at night in poor weather conditions, one engine ceased producing power and eventually stopped. During the subsequent diversion towards Jersey Airport the other engine also stopped. The pilot was able to reach the Jersey coast and make a forced landing, in which the aircraft suffered significant damage. The aircraft had operated a previous flight with the fuel system configured so that tip tank fuel was being supplied to the engines. The aircraft departed on the accident flight in the same configuration and the engines stopped when the tip tank fuel became exhausted. The investigation also noted the layout of the fuel controls and tank quantity gauges as a source of complication with regard to the presentation of fuel source information to the pilot.

Australian occurrences

The ATSB has conducted a number of investigations that involved fuel management and fuel starvation. Examples include:

• Fuel starvation involving Cessna 310R, VH-JQK, Sunshine Coast Airport, QLD, on 18 August 2022 (AO-2022-040)
• Fuel starvation and forced landing involving Piper PA-28, VH-BDB, 15 km WSW of Bankstown Airport, NSW, on 19 September 2017 (AO-2017-094)
• Fuel starvation involving Cessna 206, 3.5 NM NE of Aldinga, SA, on 3 February 2019 (AB-2019-004)
• Cessna C310R, VH-HCP, 3km E Newman Aerodrome, WA, on 26 January 2001 (200100348)

The 2013 ATSB publication, Avoidable Accidents No. 5: Starved and exhausted: Fuel management aviation accidents (AR-2011-112) focused on accidents involving fuel starvation due to fuel management, stating:

Safety analysis

At a cruise altitude of 6,000 ft the pilot reported that the right engine and, shortly after, the left engine began to surge. There were limited potential reasons for two fuel-injected engines to behave in this manner, and (almost) simultaneous dual technical failure of independent systems would be highly unlikely. The most probable contributing factor was a fuel-related issue, which was also consistent with the onset of surging described by the pilot.

The aircraft departed Horn Island Airport with sufficient total fuel for the round trip to Saibai Island Airport and a significant quantity of fuel remained in the left main tank after the accident. The aircraft therefore did not suffer fuel exhaustion. The aircraft was filled with the correct fuel type and   no fuel contamination was apparent to those on site. Additionally, the aircraft had already flown for over an hour since refuelling, which suggested that an issue with fuel quality was unlikely.

The reported fuel burn from Horn Island to Saibai Island, and the ATSB’s calculated fuel usage for the return flight, to the point of engine surging, totalled 127 l. Each of the main tanks were recorded as commencing the round trip with 120 l. This meant that the (approximately) 60 l of fuel found on site in the left main tank was correct if the flights to that point had been flown entirely on the main tanks, and assuming the breached right main tank contained the same volume. This also supported the pilot’s recollection of the occurrence, along with the fuel system cockpit control positions.

However, the calculated fuel usage was also close to the reported amount of fuel in the wing tip tanks (55 l each, 110 l total) upon refuelling at Horn Island. It was therefore considered whether the wing tip tanks had been selected for most or all of the flight duration and became exhausted. Previous occurrences have highlighted this as a possibility and it was also consistent with the small quantity of fuel found remaining within the reportedly unbreached left wing tip tank (<500 ml).

The pilot reported that they did not verify the fuel remaining in the aircraft's main or wing tip tanks when attempting to restore engine performance, following the onset of the engine surging. This oversight was possibly due to confirmation bias, as the pilot believed there was sufficient fuel in all 4 tanks at that time. The fuel controls and tank contents indicators were also probably not conducive to rapid and accurate interpretation due to the potentially confusing configuration of the fuel system panels, and the disparate location of the wing tip tank contents indicators on the right side of the cockpit. In addition, fuel records from earlier flights showed that the pilot was using the wing tip tanks when fuel remained in the aircraft's main tanks, which was not in accordance with the approved flight manual. The ATSB considered that both of these factors increased the likelihood of the wing tip tanks being inadvertently selected during part or all of the round trip, leading to the exhaustion of the wing tip tanks. However, it is noted that this scenario is inconsistent with the pilot’s recollection of switching to the tip tanks in attempting to restore power – the position to which the tanks were found selected during the wreckage examination.

The ATSB was unable to account for the fuel tank content discrepancy for either scenario. However, the aircraft wreckage was not guarded during the period between the accident occurring on 3 October 2022 and the on-site examination on 6 October 2022. Consequently, while there was not evidence of tampering, it is possible that during this time fuel was removed (syphoned or pumped) from either tank or leaked from the fuel system, through a mechanism not obvious during the on-site examination. Ultimately the fuel discrepancies were unable to be resolved from the evidence available. However, on the balance of probabilities and in the absence of any other likely mechanical or fuel related issue, the dual engine speed fluctuations and associated power loss was most probably the result of fuel starvation.

Findings

From the evidence available, the following findings are made with respect to the Engine power loss and forced landing involving Pilatus Britten-Norman Islander BN-2A, registered VH-WQA, Moa Island, Queensland on 3 October 2022.

Contributing factors

• The engine power loss was likely the result of fuel starvation.

Other factors that increased risk

• The pilot did not use the aircraft’s wing tip tanks in accordance with the flight manual. In addition to aircraft structural considerations, this also increased the likelihood of an inadvertent inappropriate fuel tank selection.
• For Britten-Norman Islander aircraft fitted with wing tip tanks, but without the alternative pilot interface per service letter number SL145, the configuration and location of the fuel controls and tank quantity gauges were probably not conducive to rapid and accurate interpretation.

Sources and submissions

Sources of information

The sources of information during the investigation included the:

• pilot of the accident flight
• aircraft operator
• UK Air Accident Investigation Branch
• Civil Aviation Safety Authority
• Queensland Police Service
• aircraft manufacturer
• Airservices Australia

Submissions

Under section 26 of the Transport Safety Investigation Act 2003, the ATSB may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. That section 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 following directly involved parties:

• pilot of the accident flight
• aircraft operator
• Civil Aviation Safety Authority
• UK Air Accident Investigation Branch.

Submissions were received from:

• UK Air Accident Investigation Branch.

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 2023 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]     Yawing: the motion of an aircraft about its vertical or normal axis.

[2]     Track: the path of the aircraft across the earth’s surface.

[3]     Stall warning system: activates an audible alert at a desired point above the stall.

[4]     Placard: a notice affixed to an aircraft that my contain warnings, limitation or reference information.

[5]     Aircraft flight manual: a document produced by the aircraft manufacturer that contains information on the specifications and operation of the aircraft.

[6]     Block time: elapsed period from when an aircraft starts to move at the beginning of the flight, to the time it comes to rest at the conclusion of the flight.

[7]     Avgas: a type of aviation fuel used in aircraft with a spark-ignited internal combustion engine.

[8]     Dipstick: a graduated tool that is inserted into a fuel tank and used to determine the level of fuel within the tank.

Executive summary

What happened

On 17 October 2022, at about 1345 local the pilot of a Mooney Aircraft Corporation M20J aircraft, registered VH-UDQ, departed Maitland Airport, NSW for a local flight. The pilot flew to Cessnock Airport, NSW and completed an orbit, then continued to Luskintyre airfield, NSW. When overhead Luskintyre airfield, the pilot broadcast their intent to track to Maitland Airport.

Witnesses near Luskintyre airfield reported observing the aircraft in what appeared to be descending to land. They further reported that smoke and flames were seen trailing the aircraft. At about 1359 VH-UDQ collided with terrain about 330 metres short of runway 30 with witnesses describing an explosion and accompanying fireball. 

The aircraft was destroyed by an intense post-impact fire. The pilot survived the collision but eventually succumbed to injuries sustained during the accident.

What the ATSB found

The ATSB determined that an O-ring seal fitted to the engine-driven fuel pump outlet fitting remained in service until it became age-affected and failed to provide an effective seal. The escaping fuel from the age-affected O-ring seal ignited and created an engine compartment fire. In response to the fire, the pilot initiated an emergency descent towards the runway but subsequently landed in a field resulting in the aircraft impacting trees. That led to a break-up of the aircraft and a severe post‑impact fire that consumed the aircraft. 

The ATSB also established that the aircraft had been recently refurbished. The refurbishment included repainting the aircraft and replacing interior furnishings with alternate materials. Neither the refurbishment activity nor details of the flammability assessment of the substituted materials, were recorded in the aircraft log books. The effect this refurbishment had on the in-flight fire, or the survivability of the pilot could not be determined.

What has been done as a result

In response to this accident, the Civil Aviation Safety Authority proposed to review Airworthiness Bulletins AWB 02-001 relating to on-condition maintenance, and AWB 85-004 regarding aircraft piston engine calendar time overhaul and most likely update them. The review would serve to highlight that ‘on-condition’ was not a ‘fit and forget’ approach to preventative maintenance. 

The ATSB has issued safety advisory notice SAN AO-2022-049-001 in conjunction with this investigation report. The SAN draws attention to the proactive replacement of O-ring seals fitted to engines and engine components, should inspection of aircraft records indicate they have been in service for a significant period.

Safety message

Piston engines, and the components necessary for their operation, installed in aircraft operating in the private or airwork category are permitted to remain in service beyond their recommended calendar time overhaul interval. As O-ring seals fitted to such engines are susceptible to deterioration due to age, being aware of their accumulated time-in-service may initiate replacement action before they fail. 

Maintaining an aircraft’s internal appearance may require the introduction of alternate materials when original products may no longer be available. Aircraft owners are encouraged to document aircraft refurbishment action in the aircraft logbook and to include details of materials if substituted, and their suitability for use in aircraft interiors. 

Summary video

The investigation

The occurrence

On 17 October 2022, at about 1345 the pilot of a Mooney Aircraft Corporation M20J aircraft, registered VH‑UDQ, departed Maitland Airport, NSW for a local flight. The aircraft travelled to the south-east before turning west towards Cessnock Airport. The pilot conducted an orbit of Cessnock Airport and by about 1353, was headed north-east towards Luskintyre airfield (Figure 1). At 1357:29, when about 2,150 ft above ground level (AGL)[1] during an orbit of Luskintyre airfield, the pilot made a broadcast on the common traffic advisory frequency[2] alerting other pilots in the area that they were overhead Luskintyre airfield and intending to track to join crosswind for runway 23[3] at Maitland Airport.

Figure 1: VH-UDQ Maitland to Luskintyre flight with inset showing flight path north of Sydney, NSW

Source: Google and AvPlan flight data. Annotated by ATSB.

Witnesses reported observing a light aircraft near the airfield that appeared to be descending to land. As VH-UDQ descended towards a line of trees to the south of their position, one witness observed flames coming from the front of the aircraft. The other witnesses reported seeing the aircraft with smoke trailing behind. No distress or emergency calls from the pilot were recorded (Figure 2).

Figure 2: Orbit and descent towards Luskintyre airfield

Source: Google and AvPlan flight data. Annotated by ATSB.

VH-UDQ continued its descent, and touched down in a grass field that was divided by a line of trees. At about 1359 VH-UDQ collided with the trees that were located about 330 metres short of runway 30 with witnesses describing an explosion and accompanying fireball. The collision with trees resulted in the break-up of the aircraft with the fuselage coming to rest about 70 m beyond the tree line. 

The witnesses proceeded to the scene and on arrival, they located the pilot lying in the field about 10 m away from the burning aircraft. The pilot advised the witnesses that the aircraft engine had caught fire and that they had suffered smoke inhalation. 

The aircraft was destroyed by an intense post-impact fire. Although the pilot survived the collision and escaped from the aircraft, the pilot sustained substantial burns. The pilot was taken to hospital but succumbed to these injuries about 10 weeks later. 

Context

Pilot information

The pilot held a Commercial Pilot Licence (Aeroplane) and a valid Class 1 aviation medical certificate. The pilot was issued with a private pilot licence in 1998 and a commercial pilot licence in 2008. The pilot, who was an associate of the aircraft owner, held the required ratings and endorsements for the flight. According to the last medical examination report in October 2021, the pilot reported that they had accumulated about 990 flying hours experience.

Aircraft information

General

VH-UDQ was a Mooney Aircraft Corporation M20J, manufactured in the United States in 1978 and assigned serial number 24-0588. The aircraft Log Book Statement for maintenance direction specified that the airframe was to be periodically inspected in accordance with Civil Aviation Safety Authority (CASA) Schedule 5 and the engine per AD/ENG/4 as amended. For special inspections and lifed component changes, the aircraft maintenance manual was to be consulted. 

Engine history

The aircraft was fitted with a 4-cylinder Lycoming engine, model IO‑360-A3B6D, serial number L‑22956-51A. Records show that the engine was last overhauled in May 1993 and had accumulated 1,806.3 hours, time-in-service since overhaul. At the time of the accident, the engine had been operating for 29.5 years since overhaul. 

The engine manufacturer recommended that the engine type was to be overhauled on accumulating 2,000 hours, time-in-service or every 12 years. Continued operation in service beyond 12 years was permitted provided the continuing airworthiness requirements of CASA AD/ENG/4 Amendment 11 for ‘on-condition’[4] operation was met. Additional maintenance related information is addressed in Airworthiness directives and airworthiness bulletins. 

This engine was installed in VH-UDQ in September 2021. The aircraft owner reported that the engine was replaced by the former owner prior to the sale of the aircraft. The engine had accumulated 27.4 hours since it was installed and the time of the accident. 

A piston engine condition report conducted at the 100‑hour inspection addressed the continuing airworthiness requirements did not identify any faults with the engine. The available aircraft records did not contain maintenance activity related to engine‑driven fuel pump replacement or repair.

Airframe information

At the last 100-hour periodic inspection on 14 September 2022, a maintenance release was issued, permitting day VFR operations and showed that the airframe had accumulated 19,128.3 hours, time-in-service.

Refurbishment

Prior to the 100-hour inspection, the aircraft had been refurbished by the current owner. The refurbishment included new external livery and replacement of interior furnishings. The interior refurbishment included re-covering of the seats with leather and the renewal of cabin and baggage compartment linings and floor panels. The owner confirmed that the backing material used for some of the cabin and baggage compartment linings were of carbon fibre construction. Details of the specific furnishings used in the aircraft were not recorded in the aircraft maintenance log books. 

Meteorological information

The Maitland weather observation recorded at 1400 was wind from 160º at 12 kt, greater than 10 km visibility, cloud bases broken at 3,500 ft and 7,900 ft above airfield elevation, temperature and dew point of 21º and 14º respectively, and QNH 1019 hPa with no rainfall recorded since 0900. The ten‑minute weather observations for Maitland recorded the wind direction from 165º at 11 kt with a maximum gust of 14 kt in the period 1350-1400.

Luskintyre airfield

Luskintyre airfield is located about 8 km to the north-west of Maitland Airport. The airfield consists of a grass runway 12/30[5] which is 760 m long. The airfield has an elevation of about 35 m above mean sea level and is surrounded by fenced pastures containing dwellings and occasional stands of trees lining boundary fences. 

Recorded data

Flight information sourced from AvPlan[6] indicated that at about 1358, VH-UDQ completed the orbit of Luskintyre airfield and commenced a descending turn to the right, reducing altitude from 1,863 ft AGL at 1358:02 to 1,597 ft at 1358:22. By 1358:32, VH-UDQ had begun a left turn and by 1358:42 was at 1,273 ft in an established turn and descending towards Luskintyre airfield. The final turn resulted in a tailwind component as the aircraft descended (Figure 3).

Figure 3: Aircraft flight - final turn towards Luskintyre airfield with time stamps 

Image description: Overhead view of flight path with displayed height referenced to the elevation of Luskintyre airfield.

Source: AvPlan and Google, annotated by ATSB

Analysis of the data when VH-UDQ was established in the final turn and approach to the tree line, indicated that the aircraft rate of descent peaked at 3,504 feet per minute. Prior to initial ground contact, when at 52 ft AGL, the rate of descent was 1,032 feet per minute and a ground speed of 108 kt was recorded (Table 1).

Table 1: Aircraft performance during descent towards Luskintyre airfield

 Source: ATSB 

Emergency procedures

The Mooney M20 pilot’s operating handbook (POH) advised pilots that in the event of an engine fire in‑flight, the emergency is to be treated in the following manner:

• Fuel Selector Valve – OFF
• Throttle and mixture – CLOSED and at IDLE CUTOFF
• Cabin ventilation and heating controls – CLOSED
• Landing gear – UP or DOWN, depending on terrain
• Flaps – Extend as necessary

The POH noted that if the fire is not extinguished, pilots should attempt to increase the airflow over the engine by increasing the glide speed and to attempt a power off landing. They should not attempt to restart the engine.

For a power off landing associated with an engine failure, the POH advised pilots to secure the engine by moving the mixture control to the idle cut-off position and to switch the magnetos off. Wing flaps were to be set to the full down position and undercarriage selected as necessary. Prior to landing, the master switch was to be in the off position and the aircraft approach speed to be 71 kts indicated airspeed. 

Site and wreckage information

The initial ground contact points were tyre marks in a grass field ahead of a tree line that separated two grass fields and was located about 330 metres before the runway 30 threshold. The tyre marks began about 67 metres from the tree line and ended about 22 metres before the tree line, indicating that VH-UDQ had become airborne again before colliding with the trees in a near-level attitude. The collision with the closely spaced trees separated the wings from the fuselage and was followed by a progressive break-up of the rest of the aircraft. The resulting wreckage trail was spread over a distance of about 70 metres, on a heading of about 282 degrees (magnetic). Ignited fuel from the ruptured wing tanks created the initial fireball reported by witnesses and generated an intense post‑impact fire that consumed the cockpit and cabin area of the fuselage (Figure 4).

Figure 4: VH-UDQ wreckage trail and location of major items

Source: ATSB

The complete aircraft was present at the accident site with all the flight control surfaces and major components accounted for. Examination of the aircraft control systems did not identify any defects that may have affected control of the aircraft. When examining the fuselage remains, the ATSB identified sections of carbon-fibre based products in the aircraft cabin and baggage compartment. 

The nose undercarriage was noted to be in the down and locked position however the left and right main undercarriages were found housed in the wheel wells. The main undercarriage actuation and locking mechanism was found to be broken and the damage attributed to the wings separating from the fuselage. 

The flap and tailplane trim actuators were retrieved from the accident site and details of their respective jack screw positions were provided to the aircraft manufacturer for comment. The aircraft manufacturer reported that in consideration of the position of the flap jack screw, the wing flaps were set in a slightly down position. The tailplane trim position could not be accurately determined. 

Assessment of ground slash marks created by propeller rotation indicated that the propeller was operating between 1,590 and 1,150 revolutions per minute.[7] The engine manufacturer reported that the propeller speed was above that expected for an engine to be windmilling. 

Due to the post-impact fire, the position prior to impact of cockpit switches and magneto selection, the fuel selector valve, the throttle and mixture controls, and the cabin ventilation and heating controls, could not be determined. 

Engine examination

A detailed engine examination was undertaken at the ATSB technical facilities in Canberra, ACT. The following observations were noted:

• Evidence of scorching and soot residue on engine hoses and components, presented as a distinct flame path that commenced at the rear of the engine and progressed to the front (Figure 5).
• The origin of the engine fire was localised to the area surrounding the engine-driven fuel pump.
• Pressure testing of the engine fuel supply and distribution hoses did not reveal any fluid leak associated with a hose defect.
• Pressure testing of the engine-driven, mechanical fuel pump detected fluid leakage at the base of the fuel outlet fitting of the pump. 

Figure 5: Underside of engine showing components and direction of flame travel

Source: ATSB

The fuel pump inlet and outlet fittings were removed to allow for examination of the O-ring seals installed between the fittings and the fuel pump body. Inspection of the outlet fitting seal revealed a deteriorated elastomer that had lost its pliability and its round, cross‑sectional shape. Further, the seal showed signs of permanent deformation and contained surface defects (Figure 6).

Figure 6: Fuel pump fittings and O-ring seals showing loss of natural shape and surface defects

Source: ATSB

The inlet fitting seal exhibited similar in-service defects as the outlet fitting seal, however the deterioration had not progressed to failing to provide an effective seal. The examination indicated that fuel under pressure was leaking past the outlet fitting O-ring. 

Maintenance records and requirements

Responsibility of registered operators – conduct and recording of maintenance

Section 3 of the Civil Aviation Act 1988, describes maintenance as:

Any task required to ensure, or that could affect, the continuing airworthiness of an aircraft or aeronautical product, including one or a combination of overhaul, repair, inspection, replacement of an aeronautical product, modification or defect rectification.

The owner or if appointed, the registered operator is responsible for the airworthiness and maintenance control of the aircraft to ensure its safe operation. Provision for the recording and certification of maintenance that is carried out are contained in the aircraft logbooks.

Maintenance activity or modifications which may include repainting of exterior surfaces or internal furnishings renewal, may change the aircraft’s empty weight and its centre of gravity position. If the change varies by more than the specified amount detailed in the aircraft’s weight and balance record, or the effect of the change is unknown, the aircraft may need to be re-weighed to determine its new empty weight and centre of gravity position. Maintaining the accuracy of the aircraft’s weight and balance information is necessary to ensure that changes do not adversely impact the performance of the aircraft as published in the aircraft flight manual.

To ensure that aircraft type certification standards are maintained and occupant survivability in the case of fire is not further degraded, modifications that include changes to materials used in interior furnishings are subject to approval by an authorised entity. The approval process will likely involve testing of the materials to evaluate their flash or flame resistance. 

Airworthiness directives and airworthiness bulletins

An airworthiness directive (AD) is a document issued by the aircraft State of Design or CASA, if an unsafe condition exists in a kind of aircraft or aeronautical product. Registered operators are to comply with the requirements of an AD or an approved alternate means of compliance. CASA also issues Airworthiness Bulletins (AWBs) to inform the aviation public of essential information or make recommendations that are not considered mandatory.

Airworthiness directives – continuing airworthiness of aircraft and aeronautical products 

CASA airworthiness directive AD/ENG/4, Piston engine continuing airworthiness requirements, was first issued in March 1995. Amendment 11 became effective on 15 January 2009 and specified the piston engine continuing airworthiness requirements. It required that registered operators of aircraft operated in either the private or airwork categories, in addition to scheduled engine maintenance, conduct additional maintenance actions to confirm the serviceability of the engine. 

AD/ENG/04 also referenced CASA Airworthiness Bulletin AWB 85-004, Aircraft piston engine calendar time overhaul, that provided guidelines for additional inspections related to engine calendar time overhaul. The purpose of this AWB was specified as: 

This AWB provides guidelines for procedures to be followed to ensure continued airworthiness of the engines that have exceeded the calendar time overhaul limits specified by the manufacturer. These guidelines are in addition to the recommendations by the manufacturers relating to inspections for corrosion.

The AWB highlights corrosion and degradation of elastomers as factors contributing to engine deterioration associated with calendar time. The AWB recommended inspecting elastomer components such as engine mounts, hoses and other elastomer related items for deterioration, however O-ring seals fitted to engine components were not specifically mentioned. 

Airworthiness Bulletins

While airworthiness bulletins are for information only, they contain useful information concerned with the airworthiness of aircraft and aeronautical products. CASA recommends that all aircraft owners, and other key stakeholders involved in the operation or maintenance of aircraft, review AWBs for applicability and take any action they consider appropriate.

Airworthiness Bulletin AWB 02-1, On-condition maintenance, was issued on 27 November 2001, and provided information relating to ‘on-condition’ maintenance. 

‘On-condition’ maintenance means that the condition of the item is monitored continuously or at specified intervals. When its performance or physical appearance is observed to no longer meet an appropriate standard, the resultant action is the removal of the item before it fails in service. Items in the engine may remain in service longer than the manufacturer recommended time‑in‑service, provided they continue to meet desired physical conditions and performance standards. The AWB identified that ‘on-condition’ is not an opportunity to fit and forget until a failure occurs. The AWB further advised that the condition of an item may require appropriate judgement to determine that failure of the item will not occur prior to the next scheduled inspection.

CASA recommended that where applicable, certificate of registration holders utilise the philosophy of ‘on condition’ maintenance to detect the potential for failures of critical items or products, especially when the time-in-service is approaching the manufacturer’s recommended overhaul period.

Flammability resistance

CASA (2011) AWB 25-016 Cabin interior and cargo compartment flammability provided guidance regarding flammability requirements for aircraft material and advised that when repairing or replacing interior material in an existing aircraft, the applicable flammability requirements are to be understood, and compliance with, is shown. 

Factors to consider are the minimum flammability requirements for the aircraft, which is dependent on the aircraft category and the standard applicable at the time that the aircraft was first certificated. Certification information is contained in a document usually referred to as the Type Certificate Data Sheet. 

The Type Certificate Data Sheet for the Mooney M20 showed that the aircraft was certified under Part 3 of the United States Civil Air Regulations as amended in 1956, which required that materials making up the cabin interior be ‘flash resistant’, or ‘flame resistant’ if the compartment could be used for (cigarette) smoking. 

For a material to be flash resistant, AWB 25-016 notes that the material is to be of a type that is not susceptible to burning violently when ignited. To be flame resistant, it is not to be susceptible to combustion to the point of propagating a flame beyond a limit when the ignition source is removed.

Elastomer type products used in aircraft systems

Aircraft and engine systems carrying petroleum-based products use fluid carrying hoses, seals and gaskets that commonly involve nitrile rubber in the manufacturing process (Brink, Czernik and Horve, 1993). Nitrile rubber is a widely used, synthetic elastomer that is resistant to oil, fuel and chemicals with inherent properties that include tear and abrasion resistance, tensile strength, resilience,[8] and compression set.[9] Distinguishing characteristics of elastomer type products are their ability to be stretched up to several hundred percent and given their ability to store energy, to return to their near original shape when significantly compressed (Brady and Clauser,1991). Hoses, seals and gaskets are normally subject to handling, inspection and replacement requirements while in service. 

Following installation, standard aviation maintenance practices described in US Advisory Circular AC-43-13, Acceptable methods, techniques and practices-aircraft inspection and repair (FAA, 1998), require that O-ring seals should not be re-used if disturbed during maintenance or disassembly of the part to which they are fitted. The O-ring seal may have swelled from exposure to fluid, hardened over time, or gained a permanent set. When installed as a gasket within a recessed area to seal a fluid carrying fitting, leaks are not normally acceptable. However, opportunities to replace or physically examine O-ring seals are reliant on scheduled maintenance where the component to which they are fitted is targeted for attention, or unscheduled maintenance in response to the component developing a defect when in service.

As specified by the Logbook Statement,[10] contained in VH-UDQ’s logbook, when considering aircraft components, reference is to be made to the overhaul and replacement schedule provided in the Mooney M20J aircraft maintenance manual. The schedule specifies that fuel and oil system flexible hoses containing elastomer type material are replaced every two to seven years, or at engine overhaul, whichever occurs first. The schedule also specifies that all other fuel and oil system components such as an engine‑driven fuel pump are treated as on-condition items and may remain in service until an inspection, or their performance reveals an unserviceable condition.

Additional information

Annex 13 to the convention on international civil aviation is published by the International Civil Aviation Organization and details the standards and recommended practices for aircraft accident and incident investigation. For statistical purposes, an injury is classified as a fatal injury when death results within thirty days of the date of the accident. The pilot involved in this accident did not succumb to injuries until after the 30-day period, and therefore, the accident is not registered as a fatal collision, despite the outcome. 

Related occurrences

Between 2014 and 2023, the ATSB investigated two accidents (described below) involving aircraft in-flight fires that resulted in a fatality or serious injuries. In one accident, the origin of the fire was attributed to burning oil due to engine failure while the other was attributed to a fuel-fed cabin fire. 

ATSB investigation AO-2014-059

On 28 March 2014, the pilot of a Cessna Aircraft Company 210L aircraft departed Numbulwar, Northern Territory, on a charter flight with one passenger on board for a return flight to Tindal, Northern Territory. 

When about 22 NM west of Numbulwar, smoke was detected in the cabin and following a ‘MAYDAY’ call advising of an engine failure, the pilot applied the cabin fire extinguisher which quickly stopped the smoke. The aircraft landed heavily and collided with trees. The pilot and passenger suffered serious injuries and the aircraft was destroyed.

The ATSB determined that a piston connecting rod broke resulting in a catastrophic engine failure. The smoke entering the cockpit was likely from burning oil.

ATSB investigation AO-2022-026

On 16 April 2022, the pilot of a Beechcraft B58 Baron aircraft, on a charter flight with one passenger on board from Broome Western Australia, commenced a straight in approach to runway 12 at the East Kimberley Regional Airport, Western Australia. 

When attempting to extend the landing gear, smoke emerged from forward of the pilot’s circuit breaker panel. The pilot made a PAN-PAN call, by which time flames were observed to be coming from the area of the source of the smoke. Attempts to extinguish the fire were unsuccessful with flames and thick smoke filling the cockpit. 

The aircraft collided with terrain about 800 m from the threshold of runway 12 and was consumed by the post-impact fire. The passenger succumbed to their injuries and the pilot received serious injuries.

The ATSB determined that a fault associated with the landing gear electrical system likely ignited fuel from the cabin heater supply line, resulting in a significant and sustained cockpit fire.

NTSB investigation

A similar occurrence in the United States was investigated by the National Transportation Safety Board. The accident was attributed to an engine compartment fuel leak and described in NTSB report, ERA17LA284 as summarised below. 

On August 20, 2017, the pilot of a Mooney M20C aircraft registered N6833N departed Palm Coast, Florida on a private flight with one passenger on board to Fort Lauderdale, Florida. 

During take‑off and climb the pilot noticed that the engine was losing power. Upon reaching 400 ft above ground level the engine lost total power. During touchdown for the subsequent forced landing to a road, the pilot noticed flames coming into the cabin near the passenger’s feet. The pilot was able to stop the aircraft and both occupants were able to egress quickly after landing. 

The pilot and passenger were not injured however the aircraft was substantially damaged. 

The NTSB investigation found that a crack in the engine’s No. 4 cylinder resulted in a rough running engine and a subsequent loss of power. The increased vibration loosened a fuel line fitting that generated a fuel leak and was the source of the in-flight fire.

Safety analysis

Introduction

On 17 October 2022, at about 1345 the pilot of a Mooney Aircraft Corporation M20J aircraft, registered VH-UDQ, departed Maitland Airport, NSW for a flight in the local area. While overhead Luskintyre airfield the pilot broadcast an intent to track to Maitland Airport. However, witnesses nearby saw the aircraft descending towards Luskintyre airfield while trailing smoke and flames. Following an initial touch-down in a grass field, VH-UDQ collided with a line of trees. The witnesses described an explosion and accompanying fireball. The pilot was seriously injured but subsequently succumbed to injuries after 10 weeks. The aircraft was destroyed.

The following analysis will consider the cause of the in-flight engine compartment fire, the O‑ring seals of the engine fuel pump, and the recording of maintenance activity in the aircraft logbook. The risk to occupant safety when substituting materials used for interior furnishings is also discussed.

Fire and emergency descent

Witnesses in the area reported flames and smoke coming from the aircraft, and the pilot reported to first responders that the engine was on fire and that they had inhaled smoke. In consideration of the leak found with the engine-driven fuel pump outlet O‑ring seal (see O‑ring seal replacement below), it was likely that fuel leaking from the pump outlet fitting produced a vapor that was ignited by heat in the lower engine compartment. 

Given that the pilot made a normal, inbound call for a landing at Maitland Airport, it is likely the pilot only became aware of the presence and/or extent of the engine compartment fire after the orbit of Luskintyre airfield and after the inbound call. However, once the pilot became aware of the fire, they abandoned their intention to land at Maitland Airport and the flight path suggests they instead intended to land at Luskintyre airfield due to its proximity. 

The approach towards runway 30 at Luskintyre airfield was at a very high rate of descent, which was consistent with an emergency reaction to a fire that the pilot perceived as high risk to their safety. The available recorded data shows that the descent from 1,522 ft to 52 ft occurred in 50 seconds. From the available evidence it was not clear if, during that time, all of the pilot operating handbook checklist items for an in-flight fire were conducted. One item included cutting off fuel to reduce the source of fire, but the calculated propeller speed at collision suggested that the engine was rotating at well above windmilling speed and was not consistent with configuring the aircraft for a power off landing.

While a landing on a runway is generally safer than in a field, the high descent rate close to the ground may have limited the pilot’s ability to reach runway 30. However, it is also possible that the pilot was becoming more affected by heat, fire and/or smoke as the aircraft descended, and made a decision to land in the paddock to expedite the landing. 

The relatively wings-level attitude when the aircraft touched down and then collided with the tree line indicates the aircraft was in an attitude appropriate for landing, albeit at a high ground speed and descent rate. The high speed at which the aircraft’s wheels first contacted the ground and the proximity of the tree line from the initial ground contact point indicated that the collision with trees and subsequent break-up of the aircraft was unavoidable. The release and atomisation of fuel as the wings separated from the fuselage was likely ignited by the pre-existing fire within the engine compartment and the generated heat ignited combustible materials of the aircraft. 

As the fuselage was subsequently consumed by the post-impact fire, it was not possible to determine whether cockpit furnishings had started to burn in-flight due to radiant heat from the engine compartment fire, or that cockpit related smoke in-flight, contributed to the reported smoke inhalation.

O-ring seal replacement

An engine compartment fuel leak during the accident flight originated from the engine‑driven, mechanical fuel pump outlet fitting due to an O‑ring seal that had deteriorated with age and had lost the ability to provide an adequate seal. Maintenance related standard practices specify that O‑ring seals are not reused, and they should be replaced when disturbed during maintenance or when a leak is detected.

Unlike the engine fluid carrying hoses that were periodically replaced per the aircraft’s maintenance schedule, the fuel pump and the O-ring seals were on condition and had not required replacement while the engine was in service. Additionally, the maintenance records did not contain evidence of other opportunities to access and physically examine the condition of the O-ring seals that are normally hidden from view when installed on the fuel pump fittings. 

O-ring seals were not specifically identified in the broad classification of elastomer type products described in CASA airworthiness bulletin AWB 85-004. Since the O-ring seals were not specifically targeted for attention, this may have resulted in them being overlooked for proactive replacement of easily accessible seals. 

In the absence of physical examination or replacement opportunity, assurance that the O‑ring seals were functional, was reliant on checking for fluid leaks at specific intervals. When last inspected it was likely that no discernible leak was detected at the fuel pump outlet fitting. However, the engine was about 17.5 years past its normal overhaul calendar life and the potential for a leak to develop between scheduled inspection intervals was increasingly likely as the O‑ring seals continued to age. A review of the records may have identified the prolonged time-in-service of O‑ring seals, and if documented, the modifications that may have affected the weight of the aircraft, or the flammability attributes of materials used during interior refurbishment.  

Flammability assessment

The interest in giving the aircraft a renewed look may have influenced the selection of products as interior furnishings. The use of alternate products was not prohibited. However, to understand the risk associated with substituting materials, an assessment of the substituted materials’ flammability was required. When conducted, the assessment would have indicated the suitability of the material in terms of potential impact on occupant safety in the case of a fire. 

Regarding material type, those having flame or flash resistant attributes were permitted for use as cabin furnishings and was dependent on whether smoking was permitted in the aircraft. Considering the common practice of not permitting smoking in aircraft interiors, assurance that occupant safety was not compromised could have been achieved by using products that at a minimum, were determined to be flash resistant. However, a record of their use and their suitability needed to be included in the aircraft records. 

Due to the intensity of the post-impact fire, and the lack of records, the ATSB was unable to determine the full extent of product or material substitution, or whether the replacement materials had fed or suppressed the fire. 

Findings

From the evidence available, the following findings are made with respect to the collision with terrain involving Mooney Aircraft Corporation M20J, registration VH-UDQ, near Luskintyre airfield, NSW on 17 October 2022.

Contributing factors

• In response to an engine compartment fire, the pilot initiated an emergency descent towards the runway but subsequently landed in a field resulting in the aircraft impacting trees. That led to a break-up of the aircraft and a severe post‑impact fire that consumed the aircraft.
• An O-ring seal fitted to the engine-driven fuel pump outlet fitting remained in service until it became age-affected and failed to provide an effective seal. The escaping fuel ignited and created an engine compartment fire. 

Other factors that increased risk

• The aircraft had recently been refurbished and changes included new external livery and replacement of interior furnishings. The aircraft records did not indicate that the refurbishment had been conducted, and consequently, there was no evidence of a flammability assessment for the materials used. 

Safety actions

Safety action by the ATSB

In response to this accident, the ATSB has issued safety advisory notice SAN AO‑2022‑049‑001 in conjunction with this investigation report. The safety advisory notice draws attention to the management of O-ring seals in engine components and their pro‑active removal from service, should inspection of aircraft records indicate a significant time‑in‑service.

Safety Action by the Civil Aviation Safety Authority

In response to the investigation finding related to deteriorated O-ring seals, the Civil Aviation Safety Authority proposed to review Airworthiness Bulletins AWB 02-001 and AWB 85-004 and most likely update them. The update would serve as a reminder to industry of some of the concepts and philosophy related to ‘on-condition’, which is not a ‘fit and forget’ approach to preventative maintenance.  

Sources and submissions

Sources of information

The sources of information during the investigation included the:

• Civil Aviation Safety Authority
• aircraft manufacturer
• registered operator
• maintenance organisation for VH-UDQ
• accident witnesses
• recorded data from AvPlan. 

References

Brady, G.S. and Clauser, H. R. (1991), Materials Handbook: An Encyclopaedia for Managers, Technical Professionals, Purchasing and Production Managers, Technicians, Supervisors, and Foremen, McGraw-Hill, Inc. New York, NY, USA

Brink, R.V., Czernik, D.E., and Horve, L.A. (1993), Handbook of fluid sealing, McGraw-Hill, Inc. New York, NY, USA.

Civil Aviation Safety Authority (2001), Airworthiness Bulletin AWB 02-1 Issue 1, On-condition maintenance, Civil Aviation Safety Authority, Canberra, ACT, Australia.

Civil Aviation Safety Authority (2005), Airworthiness Bulletin AWB 85-004 Issue 1, Aircraft Piston Engine Calendar Time Overhaul, Civil Aviation Safety Authority, Canberra, ACT, Australia.

Civil Aviation Safety Authority (2009), Airworthiness Directive, AD/ENG/4 Amendment 11, Piston Engine Continuing Airworthiness Requirements, Civil Aviation Safety Authority, Canberra, ACT, Australia.

Civil Aviation Safety Authority (2016), Airworthiness Bulletin AWB 25-016 Issue 2, Cabin Interior and Cargo Compartment Flammability, Civil Aviation Safety Authority, Canberra, ACT, Australia.

Civil Aviation Safety Authority (2017), Airworthiness Bulletin AWB 00-001 Issue 3, Airworthiness Bulletins, Civil Aviation Safety Authority, Canberra, ACT, Australia.

Federal Aviation Administration (1998), Advisory Circular 43.13-1B, Acceptable methods, techniques and practices-aircraft inspection and repair, U.S. Department of Transportation, Oklahoma City, OK, USA.

Submissions

Under section 26 of the Transport Safety Investigation Act 2003, the ATSB may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. That section 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 following directly involved parties:

• aircraft manufacturer
• Civil Aviation Safety Authority
• engine manufacturer
• National Transportation Safety Board
• registered operator/owner.

Submissions were received from the Civil Aviation Safety Authority. 

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 2024 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.

Executive summary

What happened

On the afternoon of 28 September 2022, a Boeing Company 787-9 (787), registered G-ZBKF, was being operated by British Airways as flight number BA16, on an air transport flight from Sydney, New South Wales, to Singapore. At the same time, an Airbus A330-202 (A330), registered VH-EBK, was being operated by Qantas Airways as flight number QF926, on an air transport flight from Sydney to Cairns, Queensland.

At approximately 1508 local time, BA16 was cleared for take-off from runway 16R to conduct the DEENA SEVEN (DEENA 7) standard instrument departure (SID). Shortly after, QF926 was also cleared for take-off conducting the same SID. The Departure controller cleared both aircraft, in sequence, to climb to flight level 280 via the SID.

During the departure, the following aircraft climbed faster than the controller anticipated and turned towards the next waypoint inside the preceding aircraft, resulting in a loss of separation.

What the ATSB found

The ATSB found that the actual climb performance of the A330 was greater than the climb performance of the 787 which was not expected, or identified, by the controller due to their focus on other aircraft. This resulted in the A330, while complying with the SID, turning inside the leading 787, resulting in the loss of separation.

The ATSB also found that the design of the DEENA SEVEN SID did not provide a positive method of providing separation assurance to aircraft with different performance characteristics. As the aircraft had to satisfy 2 separate conditions prior to turning, there was no way of ensuring aircraft would turn at the same distance from the airport. That is, separation could not be assured.

What has been done as a result

Airservices Australia advised that the DEENA SEVEN SID has been redesigned to remove the conditional requirements of the procedure. The change has been approved and planned to be released as part of the Western Sydney International (Nancy-Bird Walton) Airport project.  At the time of writing Airservices Australia was developing a timeline for the implementation of the redesigned SID. As such, the ATSB will continue to monitor the safety issue and provide website updates.

Safety message

Maintaining separation in high traffic terminal areas, such as Sydney, requires that both controllers and flight crews remain vigilant, maintain open communications, and use the available systems and tools to minimise the risk of errors. When sequencing departures, controllers should consider a number of factors, including how the flight duration (and the associated fuel load), will likely affect aircraft climb performance.

Standard instrument departures are designed to expedite the safe and efficient flow of air traffic operating from airports through the use of specific routings, levels, speed restrictions and waypoints. Where a SID, with limited designed separation assurance is used, it is important that air traffic controllers regularly monitor individual aircraft performance rather than rely on expected flight characteristics.

The occurrence

On the afternoon of 28 September 2022, a Boeing Company 787-9 (787), registered G-ZBKF, was being operated by British Airways as flight number BA16, on air transport flight from Sydney, New South Wales, to Singapore. At the same time, an Airbus A330-202 (A330), registered VH-EBK, was being operated by Qantas Airways as flight number QF926, on air transport flight from Sydney to Cairns, Queensland.

The Sydney Airport automatic terminal information service (ATIS)[1] indicated that parallel runways were in use, including simultaneous independent departures[2] from runways 16R and 16L.[3]

At approximately 1508 local time, BA16 was cleared for take-off from runway 16R by the Sydney tower controller. They were cleared to conduct a DEENA SEVEN (DEENA 7) standard instrument departure (SID)[4] (Figure 1) from runway 16R and to climb to 5,000 ft.

The DEENA 7 SID required an aircraft to turn right as soon as practicable after take-off, towards the DUNES waypoint[5] and then turn left to track 152° towards the DEENA waypoint. The aircraft was required to have passed DEENA and climbed above 6,000 ft prior to turning right towards the ANKUB waypoint. If the aircraft had not reached 6,000 ft at DEENA, there was a requirement to continue on a heading of 152° until passing an altitude of 6,000 ft, before turning towards ANKUB.

Passing 1,500 ft, the flight crew contacted the departure controller (Departures) and advised them that they were a heavy aircraft[6] on climb to 5,000 ft, departing via the DEENA 7 SID. Departures advised the flight crew the aircraft was identified on radar and instructed them to climb via the SID to flight level (FL) 280.[7]

Figure 1: DEENA SEVEN standard instrument departure (SID) from runway 16R

Source: Airservices Australia, annotated by ATSB

At approximately 1511, and with 6.3 NM spacing behind BA16, QF926 departed from runway 16R, also cleared on a DEENA 7 SID. During the initial climb, the flight crew contacted Departures advising that they were a heavy aircraft, passing 1,500 ft on climb to 5,000 ft, cleared on a DEENA 7 SID. Departures subsequently cleared them to climb to FL 280 via the SID.

BA16 climbed through 6,000 ft at around 25 km from Sydney Airport and initiated a turn to ANKUB in accordance with the SID. QF926 passed DEENA approximately coincident with climbing through 6,000 ft and initiated a turn to ANKUB at approximately 20 km from the airport, also in accordance with the SID (Figure 2).

Figure 2: Flight tracks of both aircraft during their departure from Sydney

Qantas Q926 is depicted in orange and BA16 is depicted in green.

Source: Google Earth with Flight Radar 24 tracks and annotated by ATSB

As QF926 was turning, the controller detected that the aircraft had climbed faster than they had anticipated and instructed QF926 to stop their climb at 9,000 ft, to which the flight crew advised they would do their best. The controller then instructed BA16 to expedite their climb through 10,000 ft.

The controller subsequently advised the flight crew of QF926 that there was a 787 above their aircraft, and they would step climb their aircraft underneath the 787. The Qantas flight crew advised they had sighted the higher aircraft.

There was a loss of separation standards (see the section titled Required separation) between the two aircraft with the minimum vertical separation reducing to 600 ft and lateral separation reducing to 2.4 NM.

The flight crew in BA16 later advised that they had received a traffic alert and collision avoidance system (TCAS)[8] traffic advisory[9] during the event and the first officer subsequently visually identified the A330. They also advised they were informed by air traffic control of the traffic, however this was not recorded on the departures frequency.
 

Context

Safety analysis

ATC                  Air traffic control

ATIS                 Automatic terminal information service

FL                    Flight level

IAS                   Indicated airspeed

MATS               Manual of air traffic services

SID                   Standard instrument departure

STCA               Short-term conflict alert

TCAS                A type of airborne collision avoidance system (ACAS).

TCU                  Terminal control unit

Sources and submissions

Sources of information

The sources of information during the investigation included the:

• involved air traffic controller
• flight crew notifications
• Airservices Australia
• Skybrary

Submissions

Under section 26 of the Transport Safety Investigation Act 2003, the ATSB may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. That section 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 following directly involved parties:

• Involved air traffic controller
• Airservices Australia
• Civil Aviation Safety Authority
• Qantas
• Air Accidents Investigation Branch, United Kingdom

A submissions was received from:

• Airservices Australia

The submission was 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 2023 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 investigation

The occurrence

Flight from Jandakot to Koorda

On 1 October 2022, a Robinson Helicopter Company R22 Beta II, registered VH‑RAS, departed Jandakot Airport, Western Australia for a private flight to an airstrip at Koorda (Figure 1 inset). The helicopter and several aeroplanes were being used to participate in a group social flying weekend, with the pilots and their passengers following a common itinerary but operating independently. The helicopter was being operated under the visual flight rules.[1]

The group arrived around midday at Koorda, and a witness recalled assisting the pilot of VH-RAS to fuel the helicopter with avgas[2] from 2 full 10 L plastic fuel containers, which they had been given by the pilot to drive to Koorda from Jandakot. This fuel was loaded into the main fuel tank. This witness also observed the pilot conduct a fuel drain during the pre‑flight inspection the next morning.

Return flight to Jandakot

On the morning of 2 October 2022, the group departed at staggered times from Koorda, on the return flight to Jandakot. VH-RAS departed at about 1140 local time, with the pilot and one passenger on board. It was reported that the pilot intended to refuel at Northam aerodrome, about 117 km to the south-west of Koorda. 

When VH-RAS did not arrive at Jandakot as expected, and after confirming it had not landed at Northam, the helicopter was reported as overdue at about 1500 and a search was commenced. The search was coordinated by the Australian Maritime Safety Authority’s Joint Rescue Coordination Centre. The wreckage was subsequently located at about 1600 on a dry salt flat in the Cowcowing Lakes region, about 13 km south-west of Koorda (Figure 1). The helicopter was destroyed, and both occupants were fatally injured. 

Figure 1: Flight from Jandakot to Koorda and from Koorda on the morning of the accident

Source: Google Earth and handheld GPS data, annotated by the ATSB

Context

Pilot information

General

The pilot was the owner of VH-RAS and held a private pilot licence for both aeroplanes (since 1976) and helicopters (since 2006). Their last flight review, conducted in VH‑RAS, was completed 11 January 2021. 

The pilot held a Class 2 Aviation Medical Certificate valid to 22 January 2024, which included restrictions requiring the wearing of distance vision correction and that reading correction must also be available while exercising the privileges of the licence. The pilot reportedly did not wear glasses while flying. However, their flight instructor and other pilots who flew with them, advised that they had not noticed any impediment with the pilot’s ability to operate the helicopter without wearing distance vision correction or at other times when reading the flight instruments.

Aeronautical experience

At their last medical in January 2022, the pilot reported that they had accrued about 3,000 hours total aeronautical experience (combined aeroplane and helicopter). It was also reported that the pilot no longer maintained their logbook, which prevented an accurate determination of their experience specific to aeroplanes and helicopters.

There were 16 entries on the current maintenance release, indicating the pilot typically flew the helicopter about twice per month, for about 20 minutes each time.[3] A review of previous maintenance releases determined that the pilot had operated the helicopter for a total of 77 hours since its purchase in 2016.[4] 

The pilot was reported to have regularly flown the helicopter accompanied by an experienced aeroplane commercial pilot and instructor, who also held a helicopter licence. Those flights would typically occur each month and involve about 45-minutes[5] of skill‑based flying practice, consisting of general handling, hovering, hovering turns and slope landing practice. No simulated emergencies or abnormal operations were practiced during those flights. In addition to those flights, they would do short flights to properties in the Perth hills or to nearby aerodromes to participate in social ‘fly-in’ breakfast events.

The pilot’s January 2021 helicopter flight review was performed by an experienced helicopter instructor and flight examiner who had conducted a significant amount of the pilot’s training and testing over the years. The flight review included simulated engine failures, autorotation[6] to confined areas, main rotor under speed recoveries and stuck tail rotor pedal emergencies. 

In addition to the flight review, the instructor recalled accompanying the pilot on other occasions during the period since January 2021. During this, the pilot had the opportunity to practice simulated emergencies under the instructor’s supervision.

The aviation community at Jandakot held the accident pilot in high regard and recognised them as somebody who had natural aptitude as both an aeroplane and helicopter pilot.

Recent history

The pilot was reported to typically retire to bed about 2100 and wake about 0500 most days. They were described consistently as being very health conscious, having exercised regularly and maintained very good overall health.

Members of the flying group that had spent time with the pilot during the weekend recalled them being their normal self, including prior to departing Koorda the morning of the accident. The pilot was reported to have slept normally, was well rested and had eaten breakfast with the group prior to arriving at the airstrip. There was no evidence to indicate the pilot was unwell nor experiencing a level of fatigue known to affect performance.

Passenger information

The passenger held a Student Pilot Licence (Aeroplane), with about 15 hours of flight experience in a Cessna Aircraft Company 152. They also held a valid Class 2 Aviation Medical Certificate, which included restrictions requiring the wearing of distance vision correction. It was stated that this was the passenger’s first flight in a helicopter. The passenger was reported to be well and that they appeared normal during the weekend, including prior to departing Koorda.

Helicopter information

General

VH-RAS (Figure 2) was a 2-seat Robinson Helicopter Company (RHC) R22 Beta II helicopter, serial number 4617, powered by a Textron Lycoming O-360-J2A, 4-cylinder carburetted piston engine. It was manufactured in the United States in 2013 and first registered in Australia the same year. It was purchased by the pilot in 2016 and had been maintained by the same maintenance organisation since that time. The helicopter was to undergo a periodic inspection every 100 hours or 12 months, whichever came first. The last periodic inspection had been completed in November 2021, with the current maintenance release issued at that time. At the time of the accident, the helicopter had accrued about 13 hours since the periodic inspection and 2,080 hours total time-in-service.

The helicopter was equipped with an intercom system and headsets, allowing those on board to talk to each other during flight. This would allow for clear communication between the pilot and passenger in the event of an abnormal situation. 

Figure 2: VH-RAS

Source: Dallas Presser, modified by the ATSB

Rotor head design

The 2-blade main rotor assembly is a semi-rigid rotor head, otherwise known as a teetering rotor head (Figure 3). Bolts secure the rotor blades to the main rotor hub at the coning hinges.[7] The teetering head allows the rotor disk to take up an attitude appropriate to the flight condition, with the coning hinges allowing the blades to achieve the optimum coning angle appropriate for the flight and loading conditions. This design permits a lighter blade structure than would otherwise be required. 

During stopping and starting of the main rotor, when the revolutions per minute (RPM) are low, the spindle tusks rest against the droop stops, which are mounted near the top of the main rotor mast. The droop stops restrict the drooping of each blade to prevent contact with the tailcone at low rotor RPM. As the main rotor RPM increases, the blades become rigid and straighten due to rotational forces, and the tusks shift off the droop stop as the blades lift. During normal flight, the rotor is free to teeter and flap around its designed flight axis via the teeter hinge, while polyurethane teeter stops limit the degree of teetering to protect the mast from direct contact with the main rotor blades. The main rotor system is also fitted with pitch change links, which connect the swashplate to the pitch horn and transmit the flight control inputs to the main rotor blades.

Figure 3: R22 main rotor head assembly

Source: United Kingdom Air Accidents Investigation Branch, annotated by the ATSB

Flight controls

The tail rotor pedals change the pitch of the tail rotor blades, and therefore the thrust of the tail rotor system, which provides directional control. The collective lever controls the amount of thrust (lift) produced by the main rotor disc. Raising or lowering the collective lever will raise or lower the swashplate,[8] which will alter the pitch on both main rotor blades to increase or decrease the main rotor thrust. The collective lever also incorporates a twist grip to provide the pilot with full manual control of the engine throttle. The cyclic[9] control tilts the main rotor disc to point the rotor thrust in the desired direction of flight. Fore‑aft movement of the cyclic provides the longitudinal (pitch) control of the main rotor disc. Left‑right movement of the cyclic provides lateral (roll) control of the main rotor disc. 

The helicopter was fitted with quick-disconnect dual flight controls (exemplar shown in Figure 4).[10] This consisted of duplicated collective and tail rotor pedal controls for both the pilot (right) and passenger (left) seating positions, and an additional control arm to the left of the T-bar cyclic control, enabling the cyclic control of the helicopter to be transferred between positions. In that configuration, the T-bar cyclic control would tilt down to be directly in front of the person controlling the helicopter. The opposite side cyclic control would still be connected and in front of the other seating position, but in a significantly higher position than normal (Figure 4 inset). The dual controls for the left seat position could be easily removed for passenger carrying operations. Notably, the R22 Pilot’s Operating Handbook (POH) included the caution: 

…remove left seat controls if person in that seat is not a rated helicopter pilot.

Figure 4: Dual flight controls, with insert showing deflection when being operated from the right 

Note: While the insert image is from a R44 helicopter, it demonstrates the position of the opposite cyclic grip when the T-Bar is oriented to the pilot sitting in the right seat, with similar see-saw orientation when being held by the left seat pilot.

Source: Robinson Helicopter Company, annotated by the ATSB

Fuel system

The fuel system consists of a main tank (left side, 69 L) and an auxiliary tank (37 L, right side). Fuel is gravity-fed (no pumps) via the gascolator to the carburettor.

Engine governor system

Under normal conditions, the governor senses engine RPM and makes adjustments to the throttle control to maintain a constant engine RPM, which leads to a constant rotor RPM in flight. It can be selected on or off using the toggle switch on the right seat collective. The POH stated that the governor may not prevent over- or under-speed conditions generated by aggressive flight manoeuvres. In the event of malfunction, the pilot can override the governor and manipulate the throttle to maintain engine RPM, until the governor can be selected off, or rendered inoperative by pulling the circuit breaker.

Carburettor heat system

The helicopter was fitted with a carburettor heat system, which directed hot air collected from a scoop installed on the engine exhaust system, via a duct, to the engine induction air box. Within the airbox was a sliding guillotine-type valve to proportion the mix of cool and heated air. The pilot could monitor the temperature of the carburettor air using the carburettor air temperature gauge on the instrument panel console.[11] The carburettor heat control knob was situated aft and rear of the cyclic, with ‘down’ being no heat and ‘up’ providing full heat, or anywhere in between as selected by the pilot. This heated air prevented the temperature within the carburettor from dropping to at or below the freezing point of water. While the increase in carburettor air temperature would result in a small reduction in power, this could be countered by increasing throttle and manifold pressure.

The helicopter also had a carburettor heat assist system, which automatically applied carburettor heat when lowering the collective, generally for descent, to reduce pilot workload. The pilot could override the heat assist. In addition, a latch was provided at the carburettor heat control knob to lock the heat assist off when not required.

Weight and balance

Weight and balance for the accident flight was calculated using the approved weight and balance record, together with the estimated fuel load, occupant weights and recovered luggage. To calculate the fuel on board departing Koorda, the ATSB used scenarios where the helicopter was fully‑fuelled prior to departing Jandakot and the average fuel consumption (as specified by RHC) was between 7 and 10 US gallons per hour (26–38 L/hr). Using those rates, the ATSB estimated that, on engine start at Koorda, the helicopter had between 40‑54 kg (55–75 L) of fuel onboard, which included the 20 L added the day before.

Irrespective of the fuel load departing Koorda, the helicopter’s lateral and longitudinal centre of gravity remained within the limits published by the manufacturer. However, the helicopter would have been at about its maximum gross operating weight if departing Koorda with 40 kg of fuel but could have exceeded the maximum gross weight by about 15 kg if departing with 54 kg of fuel. Despite this, as the helicopter was observed to have departed normally, and reached a cruise altitude, this would suggest that the overall weight did not significantly reduce the helicopter’s performance.

Meteorological information

Bureau of Meteorology forecast and analysis

The graphical area forecast prepared by the Bureau of Meteorology indicated visual meteorological conditions[12] were expected during the flight to Jandakot. Winds were forecast to be generally east-south-easterly below 5,000 ft above mean sea level, between 15–20 kt. There were no SIGMET[13] or AIRMET[14] warnings applicable to the flight.

An analysis prepared by the Bureau of Meteorology indicated a high-pressure system was situated to the south‑west of Western Australia, producing moderate east to south‑easterly winds in the vicinity of the accident site. Satellite imagery and measurements from the weather station at Cunderdin (about 80 km south of the accident site) indicated there was scattered cloud[15] in the vicinity of the site, with bases approximately 4,500–5,000 ft. There was the possibility of some thermal turbulence as the surface temperature and cloud base increased during the day, with a well-mixed air layer below the cloud. However, the analysis did not identify the existence of any hazardous weather phenomena in the vicinity of the accident site.

The air temperature at 2,500–3,000 ft was estimated to be about 10–12 °C, with a dewpoint[16] of about 6–8 °C. According to the Civil Aviation Safety Authority Carburettor icing probability chart, this temperature and dew point were within the ‘serious icing – any power’ envelope.

Pilot reports

Pilots of the other aircraft in the group reported good flying conditions during the flight back to Jandakot. Those reports were generally consistent with the Bureau of Meteorology’s analysis. There was slight variation in the pilots’ estimates of the cloud bases during their flights, ranging between 3,000–4,500 ft. Several pilots who departed Koorda that morning reported encountering some turbulence over the Cowcowing Lakes area. Most described the turbulence as mild, but one pilot reported an instance of ‘moderate’ turbulence,[17] an updraft of sufficient magnitude to startle their passenger and dislodge loose items in the cockpit. 

Recorded information

The helicopter was not fitted with a cockpit voice recorder, flight data recorder or cockpit camera, nor was it required to be. Neither the pilot nor the passenger’s mobile telephones were identified at the accident site.

Sources of data

A handheld GPS receiver and lanyard was recovered at the accident site and data was successfully downloaded. A review of the data found that during the accident flight, data was recorded at intervals varying between 1 and 19 seconds, with shorter intervals representing periods where the track and speed calculated by the GPS was rapidly changing. The information logged included GPS-calculated position, track, groundspeed, and altitude. 

While no mobile telephones were found, the passenger was using the OzRunways[18] application installed on their iPhone during the flight. The application was using the mobile telephone network to transmit data to the OzRunways’ servers every 5 seconds, which included the current position, track, groundspeed and truncated altitude in increments of 100 ft.[19]

Correlation of data

There was a strong correlation between the track log from the handheld GPS and the data transmitted by the iPhone to the OzRunways’ servers. The only significant discrepancies were during the initial climb with the handheld GPS-calculated altitude lagging behind the altitude data transmitted from the iPhone and during the last 10 seconds of recorded data (altitude, groundspeed, and position). Analysis of the last 10 seconds of data indicated several of the positions calculated by the handheld GPS required groundspeeds that exceeded the helicopter’s performance capabilities. During that period, any rapid changes in the satellite constellation used by the GPS could adversely affect the accuracy of the calculated position. The position calculated by the iPhone used additional sensors and GPS frequency bands, and those positions were more consistent with the helicopter following that trajectory during the descent.

Analysis of recorded data 

The elevation of Koorda airstrip was about 1,060 ft above mean sea level. The handheld GPS started recording position information at 1136:32 local time. At 1139:50, the track log of the GPS indicated that the helicopter was airborne and on initial climb. Soon after, the OzRunways application on the iPhone started transmitting data, also consistent with the helicopter being airborne.

There was a close correlation between the 2 tracks and groundspeeds. The helicopter’s airspeed during the initial stages of climb was estimated to be between 60 and 65 kt,[20] which was consistent with the helicopter manufacturer’s recommended climb speed. The estimated airspeed increased progressively towards about 80 kt during the latter stages of the climb. About 3 minutes 40 seconds after take-off, the iPhone and GPS altitude stabilised at about 2,700 ft (refer Appendix A for graphed data).

During the latter stages of the climb and the initial cruise, the pilot’s tracking of the helicopter towards Northam closely matched the track required[21] and any track error was generally less than 5°. This was similar to the tracking performance achieved by the pilot during the outbound flight on the previous day.

About 1 minute after reaching 2,700 ft, the tracking accuracy towards Northam began to reduce (measured as the calculated difference between the track required and track made good).[22] About 20 seconds later (1144:48), the altitude indicated by the iPhone had increased by a 100 ft increment to 2,800 ft (and the handheld GPS altitude also increased by a commensurate amount), with an estimated airspeed of 83 kt. 

About 45 seconds after reaching 2,800 ft and with an airspeed of about 76 kt (at 1145:33), the iPhone data indicated that the helicopter had departed abruptly from controlled flight, descending from 2,800 ft to 1,300 ft over a 10-second interval (a rate of descent of about 9,000 ft/min). The helicopter subsequently collided with terrain (at an elevation of 930 ft), about 6 minutes after becoming airborne at Koorda. 

Data transmitted by the iPhone (blue) and data from the handheld GPS (green) is depicted in Figure 5 and Figure 6.

Figure 5: Correlated track data for VH-RAS from Koorda to the accident site

Source: Google Earth, annotated by the ATSB 

Figure 6: iPhone track data, truncated altitude, and estimated airspeed for VH-RAS during last the 2 minutes of flight

Source: Google Earth, annotated by the ATSB 

Helicopter onboard camera

RHC introduced cockpit video cameras, which have been standard on new R66 and R44 helicopters since early 2021 and 2022 respectively and are an optional retrofit to in-service helicopters of both types. At the time of publication of this report, it remained optional for new R22 helicopters, with retrofit available to most in-service R22s. The forward-facing camera records video (encompassing a view through the windshield, pilot controls and the instrument panel), intercom audio, radio transmission and GPS data. RHC advised the recordings (up to 10 hours) can be used as a training tool, maintenance aid, or aerial-tour souvenir. 

The recording could also assist with occurrence investigations by allowing investigators to understand the circumstance/s that precede an accident, particularly when there are no survivors or witnesses. In turn, this aids the identification of important safety issues.

Wreckage and impact information

Wreckage distribution

The helicopter collided with terrain inverted, on a dry salt flat, on an easterly heading. A small distance below the salt crust was loose muddy sand and the water table was close to the surface. The main rotor head, with blades attached, and the top portion of the mast had separated but were located alongside the fuselage (Figure 7). One main rotor blade had fractured, with the outboard section located about 3 m from the main rotor assembly. The tailcone, including the tail rotor assembly, was attached to the fuselage. The horizontal and vertical stabiliser assembly had separated and was located about 6 m from the tailcone. The auxiliary fuel tank bladder was intact, however, the main tank bladder had ruptured due to impact forces. There was no fire.

Distribution of the wreckage and ground scars were consistent with an almost vertical descent and little to no rotation of the main rotor blades. Both occupants were found secured in their respective restraints, however, due to the high descent rate and inverted orientation, the impact was not considered survivable.

Figure 7: Accident site and wreckage

Source: ATSB

Wreckage examination

Detailed examination of the wreckage identified continuity of the flight and engine controls, with all fractures consistent with overstress failure, however, distortion to the fuselage precluded determining engine control position prior to impact. There was nil evidence of birdstrike found in the wreckage or the surrounding area. The examination further identified the following:

Powerplant

There was no evidence of a restriction or blockage to any part of the air induction system. Notably, the air intake SCEET hose[23] was in good condition with nil delamination and the air box filter was clear. The governor switch (located at the end of the right seat collective) was in the OFF selection. Although this may not necessarily reflect the position of the switch prior to the accident due to the damage sustained to the helicopter during the impact sequence. In addition, the governor circuit breaker was noted to be in the operational selection (not pulled due to failure or troubleshooting). 

The carburettor heat selector was fully down (no heat) and the carburettor heat assist lockout latch was unlocked. However, it was possible these positions were altered during the impact sequence. Therefore, the actual positions prior to the impact could not be determined. The air box carburettor heat slider was about mid travel, but impact forces and fuselage distortion pulled the control cable, preventing determination of the pre-impact position. Some scoring to the alternator fan backing plate and oil cooler was indicative that the engine may have been rotating at impact, although the damage was not consistent with the engine operating at full power.

Drivetrain

There was no evidence of pre-existing defects to the drive shafts, drive belts or sheaves. The drive belt tensioning clutch actuator had fractured in overstress, however, the actuator setting was consistent with normal operation and stretch of in-use belts. The main rotor transmission case had fractured open on the forward right side, but there was no evidence of failure of the transmission internal gears. The ATSB could not determine if the fracture was sustained during extreme teetering (refer to section titled Extreme teeter and mast bumping) or the impact.

Main rotor system

Damage to the main rotor head components (refer Figure 8 for main rotor head components detail) included:

• both main rotor pitch change links had fractured, in overstress, at the upper rod end thread 
• both spindle tusks had fractured and the tusks were not recovered
• the teeter stops had fractured in the centre due to severe impact forces from the spindles, the lower halves had been liberated and impact damage was visible on the mast. 

The mast had fractured near the swashplate, with no evidence of pre-existing fatigue. The fracture location was coincident with impact from a pitch horn that was free to rotate down and impact the mast, following pitch link failure. The fractured main rotor blade exhibited red paint transfer and damage consistent with it striking the forward fuselage with low energy and wrapping under the cabin.

Figure 8: Main rotor head and mast damage

Note: Diagram of typical R22 main rotor head showing VH-RAS fractured pitch links and corresponding rod end, teeter stop split in half and impact damage from mast bumping.

Source: ATSB and the United Kingdom Air Accidents Investigation Branch

Fuselage and cabin 

Both windshields had shattered, however, the bow (central pillar) was intact. The left seat quick‑disconnect dual flight controls were installed. It was reported both doors were fitted when the helicopter departed Koorda.[24] The left door or its components were not identified in the wreckage at the site, nor in an extensive search of the surrounding area. It was possible it may have been destroyed during the main rotor blade strike to the cabin, but with the fragments unable to be identified or located in the loose muddy sand that was below the salt crust. There was no evidence of main rotor blade strike to any portion of the tailcone.

Post on-site examination

The governor controller was sent to the helicopter manufacturer in the United States for testing, observed by the ATSB (via video link). However, impact damage prevented any functional testing. An internal visual inspection revealed damage consistent with impact forces and there was no evidence of electrical arcing or overheating of the circuit board components.

The ATSB collected a small amount of fuel from the gascolator and auxiliary tank. This was tested by an independent accredited laboratory. While the fuel recovered had been contaminated with ground water after the collision with terrain, the fuel content was consistent with 100 LL Avgas.

Most of the lamps for the warning and caution lights, including low rotor RPM, low fuel and engine governor-off, had been destroyed by impact forces. The only lamps available for testing (clutch, alternator, main rotor chip and main rotor temperature) were examined at the ATSB’s technical facilities in Canberra, Australian Capital Territory. The main rotor chip and temperature lamp filaments had fractured and there was no observable stretch.[25] Examination of the alternator[26] and clutch[27] lamps indicated they probably were illuminated at impact. However, it was unknown if these indications would have illuminated prior to, during, or as a result of the impact sequence. Therefore, the status of these lights prior to the impact was unable to be determined.

In addition, examination of the instruments identified possible contact transfer of material from the needle to the instrument face, at 50-55% engine and 58% rotor RPM on the dual tachometer. The normal operating range for the engine was 101% to 104%, with a maximum continuous RPM of 104%. The engine RPM had to be maintained within this tolerance for the main rotor system to provide effective lift. A possible sweep mark at 11‑15 inches of mercury was identified on the manifold pressure gauge, which was lower than would be expected in cruise conditions. This could be indicative of severe icing, however, it was also possible that the sweep between 11 and 15 was from disruption due to impact forces. There were no reliable marks identified on the carburettor air temperature gauge face or internal components.

In early November 2022, the engine was disassembled and examined at a Civil Aviation Safety Authority (CASA) authorised engine overhaul facility under the supervision of the ATSB. The engine condition was consistent with the engine’s recorded time-in-service since overhaul. No internal or external damage was identified that may have prevented the engine from operating normally prior to the accident. No defects were identified in the induction system components, core engine, or cylinder assemblies that may have affected its pre-accident operation. One magneto was operationally tested with positive results, however, the other could not be tested due to impact damage. Both magnetos were also internally examined, and resistance tested with nil defects identified. Further, the carburettor was bench tested and internally examined, with no issues identified.

In summary, examination of the helicopter’s flight and engine control systems, drive train and powerplant did not indicate any pre-existing defects that may have affected the control or normal operation of the helicopter.

Medical and pathological information

Post-mortem examinations

A full post-mortem examination was performed on the pilot and a limited examination conducted on the passenger.[28] Within the limits of this examination, the pilot’s post‑mortem did not identify any significant natural disease.

The post-mortems identified multiple fractures and injuries to both the pilot’s and passenger’s arms, hands, and legs. Research has previously been conducted into injuries sustained by control seat occupants during an accident and the extent to which those injures would be consistent with the occupant of the seat having their hands or feet on the controls. However, some research has found that the passenger and pilot may exhibit similar injuries if the passenger grasps a solid structure, such as bracing in anticipation of an impact (Cullen, 2004; Gradwell and Rainford, 2016). In addition, the inverted impact with terrain and limited protection provided by the lightweight airframe could have also contributed to the nature of the injuries sustained. Therefore, the evidence was inconclusive in determining whether the pilot or the passenger was manipulating the flight controls. Further, it could not be established if either had a medical event that prevented the other from being able to maintain control of the helicopter.

Toxicology testing conducted on pilot

Toxicology testing conducted as part of the pilot’s post-mortem examination detected the presence of paracetamol, propranolol, and quinidine/quinine. Carbon monoxide detected in the samples was less than 5%.[29]

Paracetamol is a commonly used over the counter analgesic, but generally without adverse side effects significant for its use in aviation. Propranolol, quinine and quinidine are prescription medications with some potential side effects and interactions.

A review of records from the Pharmaceutical Benefits Scheme, from June 2018, indicated that the pilot held prescriptions and a dispensing history for propranolol and temazepam, but none for either quinidine or quinine.

Quinidine and its naturally occurring stereoisomer (mirror image) quinine are natural alkaloids[30] and are individually synthesised for pharmaceutical and medical purposes. The toxicology testing technique could not differentiate between the 2 compounds. Quinidine was typically used to treat heart arrythmias and quinine, as a pharmaceutical medication, to treat malaria. Quinine is also used as a bittering agent in soft drinks such as bitter lemon or tonic water. The pilot used tonic water as a mixer with alcoholic drinks but was reported to have not consumed any during the days prior to the accident.

Propranolol is a prescription medication in the Beta-blocker class of drugs, that blocks the release of stress hormones, such as adrenaline. It can be prescribed for a variety of conditions that include migraines, benign heart palpitations and performance anxiety. The use of propranolol in an aeromedical context is discussed below (refer to section titled Civil Aviation Safety Authority review of propranolol use).

Consultant pharmacologist review

The ATSB engaged a consultant pharmacologist to review the results from the pilot’s toxicology testing. That review found the level of propranolol detected was unlikely to have been consumed during the period 24-hours prior to the accident, nor did they anticipate any adverse effects on the pilot’s ability to operate the helicopter during the accident flight.

They also concluded that if the quinine or quinidine source was from tonic water (ingested as quinine), it would have most likely have had to be consumed within the previous 24 hours and at the detected levels, they would not have expected any impairment of the pilot’s performance. If the source was quinidine (as a medication), it could have been consumed during the previous few days.

Further, the consultant pharmacologist found that the carbon monoxide levels (less than 5% saturation) in the non-preserved blood of the pilot was insignificant.

Civil Aviation Safety Authority review of propranolol use

Although propranolol was not a medication prohibited for use by pilots, the reason for its use could be of aeromedical significance and for that reason, required assessment by a designated aviation medical examiner (DAME) and/or CASA. Aeromedical impacts for its use included the blocking of the effect of adrenaline in maintaining blood pressure under g‑loading and to also reduce blood pressure at normal g-loadings but making the pilot more prone to light headedness from other causes such as dehydration.

Due to the potential side effects, CASA advised the ATSB that propranolol would not usually be approved for pilots who were likely to encounter high g-loadings[31] (high performance aircraft) or operations involving complex variations in g-loadings (push-pull effects in particular). In other cases, pilots may be approved to use propranolol if they could demonstrate their blood pressure was consistently stable.

Pilots intermittently using propranolol to treat symptoms such as performance anxiety are required to be assessed for other effects (such as low blood pressure and tiredness) and could be advised to not operate an aircraft within 24 hours of taking propranolol.

Additional medical considerations

The pilot had previously been diagnosed by their general practitioner as having a longstanding familial benign essential tremor, affecting their hand’s fine motor skills. To treat those symptoms, propranolol had been prescribed and was to be taken as needed. That tremor had also been noticed by many of the pilot’s flying friends and colleagues and was reported to be particularly evident when using hand tools such as screwdrivers, writing and pouring drinks. Flight examiners, instructors and passengers who had flown with the pilot advised that, although they were aware of the existence of the pilot’s tremor, they had not noticed it to affect the operation of the helicopter or aeroplanes being flown. 

The pilot completed an aviation medical with a CASA DAME every 2 years. That process required submission of an online applicant medical history questionnaire and then completion of a physical examination and tests with a DAME to assess the applicant’s ability to meet the relevant medical standard. This included assessing the applicant’s hand/eye coordination and checking for any symptoms or indications of neurological disease. 

The DAME had conducted the pilot’s last 3 medical renewals, with the last being conducted in January 2022, and did not recall the applicant having any tremor of clinical significance. Further, the pilot had not declared in their applicant medical history questionnaire they had symptoms or were receiving treatment of any movement disorder (including tremor), or that they had been prescribed any medications. Had a tremor of clinical significance been identified, the DAME reported that they would have referred the pilot to a neurologist for a specialist opinion. 

Extreme teeter and mast bumping 

Under certain specific flight conditions, semi-rigid rotor systems are susceptible to extreme teetering where the blades teeter beyond their normal operational range, resulting in what is commonly known as ‘mast bumping’. Mast bumping is the act of the inboard end of the blade (the spindle) or main rotor hub contacting the main rotor shaft. In R22 helicopters, this can generally be identified by extensive damage to the teeter stops and varying degrees of structural damage to, and possible fracture of, the mast. 

Once the teeter stops are damaged or fractured, both spindle tusks may also fracture. This allows the main rotor blades to flap even further resulting in excessive bending loads to the main rotor pitch change links and subsequent fracture of one or both links. Failure of a pitch link can result in the associated blade reacting to the aerodynamic and centripetal forces on the blade and rotating about its pitch axis, making the helicopter uncontrollable. 

As documented in many investigation reports worldwide, scenarios that have been linked to mast bumping include low-g and/or low rotor RPM/rotor stall, in conjunction with delayed and/or inappropriate flight control inputs.[32]

The low-g condition

‘G’ or ‘g’ is an abbreviation for the acceleration due to the earth’s gravity. Positive ‘g’ is necessary for helicopters to respond to pilot control inputs. In a low-g condition (that is, approaching the feeling of weightlessness),[33]^,[34] the pilot’s ability to control the attitude of the helicopter is greatly reduced. 

The United States Federal Aviation Administration (FAA) Helicopter Flying Handbook further described low-g effects, including:

Helicopters rely on positive G to provide much or all of their response to pilot control inputs. The pilot uses the cyclic to tilt the rotor disk, and, at one G, the rotor is producing thrust equal to aircraft weight. The tilting of the thrust vector provides a moment about the center of gravity to pitch or roll the fuselage. In a low-G condition, the thrust and consequently the control authority are greatly reduced.

Although their control ability is reduced, multi-bladed (three or more blades) helicopters can generate some moment about the fuselage independent of thrust due to the rotor hub design with the blade attachment offset from the center of rotation. However, helicopters with two-bladed teetering rotors rely entirely on the tilt of the thrust vector for control. Therefore, low-G conditions can be catastrophic for two-bladed helicopters.

… (when the helicopter) enters a low-G condition. Thrust is reduced, and the pilot has lost control of fuselage attitude but may not immediately realize it. Tail rotor thrust or other aerodynamic factors will often induce a roll. The pilot still has control of the rotor disk, and may instinctively try to correct the roll, but the fuselage does not respond due to the lack of thrust. If the fuselage is rolling right, and the pilot puts in left cyclic to correct, the combination of fuselage angle to the right and rotor disk angle to the left becomes quite large and may exceed the clearances built into the rotor hub. This results in the hub contacting the rotor mast, which is known as mast bumping.

RHC stated that extreme teetering and subsequent mast bumping can result from the pilot attempting to recover from an uncommanded right roll while in a low-g condition. Extensive investigation has shown that inappropriate recovery flight control inputs often resulted in the main rotor blade impacting the forward fuselage.

Low rotor RPM and stall

The R22 helicopter, with its low rotor system mass and relatively high RPM, is described as ‘low inertia’. In low inertia systems, rotor RPM is gained and lost very easily. Low rotor RPM occurs when the rotor can no longer produce enough lift to support the weight of the helicopter, and it will start to descend. If this situation is not quickly and effectively managed, the rotor RPM can reduce to a point where one, or both, main rotor blade/s stall. According to RHC safety notice[35] SN-24 Low RPM rotor stall can be fatal, rotor stall recovery is ‘virtually impossible’.

Low-rotor RPM can occur at almost any time, during power-on and power-off operations and is usually the result of improperly coordinating the collective and throttle, including overpitching or a failure to quickly lower the collective in an emergency such as engine failure or power reduction.

RHC safety notice SN-10 Fatal accidents caused by low rotor RPM rotor stall included:

A primary cause of fatal accidents in light helicopters is failure to maintain rotor RPM. To avoid this, every pilot must have his reflexes conditioned so he will instantly add throttle and lower collective to maintain RPM in any emergency. 

…

Power available from the engine is directly proportional to RPM. If the RPM drops 10%, there is 10% less power. With less power, the helicopter will start to settle, and if the collective is raised to stop it from settling, the RPM will be pulled down even lower, causing the ship to settle even faster. If the pilot not only fails to lower collective, but instead pulls up on the collective to keep the ship from going down, the rotor will stall almost immediately. When it stalls, the blades will either "blow back" and cut off the tailcone or it will just stop flying, allowing the helicopter to fall at an extreme rate. In either case, the resulting crash is likely to be fatal.

No matter what causes the low rotor RPM, the pilot must first roll on throttle and lower the collective simultaneously to recover RPM before investigating the problem. It must be a conditioned reflex. In forward flight, applying aft cyclic to bleed off airspeed will also help recover lost RPM.

The low rotor RPM warning lamp and horn will activate when the rotor RPM reduces to 97% or below. The warning lamp is located on the top of the instrument panel and the horn can be heard in the cabin and in both headsets. The POH stated that ‘catastrophic rotor stall could occur if the rotor RPM ever drops below 80% plus 1% per 1,000 ft of altitude’. Further, the FAA Helicopter Flying Handbook stated that ‘low inertia rotor systems can become unrecoverable in 2 seconds or less if the RPM is not regained immediately’.

Potential factors leading to low-g and/or low rotor RPM/rotor stall

As a possible explanation for delayed or inappropriate control inputs that may have preceded a low-g condition or low rotor RPM and stall, the investigation explored the following scenarios:

Low-g pushover 

RHC safety notice SN-11 Low-g pushovers – extremely dangerous stated ‘pushing the cyclic forward following a pull-up or rapid climb, or even from level flight, produces a low-g (weightless) condition’. It also stated that severe in-flight mast bumping usually results in main rotor shaft separation and/or rotor blade contact with the fuselage. The following warning was included:

Never attempt to demonstrate or experiment with low-G manoeuvres, regardless of your skill or experience level. Even highly experienced test pilots have been killed investigating the low-G flight condition. Always use great care to avoid any manoeuvre which could result in a low-G condition. Low-G mast bumping accidents are almost always fatal.

The ATSB spoke with the pilot’s regular helicopter flight instructor and other pilots who routinely flew with the pilot in VH-RAS. They all reported that the pilot demonstrated awareness of low-g flight conditions and the danger in entering a pushover manoeuvre, whether deliberate or unintentional. It was therefore determined to be unlikely that the pilot initiated a deliberate low-g pushover.

Turbulence

The FAA Helicopter Flying Handbook stated that ‘turbulence, especially severe downdrafts, can also cause a low-g condition and, when combined with high airspeed, may lead to mast bumping’. RHC safety notice SN-32 referred to flying in high winds or turbulence,[36] firstly stating that it should be avoided. It continued, noting that ‘a pilot’s improper application of control inputs in response to turbulence can increase the likelihood of a mast bumping accident’. In addition, in March 2024, the ATSB published the safety advisory notice Anticipate turbulence and slow down, which included:

Awareness of conditions likely to produce turbulence, and slowing down prior to encountering turbulence, could increase the time available to recognise and respond to a low-g condition in Robinson Helicopters.

Several pilots who departed Koorda that morning reported encountering some turbulence over the Cowcowing Lakes area.[37] RHC recommended a reduction in airspeed when encountering turbulence. The recorded data for VH-RAS did not indicate any significant deviation to altitude or airspeed typically seen in moderate, severe or extreme turbulence. In addition, the data did not indicate any significant reduction in airspeed in the last minutes of the flight, as recommended by RHC when encountering significant turbulence.[38]^,[39] 

RHC also advised helicopters that are lightly loaded may be more susceptible to turbulence than heavy helicopters. As the weight of VH-RAS was calculated to be at, or just above, its maximum weight at the time of the accident, this potentially reduced any effects of turbulence that may have been encountered.

Overpitching

As the collective is raised, there is a simultaneous and equal increase in pitch angle of both main rotor blades. An increase in pitch angle also results in increased drag on the main rotor blades. To counter this adverse effect, the R22 has a throttle correlator mechanism attached to the collective control that increases the engine’s throttle when the collective is raised.

Overpitching is a condition that happens when the collective pitch is increased to a point where the angle of attack of the main rotor blades creates extra drag and maximum engine power cannot maintain or restore normal operation rotor RPM. One typical scenario often occurs during take-off, where the pilot raises the collective lever to a point beyond the full throttle position (where full throttle was required to maintain RPM), then there will be more power required by the rotors than power available from the engine, resulting in a rotor RPM decay. However, as the helicopter had departed Koorda without incident, this was considered unlikely for this occurrence.

Overpitching can also occur if the pilot raises the collective lever at a rate that is faster than the correlator will open the throttle (to avoid a bird for example), while not compensating for the increased drag by manually increasing the throttle. In this case, the rotor RPM may rapidly decay to a level that is too low for the engine power available to recover.

In addition, overpitching is more likely to occur at a high weight and/or high altitude, where the rotor blades are already operating at larger pitch angles.

Any application of collective to arrest the descent further increases rotor drag and reduces rotor RPM. The situation can rapidly deteriorate resulting in the rotor blades effectively stalling and significantly reduced lift. 

Potential for negative transfer of learning in an emergency situation

The pilot had considerable aeroplane operational experience, including flight activity endorsements to conduct aerobatics, spinning and formation flying. The investigation considered if the pilot may have reacted to an unexpected situation with an inadvertent ‘aeroplane’ control input, as described in SN-29 Airline pilots high risk when flying helicopters, which could have induced a low-g scenario. Specifically, the notice stated that:

…The airplane pilot may fly the helicopter well when doing normal maneuvers under ordinary conditions when there is time to think about the proper control response. But when required to react suddenly under unexpected circumstances, he may revert to his airplane reactions and commit a fatal error. Under those conditions, his hands and feet move purely by reaction without conscious thought. Those reactions may well be based on his greater experience, ie., the reactions developed flying airplanes.

Records indicated the majority of the pilot’s recent flying had been conducted in VH-RAS.[40] In addition, people who had flown in VH-RAS reported the pilot to being cognisant of the dangers of inappropriate control inputs when operating an R22 helicopter. Therefore, while a scenario where the pilot inadvertently used aeroplane flight control actions was considered a possibility, it was assessed to be unlikely in this occurrence given the pilot’s helicopter experience.

Inadvertent or inappropriate dual control input 

Several sources reported that, on occasion, the pilot had offered passengers with flying experience the opportunity to ‘fly’ the helicopter using the left seat dual flight controls. Some passengers reported that they had flown the helicopter using the cyclic control, while the pilot retained control of the collective. One of those passengers had remarked that the pilot had been closely supervising them and had cautioned them about making abrupt forward control inputs. Another passenger recalled that they had been offered but decided to decline the opportunity to fly the helicopter. In this case, the pilot had emphasised the sensitivity of the flight controls and highlighted the hazards with making abrupt control inputs. 

RHC safety notice SN-20 Beware of demonstration or initial training flights noted that, ‘a disproportionate number of fatal and non-fatal accidents occur during demonstration or initial training flights’. Further, if a ‘student’ was to make sudden large control movements in the wrong direction, an experienced instructor may not necessarily be able to recover control of the helicopter. The notice also highlighted the importance of thoroughly explaining to students the ‘extreme sensitivity of the controls in a light helicopter’. 

RHC safety notice SN-44 Carrying passengers also identified that carrying a passenger can potentially increase risk as they add workload and distraction. The notice specifically stated:

• Always remove passenger-side controls. 
• Caution passengers against inadvertently bumping the cyclic center post.

Another potential source of confusion for someone inexperienced in flying helicopters is the throttle operation. The throttle twist control, located on the end of each collective, is operated by rolling the hand away from the body (or outboard) to increase engine power, and rolling toward the body to decrease. This is opposite to throttle controls fitted to vehicles such as motorcycles and the steering arm on boat outboard motors.

Despite the above, without any onboard recording devices such as a cockpit camera, the ATSB was unable to determine if the passenger was in control or if they had inadvertently bumped or moved the controls. 

Other scenarios

The investigation considered other scenarios that may induce a large control input, such as avoiding a bird or a door opening or separating in-flight. Bird avoidance may have resulted in an abrupt control movement, however, nil evidence of a strike was found in and around the wreckage. Further, one of the pilots in the group and a local resident stated that they did not observe any bird activity in the area. 

If a door had opened, this had potential to temporarily distract the pilot and/or the passenger, or result in an inadvertent control input while trying to close the door. [41] Noting that the left door could not be identified at the accident site, operating the helicopter with one or both doors removed is permitted, with the manufacturer advising to calculate weight and balance as required[42] and ensure loose articles are secured in the cabin. Liberation of the left door had the potential to strike the tail cone or tail rotor assembly, however, there was no evidence of the door coming into contact with any portion of the tailcone.

While these scenarios remained a possibility, there was insufficient evidence to conclude probability.

Engine power loss scenarios 

While the governor switch was found in the OFF selection, the position prior to impact could not be determined and post-accident functionality testing could not be performed due to the damage sustained. However, it remained a possibility for an intermittent issue (for example a cylinder misfire, fouled spark plug or sticky valve), to briefly affect engine operation, which did not result in any visible damage. An intermittent loss of engine power will typically result in a ‘kick’ (nose left yaw). It was possible that a sudden yaw may have contributed to either the pilot or passenger making a large control input, but there was insufficient evidence to conclude probability.

There have also been cases where an engine stoppage was inadvertently induced by rolling off the throttle too fast or the mixture control being pulled instead of the carburettor heat or another control.[43] 

Acknowledging the limitations of the Civil Aviation Safety Authority Carburettor icing probability chart and variations to induction systems depending on engine installation, the investigation also considered the possibility of carburettor icing. 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 airflow to the engine. This is more likely to occur at low engine power settings due to the added cooling effect of the partially-closed throttle butterfly.[44]^,[45] 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. The effect of carburettor icing includes reduced power output, rough running and in some cases engine failure. Further, the risk of ice build-up in the carburettor can be high even with no visible moisture and at temperatures of up to 38 °C. 

The RHC safety notice SN-25 Carburetor Ice also stated that, ‘avoidable accidents have been attributed to engine stoppage due to carburetor ice. When used properly, the carburetor heat and carb heat assist systems on the R22 and R44 will prevent carburetor ice’. The European Union Aviation Safety Agency safety information bulletin No: 2010-03 Carburetor Icing Prevention, also noted that:

In a helicopter, icing can develop quite insidiously and the effect on the engine is less obvious. Furthermore, should the engine stop, an immediate entry to autorotation is necessary to prevent catastrophic reduction of rotor RPM, and the descent rate, usually around 2 000 feet per minute, is such that there is rarely time to attempt a restart. 

RHC safety notice SN-31 Governor can mask carb ice stated that, when ice begins to form in the carburettor, a properly functioning governor will increase throttle to maintain engine RPM, which will also result in constant manifold pressure. Once the governor has opened the throttle completely it can no longer maintain rotor RPM and the engine RPM will reduce, which may be the first point a pilot is aware of the icing, if they had not been monitoring the manifold pressure and carburettor air temperature gauges.[46]^,[47] Further, application of heat at this late stage will melt any ice, which could result in engine stoppage.

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. Also, the examination could not determine if carburettor heat was in use prior to impact nor were there any reliable needle impact marks identified on the carburettor air temperature gauge. Therefore, while local conditions were conducive to the possibility of the formation of carburettor ice, it was not possible to establish if VH-RAS was affected by carburettor icing at any stage of the flight.

Overall, the ATSB’s examination of the engine did not identify any issue that may have affected operation and noting that the engine was likely rotating at impact, may suggest that an engine stoppage had not occurred. However, the investigation could not conclude an intermittent engine issue and/or that carburettor icing occurred, resulting in a reduction in power and the requirement to enter autorotation. Equally, the inverted portion of the descent had the potential to interrupt fuel flow to the engine via the gravity‑fed fuel system and reduce engine power.

In the event of an engine power failure above 500 ft above ground level, the emergency procedures section of the POH stated to:

• Lower collective immediately to maintain rotor RPM
• Establish a steady glide at approximately 65 KIAS (knots indicated airspeed)
• Adjust collective to keep RPM between 97 and 100% or apply full collective down if light weight prevents attaining above 97%.

Airspeed and descent rates may be variable during controlled descent. Descent rates during autorotation vary depending on helicopter configuration and local conditions, however, anywhere between 1,400 to 1,900 ft/min can be expected.[48]

However, using the recorded data, the investigation estimated that the helicopter collided with terrain between 10 to 15 seconds after departing from controlled flight. The descent rate of about 9,000 ft/min during this period was inconsistent with the helicopter being established in an autorotative descent. 

Low-g or low rotor RPM signatures

The ATSB liaised with RHC and reviewed investigation reports from the United Kingdom, United States, Europe, and New Zealand. Excessive teeter and mast bumping, such as fractured tusks and teeter stops, were associated with a combination of these scenarios. 

A sudden roll to the right is often associated with a low-g event. Typical low-g wreckage signatures include a main rotor blade strike to the cabin/forward fuselage (where the blade/s cuts through the helicopter at high velocity) and mast separation (often at the mast bump/teeter stop location). In addition, a long/large debris trail is indicative of in‑flight break‑up initiating when the helicopter was within normal operating parameters.

Low rotor RPM and rotor stall signatures identified in many investigation reports included a rotor strike to the tailcone, main rotor blade coning,[49] spindle impact marks on the main rotor head, little to no rotational ground marks, and the helicopter nosing over and a vertical descent with a small debris field. However, it has been documented that, since RHC changed the main rotor blade construction from stainless steel skin to aluminium skin, damage to blade skins due to extreme coning as a result of low rotor RPM is now less than previously observed.[50]

A rotor strike to a solid object, such as the fuselage, a tree, or the ground, can stop a low powered engine. This is in contrast to where rotor blades often fracture into multiple small pieces during a rotor strike with full engine power. 

Similar occurrences

A review of investigation reports from Australia, New Zealand, United Kingdom, United Sates, South Africa, and Europe identified at least 3 reports that had signatures very similar to this occurrence. These are summarised below.

German Federal Bureau of Aircraft Accident Investigation BFU21-0949-3X  

On 17 October 2021, recorded data from a Robinson R44 Raven II showed that, while in cruise about 1,900 ft above mean sea level, the helicopter abruptly climbed about 200 ft, followed almost immediately by a rapid descent. Coincident to this, the recorded ground speed rapidly decayed from about 100 kt to zero. The pilot and 2 passengers were fatally injured.

The mast had fractured below the swashplate and the rotor head, with the blades attached, came to rest on top of the upright destroyed fuselage. Typical mast bump and in‑flight break‑up signatures were found on the mast and rotor head. One main rotor blade showed damage consistent with it striking the forward fuselage. In addition, the left seat (passenger) dual flight controls were installed.

The investigation concluded the helicopter likely entered a low-g condition that led to mast bumping ‘due to erroneous control inputs, which resulted in the rotor hitting the fuselage and in‑flight break-up’. Of note, the Bureau indicated that it was quite possible that the passenger was allowed to control the helicopter during the flight and unknowingly initiated a pull‑up or push over. However, it could not be established with any certainty who was in control of the helicopter at the time of the occurrence.

ATSB investigation AO-2020-061

On the afternoon of 2 December 2020, a RHC R44 Raven I, registered VH-HGU, departed Goulburn Airport, New South Wales with a student pilot and instructor on board. The helicopter flew east, and the last recorded automatic dependent surveillance broadcast detected it descending into a valley in the Bungonia State Conservation Area. A search commenced when the helicopter did not return as expected, and the wreckage of VH-HGU was found in a valley, approximately 4 km north-west of its last ADS-B transmission. Both pilots were fatally injured, and the helicopter was destroyed.

Although fire and impact damage had destroyed some parts of the helicopter, the evidence available gave no indication that the helicopter was operating abnormally prior to an in‑flight break-up. Components recovered near the beginning of the 275 m wreckage trail indicated that a main rotor blade had struck the left side of the fuselage at the beginning of the break‑up sequence. No evidence was found to indicate pre-existing mechanical defects or issues that could have prevented normal engine operation. While a transient condition such as a partial or complete power loss could not be ruled out, such an event should not have resulted in an in-flight break-up.

The investigation concluded that, while flying in the vicinity of the valley, the helicopter entered a low-g condition due to turbulence, inappropriate control inputs, or a combination of both. This condition, probably in combination with inappropriate recovery control inputs resulted in extreme teetering of the main rotor. A mast bump occurred as a result, and the helicopter subsequently broke up in-flight.

United Kingdom Air Accidents Investigation Branch G‑CHZN

On 6 January 2012, after about 1 hr and 28 minutes into a private flight, and when about 1,440 ft above the ground, the Robinson R22 disappeared from air traffic control radar. Witnesses reported hearing a pop sound and some described the helicopter rolling to the left, with one person saying the helicopter pitched up prior to the left roll. The helicopter then fell, inverted, to the ground. The pilot, the only person on board, sustained fatal injuries.

Both main rotor blades had separated from the hub. There was evidence of rotor blade strike to the forward left fuselage. The tailcone remained attached to the fuselage and there was no evidence of rotor strike. An engine examination did not identify any issue or failure that would prevent it from operating normally. The local conditions at the time were calculated to be conducive to moderate or serious carburettor icing at any engine power.

The investigation explored low-g and low rotor RPM, with the wreckage exhibiting signatures from both scenarios. Pilot incapacitation, carburettor icing and avoiding a bird were also considered as possible contributing events. The report concluded that the mast bumping ‘was probably caused by a loss of rotor RPM (not followed by rapid lowering of the collective), a low-g pushover, a large abrupt control input – or a combination thereof’. Further, it was noted that low-g or a large abrupt control input ‘could have been generated for a number of reasons, and the light control forces in the R22 make it relatively easy to enter such conditions’.

Safety analysis

Executive summary

What happened

On the morning of 19 September 2022, the pilot of an Air Tractor AT-502B aircraft, registered VH‑KDR and operated by Smart Air Services, was completing aerial spray operations near Chinchilla, Queensland. After becoming concerned at not being able to contact the pilot, the spray company alerted nearby farmers to assist, and the aircraft was subsequently found in the paddock being sprayed. The pilot was fatally injured and the aircraft was destroyed.

What the ATSB found

The ATSB found that, while the pilot was conducting a spray run parallel to the paddock fence line at about 8 ft above the ground, a large bird (Australian bustard) struck the right Perspex windshield. During the impact, the right windshield was shattered, with the bird carcass found inside the aircraft.

The site examination showed that, after the birdstrike, the aircraft continued flying for a further 310 m, or about 5 seconds. However, without any flight data, it could not be exactly determined what occurred during this time. Despite this, the ATSB established that, it was likely that the birdstrike affected the pilot’s ability to control the aircraft, resulting in a collision with terrain.

Safety message

Birdstrikes are relatively common with over 1,300 occurrences reported to the ATSB in 2022. While these strikes typically result in nil-to-minor damage to an aircraft and nil-to-minor injuries to occupants, there have now been 2 confirmed fatal accidents in Australia. This accident highlights the ongoing hazards of birdstrikes to aircraft, particularly during low-level spraying operations.

Where applicable, there are published recommended practices available from the Australian Aviation Wildlife Hazard Group (2014) to manage risk of wildlife strikes for organisations that may be involved in a wildlife strike occurrence. Practices include considering the likelihood of a wildlife strike and proximity of known wildlife to aircraft manoeuvring areas and flight paths, as well as identifying wildlife characteristics such as their agility, speed, manoeuvrability, and their ability to avoid aircraft. Other considerations include the consequence of a strike, including the effects of the mass of the wildlife species, the velocity of the aircraft involved, the resultant impact force(s) and the damage that could result; and if adequate control of the aircraft can be maintained following a strike event. If a birdstrike does occur, available protections for a pilot include the wearing of flight helmets, 4‑point seatbelts, and the fitment of thicker aircraft windshields, where possible.

The investigation

The occurrence

On the morning of 19 September 2022, the pilot of an Air Tractor AT-502B aircraft, registered VH‑KDR and operated by Smart Air Services, was conducting aerial spray operations on a property near Chinchilla, Queensland. The operations base for the day was at a private airstrip south-east of Chinchilla, where the aircraft was to be loaded with fungicide, pesticide, and fuel. The 2 farms to be sprayed were located several kilometres to the north of the airstrip (Figure 1).

Figure 1: Location of the airstrip and area of operations

Source: Google Earth, annotated by the ATSB

Numerous spray runs were completed on the paddocks during the morning between 0704 and 1012 local time, with the pilot returning to the airstrip on 4 occasions to reload the spray chemicals and to refuel the aircraft. Recorded GPS data from the aircraft’s Satloc[1] system identified that, at 1058, the pilot departed the airstrip to continue herbicide spraying on the northerly paddock. By 1125, the pilot had completed 17 spray runs within the paddock in a racetrack pattern (Figure 2).

At the conclusion of run 17, the aircraft was turned onto a northerly heading, flown parallel to the fence line, before being turned onto a reciprocal heading to spray this area (Figure 2). The final GPS data point recorded by the Satloc system was at 11:25:41, 3 seconds after the pilot had commenced the southerly fence line spray run.

After 1200, the loader at the airstrip attempted to phone the pilot to ask whether more fuel was required, however, no response was received. The loader was concerned at not being able to contact the pilot, and phoned the company operations manager, who then contacted nearby farmers to assist with locating the aircraft. At about 1215, a local farmer found the aircraft wreckage in the paddock where the pilot had been spraying. The pilot was fatally injured and the aircraft was destroyed.

Figure 2: Accident site and spray runs (orange)

Source: SatLoc data provided by the aircraft operator overlaid on Google Earth, annotated by the ATSB

Context

Pilot information

The pilot held a valid Class 1 Aviation Medical Certificate and a Commercial Pilot Licence (Aeroplane). The pilot also held aerial application and low-level ratings, as well as tailwheel and retractable undercarriage, manual propellor pitch control, and turbine engine endorsements.

The pilot had accumulated 3,738 hours of aeronautical experience, of which 2,681 hours was in turbine agricultural aircraft.

Aircraft information

General information

VH-KDR was an Air Tractor Incorporated 502B single-seat low-wing tailwheel aircraft manufactured in the United States in 1996. It had a total wingspan of 15.8 m and was powered by a Pratt & Whitney Canada PT6A‑34AG turboprop engine. It was first registered in Australia on 6 August 2004. It was issued with a Special Certificate of Airworthiness in therestricted category[2] on 10 August 2004. A maintenance release was issued on 9 September 2022.

Windshield information

The Air Tractor 502B windshield is constructed of 3 pieces being either glass or Perspex (Figure 3). The centre windshield is manufactured from glass. The thickness of the glass on VH‑KDR was 1/4-inch, which was the standard thickness when the aircraft was manufactured. The 2 side pieces were manufactured from Perspex.

Figure 3: Air Tractor AT-502B windshield

Source: Storm Aeronautics

The ATSB contacted Air Tractor regarding the availability of thicker side windshields and they advised that there had been no consideration for using thicker material. The manufacturer stated that the windshield design on the Air Tractor is standard for the agricultural aviation industry, and in previously reported birdstrikes where the bird had entered the aircraft through either side windscreen, the bird missed the pilot and hit the back wall. Further, they described difficulties in replacing the Perspex windshield with glass due to the curvature of the design, which would increase the likelihood of glare and would add weight.

In 2009, Air Tractor issued a service letter advising that thicker centre windshields were being installed in current production aircraft. The thicker windshields were 3/8-inch thick glass compared to the original 1/4-inch. The letter also advised that the thicker windshields could be installed in existing aircraft, though not mandatory. This letter was issued after operators had reported on birdstrikes that had resulted in the bird carcass entering the cockpit through the centre windshield (refer to section titled Related occurrences).

Meteorological information

Based on information recorded from a private weather station located in the paddock being sprayed, the air temperature around the time of the accident was 26.3 ^°C, and the average wind speed was about 9 kt from the west-south-west. The calculated crosswind component was about 8 kt.

Wreckage and impact information

The ATSB’s on-site examination found that the aircraft had impacted the ground, adjacent to the fence line. The aircraft’s orientation was consistent with the direction of travel. The left wing had initially impacted the ground, followed by the fuselage, in a near vertical attitude. The propeller and engine were buried in the soft earth, with the spread of wreckage contained to a small area. Ground scars and damage to the left wing indicated that the wing struck the ground at about 30° to the horizontal (Figure 4). The length of the ground scar from the point of impact to the fuselage was about 12.6 m.

Figure 4: Aircraft wreckage at the accident site

Source: ATSB

Examination of the wreckage identified that:

• Inspection of the propeller and engine indicated the engine was delivering power at the time of the impact.
• Fuel and chemical product residues were also found at the accident site.
• There was no post-impact fire.
• Due to the extent of damage to the cockpit in the accident sequence, it was not possible to determine the position of the engine controls prior to the impact. However, where possible, control continuity was established.
• It was very likely the aircraft was trimmed ‘nose-up,’[3] which was reported by the operator to be the configuration[4] typically used when conducting low-level aerial spraying.
• There were no pre-existing technical failures with the aircraft that would have contributed to the accident.

A large bird carcass was found in the cockpit and severed bird wings were located about 310 m north of the wreckage, in-line with the aircraft’s track along the fence line, which indicated the approximate location of the initial birdstrike (Figure 5). Bird feathers were found along flight path in between location of wings and aircraft wreckage. Biological residue from the bird was found outside the right cockpit window and among the pieces of Perspex from that window. Further biological residue was found on the left side of the cockpit behind the pilot’s seat, indicating the bird passed through the right cockpit window in an upward trajectory towards the left side. There was no evidence of the bird striking any other part of the aircraft.

Figure 5: Site map with key locations

Source: Google Earth, annotated by the ATSB

Recorded data

The SatLoc data’s final recorded point was about 800 m north of the wreckage, and 500 m from the location of the severed bird wings (Figure 2). Based on the aircraft's previously recorded ground speed and direction of travel, there were about 5 seconds between the birdstrike occurring and the impact with the ground. The ground speed of the aircraft was about 116 kt (200 km/h), which was a similar speed to other runs, and the height was about 8 ft above ground level,[5] which was a similar height to the previous runs.

Medical and pathological information

The post-mortem report indicated that the pilot did not have any pre-existing medical conditions nor consumed any substances likely to have contributed to the accident. The report indicated that the pilot sustained significant head injuries in the accident. It was also noted that feathers were found on the pilot’s neck and chest, but it could not be determined whether they were the direct result of the birdstrike or from the accident sequence. Therefore, it was unknown whether these injuries sustained were due to the birdstrike or collision with terrain, or both.

Bird identification

Recovered biological specimens of the bird, including wing feathers and residue from the carcass, were analysed by the Australian Centre for Wildlife Genomics at the Australian Museum. The specimens were identified as a Ardeotis australis, commonly known as an Australian bustard (or bush/wild/plains turkey) (Australian Museum 2020).

The Australian bustard is a large bird, 80 to 120 cm in length and a wingspan of up to 230 cm (Ziembicki 2010). An average adult bird weighs 4.5 kg, however, males can weigh up to 8 kg (Bird Fact 2022). Australian bustards are capable of flying but are mostly ground dwelling. They are Australia’s heaviest flying bird. The bird carcass recovered had a wingspan of at least 120 cm and a length of 100 cm; the gender was unknown.

Survival aspects

First responders to the accident site identified that the pilot had been fastened into the seat by the aircraft’s 4‑point restraint harness. A flight helmet was found within the wreckage, which the pilot had been wearing, but it had become dislodged and sustained damage to the right-side shell and left‑side earpiece during the accident sequence. The left-side helmet strap had separated from the earpiece. Yellow paint transfer from the rear of the cockpit was found on the back of the helmet. Bird biological residue was found on the inner surface of the helmet visor.

Severe compression of the forward fuselage and cockpit region had occurred as a result ground impact forces. The accident was not considered survivable.

Related occurrences

Birdstrikes are a recognised hazard in aviation and there are mitigators in place around airports, however, there are challenges when operating outside of these areas. A review of Australian and international data and investigation reports found the following occurrences involving birdstrikes during aerial spraying operations.

Australian data

A review of the ATSB occurrence database from 2000 to 2022 found that 30 birdstrike occurrences were reported during aerial spraying operations. Of those, one resulted in minor injury, 20 resulted in minor damage, and one resulted in substantial damage. For example:

• On 4 February 2000, the pilot of an Air Tractor 502B was conducting aerial spraying operations 15 km west of Moree, New South Wales. As the aircraft approached the start of a spray run at approximately 50 ft above ground level, a large bird struck the front windshield. The pilot sustained minor injuries and the aircraft sustained minor damage.
• On 9 November 2011, an Air Tractor 802A struck an Australian bustard while conducting aerial spraying operations, 37 km south-east of Emerald Airport, Queensland. The impact damaged the wing, resulting in a loss of control and the aircraft sustained substantial damage. The pilot was uninjured.

In 2022, there were 2 fatal accidents involving birdstrikes, including this accident. The other accident was a private helicopter flight, investigated as AO-2022-034. This was the only fatal birdstrike confirmed in Australia.

United States data

A review of the United States Federal Aviation Administration wildlife strike database identified 2 birdstrikes resulting in damage to the windshield while conducting aerial spraying operations:

• On 15 April 2000, the pilot of an Air Tractor 502B reported that a red-tailed hawk struck and damaged the windshield during a spray run near Prescott, Washington, resulting in minor injuries.
• On 9 June 2017, the pilot of an Air Tractor 502B reported a birdstrike occurring while conducting spraying operations about 10 miles west of Salt Lake City Airport, Utah. The birdstrike damaged the right front window, which resulted in the door opening. The pilot was uninjured and conducted a precautionary landing. The bird species was unknown.

In addition, Air Tractor reported another 3 instances involving birdstrikes resulting in minor windshield damage and nil injuries:

• On 30 October 2015, an Air Tractor AT-402B struck a bird during mosquito control activity.
• On 24 August 2020, while returning from a drop during firebombing operations, the windshield of an Air Tractor AT-802A broke due to a birdstrike.
• On 24 October 2021, an ibis entered the cockpit through the windshield of an Air Tractor AT‑802.

South African data

On 11 September 2008, the pilot of an Air Tractor 502B was conducting aerial spraying operations on a farm in the Caledon region of South Africa. During a spray run, a witness on the ground, who was approximately 700 m from the aircraft heard an aircraft noise and immediately after, an explosion. The aircraft wreckage was located in a field in an inverted position. The pilot was fatally injured.

Bird remains and feathers were identified inside the aircraft cockpit. A piece of glass windshield was also found within the cockpit that was covered in bird remains. The carcass of a blue crane was found near the wreckage. The investigation was unable to determine if the pilot was incapacitated or fatally injured as a result of the collision with the bird before the impact with the ground.

As a result of the investigation, the South African Civil Aviation Authority (investigation CA 18/2/3/8548) made the following recommendation:

Request the Air Tractor manufacturer to investigate the possibility of replacing the existing windshield with a thicker, stronger windshield to be used by clients operating these aircraft in areas of high bird activity.

The following year, Air Tractor published a service letter with details of a new, thicker windshield available to be fitted (refer to section titled Windshield information).

Safety analysis

Birdstrike and resulting collision with terrain

The location and impact point of the aircraft wreckage was consistent with the previously recorded flight path. Further, the presence of the Australian bustard carcass inside the aircraft wreckage, and the identification of the bird’s wings around 310 m north of the wreckage, indicated that the aircraft had experienced a birdstrike while flying the southbound spraying track at about 8 ft above ground level.

Considering the recorded ground speed of the aircraft on the 17 previous spray runs and up to the last recorded data point, it was likely that the aircraft struck the bird while flying at about 116 kt (200 km/h). This resulted in a collision with sufficient energy to break the right Perspex windshield as evidenced by the biological residue found on the aircraft and broken windshield pieces.

Given the expected track and speed of the aircraft, and the location of the bird wings to the wreckage, the aircraft sustained flight for about 5 seconds following the birdstrike. As the aircraft was operating around 8 ft above ground level, and the wingspan (total of 15.8 m) was longer than this height, it was likely the aircraft entered a short climb to result in a near vertical attitude at impact, indicative of a loss of control. There was also significant damage to the pilot’s helmet as well as a small amount of bird biological residue found in the visor. However, in that time, the ATSB could not determine exactly what occurred as a result of the bird entering the aircraft as no flight data was recorded nor was the extent of the pilot’s injuries from the birdstrike known. That is, whether the pilot sustained a level of incapacitation due to a direct strike from the bird, or whether they experienced a level of startle or distraction following the event.

Therefore, with no other reasonable explanation and having excluded a pre-existing medical condition, it was likely that the birdstrike affected the pilot’s ability to control the aircraft when operating at low-level, resulting in a loss of control and collision with terrain.

Findings

From the evidence available, the following findings are made with respect to the birdstrike and collision with terrain involving Air Tractor 502B, VH-KDR, on 19 September 2022.

Contributing factor

• During aerial spraying operations at low-level, a large bird (Australian bustard) struck the right windshield of the aircraft and entered the cockpit. This likely affected the pilot’s ability to control the aircraft, resulting in a collision with terrain.

Sources and submissions

Sources of information

Sources of information during the investigation were gathered from:

• Smart Air Services Pty Ltd
• the aircraft loader
• Air Tractor Incorporated
• Pratt & Whitney Canada
• Civil Aviation Safety Authority
• maintenance organisation for VH-KDR
• Queensland Police Service.

References

Australian Museum (2020) Australian Bustard. Retrieved from https://australian.museum/learn/animals/birds/australian-bustard/

Bird Fact (2022) Australian Bustard - Ardeotis australis. Retrieved from https://birdfact.com/birds/australian-bustard

Ziembicki, M (2010) Australian Bustard. CSIRO Publishing: Collingwood.

Submissions

Under section 26 of the Transport Safety Investigation Act 2003, the ATSB may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. That section 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 following directly involved parties:

• Smart Air Services Pty Ltd
• the aircraft loader
• Air Tractor Incorporated
• Pratt & Whitney Canada
• Civil Aviation Safety Authority
• Transportation Safety Board of Canada
• United States National Transportation Safety Board.

Submissions were received from Air Tractor Incorporated. The submission was 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 2023 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.

Executive summary

What happened

On 31 August 2022, an Ayres Corporation S2R-T15, registered VH-IWI, was conducting aerial application operations. Around 100 m into a take-off roll, the pilot heard a bang as the left wing hit the ground and the aircraft performed a ground loop. No injuries were sustained.

Inspection revealed that the left main landing gear shock assembly had failed, with the left main landing gear folding outwards and contacting the bottom of the wing.

What the ATSB found

The ATSB examination identified that the lower tube of the left shock strut assembly failed at a fatigue crack, which led to the collapse of the left main landing gear and the aircraft wing to strike the ground. It is very likely that this lower tube was a part that Thrush Aircraft had instructed owner/operators to replace or modify in 1994 in accordance with a service bulletin. The reason the part was not replaced or modified was not identified.

Safety message

Manufacturers issue service bulletins to inform owners and operators about critical and useful information on aircraft safety, maintenance, or product improvement. The ATSB strongly encourages compliance with service bulletins pertaining to aircraft safety.

Additionally, on acquisition of an aircraft, it is important to review maintenance documentation to determine whether all the appropriate manufacturer issued instructions have been addressed.

 

The investigation

Decisions regarding the scope of an investigation are based on many factors, including the level of safety benefit likely to be obtained from an investigation and the associated resources required. For this occurrence, a limited-scope investigation was conducted in order to produce a short investigation report, and allow for greater industry awareness of findings that affect safety and potential learning opportunities.

The occurrence

On 31 August 2022, an Ayres Corporation S2R-T15, registered VH-IWI, was conducting aerial application operations from Trangie Airport. The pilot was the only person on board.

Three application flights were scheduled for the day, and after performing the daily aircraft inspection, the pilot commenced the first flight at around 0800 local time. Once the first flight was complete, the pilot obtained fuel and chemical, and taxied back to the airstrip to commence the second flight. The aircraft was around 100 m into the take-off roll when the left wing struck the ground, resulting in the pilot’s view being obstructed with chemical, and the aircraft started to rotate. In an attempt to stop the aircraft, the pilot applied reverse thrust. After completing a ground loop, the aircraft came to rest near the left edge of the runway. No injuries were sustained.

Initial inspection revealed that the left main landing gear shock assembly had failed, with the left main landing gear folding outwards and contacting the bottom of the wing.
 

Context

Operational and maintenance history

VH-IWI was manufactured by the Ayres Corporation in 1980 and first registered in Australia on the 24 December that year. The aircraft was powered by a Pratt & Whitney PT6A -15AG engine. It had accumulated 14,474.5 hours total time in service. The last maintenance took place 2 days prior to the occurrence and the right landing gear was replaced in 2012.

The Ayres Corporation[1] had issued 2 service bulletins associated with the main landing gear shock strut assembly.[2] The first, SB-AG-31,[3] was issued in 1992 in response to failures that occurred in the upper tube. This service bulletin described a modification to strengthen the upper tube, adding 4 rosette welds to the existing 2 rosette welds.[4] An entry in the aircraft maintenance logbook indicated that shock struts were welded to comply with SB-AG-31.

The second service bulletin, SB-AG-36,[5] was issued in 1994 in response to failures that occurred in the lower tube and included strengthening the lower tube. SB-AG-36 described 5 methods to comply with the bulletin:

• (1) replace shock assembly with P/N 50116-29 (new)
• (2) replace shock assembly with P/N 50116-28F1 (re-worked)
• (3) replace lower tube P/N 50116-12 with P/N 50116-12F (modified)
• (4) replace lower tube P/N 50116-12 with P/N 50116-30 (new design)    
• (5) modify old lower tube P/N 50116-12

Option (5) described the removal of a plug from the top of the tube, removal of a 0.12” (3.0 mm) thick and 1 3/8” (34.9 mm) long reinforcing tube, insertion and welding (including rosette welds) of a 0.312” (7.9 mm) thick and 4.25” (108.0 mm) long reinforcing tube P/N 50116-100, and attachment of a plug by welding. No logbook record indicating compliance with SB-AG-36 was found. The current maintainer stated they were performing maintenance under the assumption that all previous maintenance had been performed according to the manufacturer’s instructions. The reason the lower tube was not replaced or modified when the service bulletin was issued was not identified.

The operator of VH-IWI acquired the aircraft in 2018. The maintainer commenced maintaining the aircraft in the 12 months leading up to the occurrence. They were not the maintainer of the aircraft when either SB-AG-31 or SB-AG-36 were issued.

ATSB technical examination

Examination of the damaged main landing gear shock strut assembly was conducted at the ATSB’s technical facilities. The overall condition of the assembly was aged, and the rubber shock biscuits were cracked (Figure 1).

Figure 1: Left main landing gear shock strut assembly

Schematic and photo showing the main landing gear shock struct assembly with components labelled. The assembly was dirty and the shock biscuit exhibited cracks.

Source: Thrush Aircraft (schematic), ATSB (photo and annotations)

The grease between the upper and lower tube was discoloured; however, there was fresh blue grease around the lubricator and within the lower tube near the point of fracture. A substantial amount of new grease appeared to have migrated out of the lower tube during the occurrence.

The left shock strut fractured at the hole where the lower steel tube attached to the slider plate
(Figure 1). This allowed the bottom portion of the lower tube to separate from the remaining shock assembly. The fracture was consistent with fatigue cracking, followed by unstable crack growth and ductile overstress (Figure 2).

Figure 2: Fractured lower tube exhibiting characteristics of fatigue fracture followed by overstress

Fracture surface exhibiting characteristic of fatigue cracking followed by overstress.

Source: ATSB

The construction of the fractured lower tube consisted of a steel tube, around 3 mm thick and 400 mm long, reinforced inside with a smaller diameter section of tube, around 3mm thick and 35 mm long, over the region of the through-hole. The reinforcing tube was welded to the lower tube at the top end. There was no end cap on the lower tube, which was present on a more recently-manufactured lower tube. There were no rosette welds connecting the lower tube and the reinforcing tube. The upper tube of the left shock strut assembly was also examined; there were 6 rosette welds present on the upper tube.

There was a handwritten marking on the side of the lower tube (Figure 3) which was identified as the characters ‘50116T012 25[?]’, with an indistinct character at the end.

Figure 3: Marker writing on the lower tube

Writing on the lower tube written in black marker. The characters are 50116T012 25[?].

Source: ATSB

Advice from Thrush Aircraft

When asked about the writing on lower tube Thrush Aircraft informed the ATSB that first 5 characters represented a drawing number,[6] the 3 numbers to the right of the ‘T’ identified the component on that drawing. The drawing number, together with the component identifier, was considered the ‘part number’ and these numbers, 50116-012, were consistent with the designation of the pre-1994 lower tube design. Thrush Aircraft suggested the final 3 digits might be an employee or serial number.

A representative from Thrush Aircraft concluded that, based on the available evidence, the failure was consistent with the failures that prompted SB-AG-36.

Safety analysis

The ATSB examination identified that the lower tube of the left shock strut assembly failed at a fatigue crack. The crack had initiated from the hole where the lower tube attached to the slider plate and propagated around the circumference of the tube. The final fracture most likely took place early in the take-off roll which led to the collapse of the left landing gear and the aircraft wing striking the ground.

The dimensions of the reinforcing tube[7] and the writing on the side of the part were most consistent with the original part, noting that there was no plug. This part was to be replaced or modified to comply with SB-AG-36. It was not possible to determine the exact age of the lower tube. It is very likely that this lower tube was manufactured prior to 1994 and would need to have been replaced or modified to comply with the service bulletin.

The construction of the upper tube was consistent with SB-AG-31, which was issued 2 years before SB-AG-36.

Findings

ATSB investigation report findings focus on safety factors (that is, events and conditions that increase risk). Safety factors include ‘contributing factors’ and ‘other factors that increased risk’ (that is, factors that did not meet the definition of a contributing factor for this occurrence but were still considered important to include in the report for the purpose of increasing awareness and enhancing safety). In addition, ‘other findings’ may be included to provide important information about topics other than safety factors.  These findings should not be read as apportioning blame or liability to any particular organisation or individual.

From the evidence available, the following findings are made with respect to the ground strike involving Ayres Corporation S2R-T15, registration VH-IWI at Trangie, New South Wales on 31 August 2022.

Contributing factors

• A fatigue crack initiated in the lower steel tube of the left shock assembly, located at the hole where the lower tube connected to the slider plate. The fatigue crack grew until there was a final overstress failure of the lower tube during the take-off roll.
• For reasons not identified, it was highly likely that the lower tube of the left landing gear shock assembly was not replaced or modified to comply with SB-AG-36, which was issued to prevent similar failures.

Sources and submissions

Sources of information

The sources of information during the investigation included the:

• pilot of the occurrence flight
• Civil Aviation Safety Authority
• Thrush Aircraft
• maintenance organisation for VH-IWI

References

Main Landing Gear Shock Strut Inspection and Repair, Service Bulletin: SB-AG-31, Ayres Corporation, 3 November 1992. [Online]. Available: https://thrushaircraft.com/support/technical-publications/Service%20Bulletins/sb-ag-31%20MAIN%20LANDING%20GEAR%20SHOCK%20STRUT%20INSPECTION%20AND%20REPAIR.pdf [Accessed: 18 November 2022].

Main Landing Gear Shock Strut Modification, Service Bulletin: SB-AG-36, Ayres Corporation, 25 March 1994. [Online]. Available: https://thrushaircraft.com/support/technical-publications/Service%20Bulletins/sb-ag-36%20MAIN%20LANDING%20GEAR%20SHOCK%20STRUT%20MODIFICATION.pdf. [Accessed: 18 November 2022].

Maintenance Manual Model S2R-T15 Model S2R-T34, Serial Numbers T15-020 & Subsequent Numbers T34-091 & Subsequent Numbers, Thrush Aircraft Inc., 25 October 1990. Revised: 24 July 1991.

Submissions

Under section 26 of the Transport Safety Investigation Act 2003, the ATSB may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. That section 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 following directly involved parties:

• pilot of the occurrence flight
• maintenance organisation for VH-IWI
• Civil Aviation Safety Authority
• Thrush Aircraft

No submissions were received.

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 2023 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]     In 2003, Ayres Corporation’s assets were purchased by Thrush Aircraft, the current producer of the aircraft.

[2]     The main landing gear shock strut reduces the landing loads transmitted from the landing gear to the fuselage.

[3]     Main Landing Gear Shock Strut Inspection and Repair, Service Bulletin: SB-AG-31, Ayres Corporation, 3 November 1992. [Online]. Available: https://thrushaircraft.com/support/technical-publications/Service%20Bul…. [Accessed: 18 November 2022].

[4]     A rosette weld, also known as a plug weld, is a process that fuses two metals together by making a weld inside small circular holes.

[5]     Main Landing Gear Shock Strut Modification, Service Bulletin: SB-AG-36, Ayres Corporation, 25 March 1994. [Online]. Available: https://thrushaircraft.com/support/technical-publications/Service%20Bul…. [Accessed: 18 November 2022].

^[6]     The drawing number together with the component identifier was considered the ‘part number’.

[7]     The reinforcing tube was around 3mm thick and 35 mm long, while the pre-1994 reinforming was described as 0.12” (3.0 mm) thick and 1 3/8” (34.9 mm) long.

Introduction

At 0715 local time on 29 August 2022, a Cessna R182 Skylane RG (R182), registered VH-EHM, and operated by Executive Helicopters, departed a private property (Lighthouse) north-east of Roma, Queensland for an air transport flight under visual flight rules (VFR) to Archerfield Airport, Queensland. The aircraft departed with the pilot, 2 passengers, and a passenger’s pet dog on board.

As the aircraft was crossing the elevated terrain of the Great Dividing Range, the pilot encountered forecast low cloud and reduced visibility, and was unable to find a way across that was clear of cloud. With limited fuel remaining, the pilot diverted to Dalby Airport and refuelled the aircraft before departing once again. About an hour later, the aircraft collided with terrain in the D’Aguilar Range about 36 km north-west of Archerfield. All occupants of the aircraft were fatally injured.

The ATSB did not identify any aircraft defects or anomalies that may have contributed to the accident. As such, the following analysis will examine the:

• pilot’s pre‑flight planning and decision making.
• continuation of the flight into adverse weather.
• operator’s risk management.
• search and rescue aspects.

Pre-flight planning and diversion to Dalby

Pre-flight weather assessment

While at Lighthouse, the forecast weather at the destination (Archerfield) was suitable for a VFR arrival. However, the en route forecast predicted low cloud, rain, fog and associated reduced visibility, particularly around the elevated terrain of the Great Dividing Range, with cloud down to ground level in some areas. Although the forecast predicted acceptable visibility below the cloud layers, the weather conditions around the elevated terrain of the range – which had to be crossed to reach Archerfield – were unsuitable for visual flight.

About 6 minutes before the flight departed Lighthouse, the pilot requested weather information through the National Aeronautical Information Processing System (NAIPS). This was the pilot’s first NAIPS weather request since their departure from Archerfield the previous day. The exact weather information reviewed by the pilot could not be determined, nor is it not known whether the pilot accessed other non‑approved sources of weather. Nevertheless, the information available as part of the NAIPS request was sufficient for the pilot to assess the weather en route and to inform the pilot’s pre-flight decision making. However, the minimal time between the NAIPS weather request and subsequent take-off (6 minutes), provided limited opportunity for a thorough review of the en route weather and was consistent with a strong motivation to conduct the flight despite the forecast weather conditions.

Fuel planning

Although fuel was available at Lighthouse, and there was sufficient fuel on-board to complete the flight in ideal conditions, there was insufficient fuel to account for weather-related diversions or meet contingency fuel requirements. Furthermore, depending on the fuel flow data used for the calculations, there was also insufficient final reserve fuel on-board.

The limited pre-flight planning information available meant that the investigation was unable to determine the factors considered by the pilot with respect to fuel planning prior to departure.

Flight notification and search and rescue

Civil Aviation Safety Regulations (CASR) required that, before undertaking the flight, the pilot leave details of the flight and arrival time with a ‘responsible person’ who knew when and how to contact the Joint Rescue Coordination Centre (JRCC) and would do so if the flight was overdue.

The operator's documented procedures reflected this requirement and stipulated that this ‘responsible person’ could be someone from the operator or a representative of a client onboard.

The operator reported that the pilot did not always follow this documented procedure and would sometimes provide flight details to a family member, as occurred on the accident flight. Providing the flight note to this family member was still sufficient to meet the regulatory requirement if they were aware of their obligations and expected actions. However, the investigation identified various deficiencies in relation to the pilot’s compliance with this requirement for the accident flight:

• The family member was notified about 30 minutes after departure from Lighthouse, limiting search and rescue (SAR) functions during that initial part of the flight.
• An unspecific arrival time of ‘late morning’ was provided which was not suitable for the purposes of timely SAR action.
• Although the flight diverted to Dalby, the pilot did not notify the family member when there was a change to the flight route and arrival timing.
• The pilot had not briefed the family member about the correct actions to take for an overdue flight, resulting in a delay of over 2 hours in notifying the operator that the flight was overdue. This substantially delayed the subsequent emergency SAR response and potential medical attention (in the event of a survivable accident).

The accident was non-survivable, and consequently, a timely SAR response and immediate medical aid would not have altered the outcome. However, in different circumstances, SAR notification and the carriage and use of manually activated emergency beacons would likely have played a key role in increasing the chances of post‑accident survival by facilitating rapid medical aid to treat injuries.  

Departure and diversion to Dalby

While the forecast indicated that visual meteorological conditions (VMC) would probably not be maintained while transiting the elevated terrain of the Great Dividing Range, the decision to depart Lighthouse towards Archerfield was in itself reasonable. With adequate fuel on-board and escape routes planned, actual weather conditions could be assessed and managed in‑flight. However, doing so placed additional pressure on the pilot’s in‑flight decision making capabilities.

After encountering the forecast cloud, the pilot continued to manoeuvre to try to find a way through the Great Dividing Range with the aircraft’s fuel running low. The pilot then diverted to Dalby, landing with less than the required final reserve fuel. Although this may have been influenced in part by the aircraft being equipped with instrumentation capable of indicating the fuel state accurately, this indicated that the pilot’s planning before and during the flight was ineffective, and there was likely a motivation to complete the flight even though reduced safety margins existed.

Summary

The operator’s small size, limited flight operation personnel, and single-pilot operations, required the accident pilot to exercise a high level of pre-flight and in-flight decision-making autonomy.

The decision to depart into forecast weather that was not conducive to visual flight, with limited time for pre-flight weather review, insufficient required fuel on-board, and without an appropriate flight notification system, combined to reduce safety margins and presented an increased risk to occupants on board.

Although not contributory to the accident outcome, these factors were indicative of inadequate pre-flight planning. However, having attempted to cross the Great Dividing Range, the pilot did make an effective in-flight decision to return to Dalby to refuel. This provided an opportunity to re‑assess the weather and its suitability for continued flight, as well as providing an opportunity to identify alternative options for the passengers’ continued travel.

Departure from Dalby and continuation of flight

Pre-flight

The updated weather forecast information available to the pilot while at Dalby, predicted en route weather conditions similar to those already experienced but the forecast cloud between 1,500‑2,500 ft was expected to start dissipating within about 2 hours (by 1100), and clear completely by 1200. Also, the forecast deteriorating weather conditions with reduced visibility and low cloud cover at Amberley and Archerfield were also expected to clear by 1100. These deteriorating periods had forecast conditions that were less than the VMC criteria necessary to achieve the minimum height of 1,000 ft AGL required to transit Amberley and land at Archerfield. However, the conditions were within the special VFR criteria which meant that a transit and landing could have been possible had the pilot requested a special VFR clearance from air traffic control at Amberley and Archerfield.

As was the case before departing Lighthouse, although good visibility below the cloud layers was forecast en route, conditions around the elevated terrain of the Great Dividing Range were not suitable for visual flight.

The pilot spent about 10 minutes on the ground at Dalby, almost all of which was spent refuelling, and there were no NAIPS logins recorded during time on the ground. Therefore, it was unlikely that the pilot reviewed the updated forecast either from NAIPS or another source before take-off (for example, via radio), although the pilot likely reviewed it in-flight, shortly after take-off. This suggested that there was little consideration given to alternative means of getting the passengers to their destination (for example, ground transport or waiting until the weather improved as was forecast to occur within the next 2-3 hours). The short time spent on the ground provided further evidence of a motivation to complete the flight despite having already encountered the forecast cloud in-flight.

After departure

Across the Great Dividing Range

The flight track showed that, following departure from Dalby, the pilot attempted to find a different path across a section of the Great Dividing Range to get to the low-lying terrain west of Amberley, this time tracking south of Toowoomba. The pilot requested updated weather through NAIPS about 13 minutes after take-off. However, re-assessing the weather in‑flight removed the ability of the pilot to decide against taking off again. Research indicates that continuing with an initial plan or strategy despite indications that an alternative course of action may be safer can be stronger after take-off, making the decision to discontinue the flight more difficult.

The forecast and observed weather conditions, as well as the pilot’s flight track, indicated that a cloud layer was obscuring terrain on Main Range. The pilot’s first attempt to cross this range was unsuccessful, likely due to cloud, with a 180° turn performed to try to cross at a different location. The pilot then tracked south-east and between 2 peaks at low level, before heading down a long valley towards low-lying coastal terrain.

The flight track suggested familiarity and knowledge of the terrain with the pilot possibly heading towards known locations that presented the best chance of transit. However, the aircraft crossed over the terrain with minimal separation. This was likely indicative of the pilot’s intention and motivation to continue and complete the flight, despite the forecast and observed weather conditions.

Amberley and surrounding area

After crossing Main Range, the pilot descended the aircraft towards Gatton before tracking towards Lowood, following the low-lying terrain below cloud. Weather observations from Gatton, Amberley, and Lowood Golf Course indicated a low cloud base but with good visibility below the layer. However, a weather report about 15 minutes after the accident indicated the forecast deteriorating weather at Amberley had reduced the observed visibility to 4,000 m and lowered the scattered cloud to 700 ft. These conditions would have meant that the area was not suitable for visual flight at that time.

Lake Manchester VFR route

The Lake Manchester VFR route, intended to be used in VMC, had narrow lateral and vertical separation between the hilly terrain of the adjacent D’Aguilar Range and controlled airspace, which increased the difficulty of this route in poor weather. The flight track indicated that the pilot was trying to follow the route to get to Archerfield. However, the adjacent range was obscured by cloud, so it was likely difficult to orientate the aircraft with identifiable ground features to assist with navigation. In addition, the aircraft’s relatively high speed for the conditions reduced the time available for decision making.

After turning east over Fernvale, the aircraft tracked towards a mountain at an altitude below its peak before conducting another 180° turn at about 1,200 ft and less than 2 km from rising terrain. About 10 seconds later, the pilot conducted a right turn at about 140 kt ground speed towards the rising terrain of the D’Aguilar Range, which had low cloud cover obscuring some of the peaks, and shortly after, impacted terrain.

Evidence from the site of the accident and the sound of the aircraft up until the impact indicated:

• no abnormal engine or propeller sounds up to the time of impact
• no pre-impact issues with the aircraft, engine, or propeller
• a relatively high-speed impact
• a shallow descent at the time of impact with the trees and terrain
• an angle of bank at the time of impact similar to previous turns completed during the flight.

As a result, the ATSB concluded that the aircraft was likely under control during the impact with the pilot very likely unaware of the aircraft’s proximity to the terrain. Although the pilot was knowledgeable and experienced with the location of the D’Aguilar Range relative to Fernvale, the low cloud over the range and the flight track indicated the aircraft was very likely in instrument meteorological conditions (IMC) in the final stages of the flight. The pilot very likely lost visual reference with the ground, limiting the ability to geographically orient themselves and avoid the surrounding hills, eventually leading to a controlled flight into terrain. The high speed would also have limited the pilot’s ability to react and avoid any terrain identified in the final moments of the flight.  

Pilot decision making

On the day of the accident, the pilot was conscious of the adverse weather forecast and encountered during the flight prior to landing at Dalby. The pilot then made a deliberate decision to depart Dalby and continue the flight into known unsuitable weather conditions at cruise speed, low altitude, and towards a known area of rising terrain obscured by cloud. This was inconsistent with the prudent judgement, behaviour and decision making capability attributed to them by others. Nevertheless, the available information suggests the pilot was probably influenced by plan continuation bias – an internal pressure or desire to get to the destination – to continue the flight, which probably became stronger as they got closer to Archerfield Airport. However, due to a lack of information, the ATSB was unable to conclusively determine the pilot’s motivations in choosing to continue the flight in unsuitable circumstances with reduced safety margins.

Operator risk management

Although inadvertent IMC was a well-known, high consequence risk, the operator's hazard and risk register, which formed part of the organisation's safety management system (SMS), did not identify this hazard. Consequently, there was no specific assessment of its risk to operations and no specific risk mitigation in place.

Although not documented in the operator’s SMS, the risk of inadvertent IMC was being managed by requiring that pilots not commence flights unless VFR conditions were forecast for the route and at the destination, and to monitor and plan alternative actions in-flight if weather deteriorated. In this accident, the pilot did not follow these procedural controls. The controls relied on an individual pilot’s pre-flight assessment of weather, and on in-flight weather-related decision making, neither of which were addressed by the operator through any specific training, procedures, or review.

In addition, the operator did not have any procedures or training in place beyond the minimum regulatory instrument flying component of fixed wing flight reviews for pilots to recover from an inadvertent IMC encounter. Historically, relying on pilots to avoid IMC has been the primary strategy in preventing inadvertent IMC accidents. However, the continuing occurrence of these accidents demonstrates that this strategy alone is insufficient.

As the hazard of inadvertent entry into IMC was not identified in the risk register, no conclusions could be drawn about the nature and effectiveness of the specific risk controls that might have been implemented, nor whether the pilot would have utilised them. Therefore, while important, the absence of the hazard being identified in the operator’s SMS was not considered contributory to the accident.

On-board recording devices

There was no regulatory requirement for the aircraft to be fitted with on-board recording devices. However, this and numerous other investigations have shown that the lack of such devices limits the ability to understand and determine all of the factors that contributed to an accident. In turn, important safety issues that present a hazard to current and future operations were potentially not identified. Conversely, other investigations where some form of recording device was on board, provided valuable information regarding the accident.

In this investigation, flight track data was available to identify the aircraft’s movement. However, the ATSB was unable to determine why a highly experienced commercial pilot continued the flight into a known area of high terrain obscured by cloud, nor could the circumstances of what was occurring in the cockpit, especially in the final minutes before the accident, be determined. Recorded information may have provided some insights into the pilot’s decision making during the flight and further detail on how the accident developed.

The use of lightweight recorders on smaller aircraft conducting commercial operations has the potential to provide a relatively simple and cost-effective way of achieving many of the benefits that are provided by traditional recorders fitted to larger aircraft.

Findings

From the evidence available, the following findings are made with respect to the VFR into IMC and controlled flight into terrain involving Cessna R182, VH-EHM, 36 km north-west of Archerfield Airport, Queensland on 29 August 2022

Contributing factors

• After encountering and manoeuvring around forecast low cloud, insufficient fuel remained on board the aircraft to complete the flight and the pilot diverted to Dalby to refuel. After refuelling, the pilot departed toward forecast en route weather unsuitable for visual flight.
• The pilot continued the flight at low level, at cruise speed, into weather conditions unsuitable for visual flight. This very likely resulted in the pilot experiencing a loss of visual reference leading to controlled flight into terrain.

Other factors that increased risk

• Although in this instance the accident was not survivable, several deficiencies were identified that delayed a search and rescue response:
  • Contrary to operator procedures, the pilot provided flight notification information to a family member not associated with the operator.
  • The family member was not provided with any information on appropriate actions to be taken in the event of the aircraft being overdue.
  • The pilot provided flight notification information to the family member about 30 minutes after take-off from Lighthouse which limited search and rescue functions for that time.
• The aircraft likely departed Lighthouse for the flight to Archerfield Airport with insufficient fuel to account for weather-related diversions and the required contingency and final reserve fuel.
• The operator's hazard and risk register, which formed part of the organisation's safety management system, did not identify inadvertent entry into instrument meteorological conditions as a hazard, which reduced the ability of the organisation to effectively manage the related risk. (Safety issue)

Other findings

• While flight tracking data was available, the aircraft was not fitted with an onboard recording device. This could have provided valuable information to better understand the pilot’s in-flight weather-related decision making.

Safety issues and actions

Operator risk management

Safety issue number: AO-2022-041-SI-01

Safety issue description: The operator's hazard and risk register, which formed part of the organisation's safety management system, did not identify inadvertent entry into instrument meteorological conditions as a hazard, which reduced the ability of the organisation to effectively manage the related risk.

Glossary

Sources of information

The sources of information during the investigation included:

• the pilot’s next of kin and former colleagues
• the passengers’ friends and colleagues
• Bureau of Meteorology
• Geoscience Australia
• Airservices Australia
• Executive Helicopters (operator)
• Civil Aviation Safety Authority
• OzRunways
• Avdata
• Fernvale Rural Fire Brigade
• Queensland Police Service
• witnesses
• CCTV video footage of weather near the aircraft’s flight path.

References

Australian Transport Safety Bureau. (2005). General Aviation Pilot Behaviours in the Face of Adverse Weather. B2005/0127

Australian Transport Safety Bureau. (2013). A review of the effectiveness of emergency locator transmitters in aviation accidents.

Australian Transport Safety Bureau. (2019). Avoidable Accidents No. 4: Accidents involving Visual Flight Rules pilots in Instrument Meteorological Conditions. AR-2011-050

Australian Transport Safety Bureau. (2021). Collision with water involving de Havilland Canada DHC-2, VH-NOO. AO-2017-118

Australian Transport Safety Bureau. (2021). VFR into IMC and controlled flight into terrain involving Pilatus Britten-Norman BN2A, VH-OBL. AO‑2018‑078

Australian Transport Safety Bureau. (2023). VFR into IMC, loss of control and collision with terrain involving Airbus Helicopters EC130 T2, VH-XWD. AO-2022-016

Civil Aviation Safety Authority. (2019). Safety behaviours: human factors for pilots: Resource booklet 7 Decision making (2^nd edition)

Collins, J. (2022). Training and Safety Tip: Escaping the VFR-into-IMC Trap. Aircraft Owners and Pilots Association (AOPA).

Endsley M.R. (1995). Toward a theory of situation awareness in dynamic systems. Human Factors, Volume 37, No. 1, 32-64

Federal Aviation Administration. (2023). Information for Operators (InFO): Terrain Awareness and Warning Systems (TAWS) Nuisance Alerts

General Aviation Joint Steering Committee (GAJSC). (2021). Controlled Flight Into Terrain (CFIT) Working Group

General Aviation Joint Steering Committee (GAJSC). (2021). Controlled Flight Into Terrain (CFIT) Working Group: Appendix 6 — Detailed Implementation Plans for Safety Enhancements

Harris M.R., Fein E.C. and Machin M.A. (2022). A Systematic Review of Multilevel Influenced Risk-Taking in Helicopter and Small Airplane Normal Operations. Front. Public Health 10:823276. doi: 10.3389/fpubh.2022.823276

International Civil Aviation Organization. (2018). Safety Management Manual (4th ed). ICAO Doc 9859, Montreal

O'Mahony et al. (2023). VFR Into IMC Through the Lens of Behavioral Economics. Journal of Air Law and Commerce. Volume 88, Issue 1, Article 4 https://scholar.smu.edu/jalc/vol88/iss1/4

Smith, M. (2022). 178 Seconds: VFR into IMC is an Insidious Trap. Aircraft Owners and Pilots Association (AOPA)

Stanton A. (2022). 'Gathering Clouds' A Study of Plan Continuation, Risk, Rules, and Pilot Behaviour. https://doi.org/10.25904/1912/4844

Wilson, D.R., & Sloan, T.A. (2003). VFR Flight Into IMC: Reducing the Hazard. Journal of Aviation/ Aerospace Education & Research, 13(1). https://doi.org/10.15394/jaaer.2003.1567

Submissions

Under section 26 of the Transport Safety Investigation Act 2003, the ATSB may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. That section 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 aircraft operator and the Civil Aviation Safety Authority.

A submission was received from the aircraft’s operator.

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

Appendices

Appendix – Fuel calculations

Background

An assessment was undertaken to determine the likely quantity of fuel on-board the aircraft when it departed Lighthouse, and if that quantity was sufficient for the intended flight to Archerfield.

Calculations – Fuel on-board at Lighthouse

Using the maximum capacity of the fuel tanks and the fuel uplift data at Dalby, the useable fuel remaining in the tanks on landing at Dalby could be calculated (20.7 L). The total fuel used during the flight (taxi, climb, and cruise) from Lighthouse to Dalby was estimated using the pilot’s operating handbook (POH),[48] the operator’s operations manual, and the forecast weather.

Taxi

Total taxi fuel was calculated from the POH as the difference between the maximum ramp weight (1,411.6 kg) and maximum take-off weight (1,406.1 kg). The operator’s operations manual also provided an estimated taxi fuel amount.

Climb

Climb fuel consumption rate data in the POH was available for maximum aircraft weight, and International Standard Atmosphere (ISA) conditions and for normal climb performance. Similar temperature conditions to International Standard atmosphere (ISA) were forecast. The main cruise altitude from the aircraft’s OzRunways flight track data was 3,400 ft between Lighthouse and Dalby. The POH data provided fuel consumption for 3,000 ft and 4,000 ft so an interpolation calculation was conducted.

Cruise

Fuel consumption from Lighthouse to Dalby was calculated using data provided in the POH, operator’s operations manual, and forecast weather. As the cruise power setting for the flight was not known, cruise fuel consumption rates were calculated for 3 common engine power settings (75%, 65% and 55%). The POH data assumed maximum take-off weight and standard ISA conditions. The most relevant data for the 3,400 ft cruise altitude was contained in the performance chart for 4,000 ft altitude. For calculations conducted using the operations manual data, the provided ‘block fuel flow’ was used as the cruise and climb fuel flow.

Cruise time was based on the OzRunways flight track data after reaching cruise level until landing at Dalby. Although about two-thirds of the flight was at a similar cruise altitude, the remaining third was made up of 2 climbs and 2 descents along with heading changes. It was assumed that the fuel consumption rate during these phases averaged out to be similar to the cruise fuel consumption rate (since a higher fuel consumption rate would occur during climb with a lower rate during descent). Therefore, the cruise fuel consumption rate was used for the flight time between initially reaching cruise altitude and landing at Dalby.

The actual weight of the aircraft at the same power level and operating altitude can affect the cruise speed of the aircraft and thus flight time and total fuel consumption. POH fuel consumption and related true airspeed (TAS) data was provided at maximum weight but the aircraft during the flight was below this weight. The estimated aircraft departure weight based on the operations manual fuel usage was about 1,269 kg. However, the difference in TAS at this weight was considered minimal so no correction to the cruise time was applied.

The estimated fuel on-board at Lighthouse for each scenario is shown in Table 2.

Table 2: Estimated fuel on-board VH-EHM at Lighthouse

Calculations – Fuel required from Lighthouse to Archerfield

The Civil Aviation Safety Regulations defined the fuel requirements for Part 135 operators. The following fuel was required for a Part 135 flight:

• taxi fuel – fuel used before take-off
• trip fuel – fuel for take-off, climb, cruise, descent, and landing
• contingency fuel – 10% of trip fuel for a piston engine aeroplane to compensate for unforeseen factors
• final reserve fuel – 45 minutes flight time for a piston engine aeroplane and is useable fuel remaining on completion of final landing at the aerodrome
• Destination alternate fuel (if required)
• Holding fuel (if required)
• Additional fuel (if required)

An estimate on the fuel required to complete the flight from Lighthouse to Archerfield (taxi, climb, cruise) was determined using data available in the POH, operations manual and the forecast wind conditions.

The anticipated ground speed was calculated based on the TAS and forecast wind conditions. The Bureau of Meteorology grid point wind and temperature chart forecast 13‑21 kt wind from 70° at 2,000 ft and 14-20 kt wind from 50-80° at 5,000 ft for most of the planned flight which was on a 115° track. Therefore, an estimated 10 kt headwind was used for the calculations. For the different power settings, the ground speed was used along with the cruise distance to determine the total time and fuel consumption during cruise. For the operations manual fuel flow calculations, the total planned distance to Archerfield was used as the ‘block fuel flow’ accounted for climb, cruise, and descent fuel consumption.

Cruise fuel planning was conducted to overhead the destination aerodrome at cruise level. Manoeuvring and approach fuel consumption was not considered.

Contingency fuel

Contingency fuel was 10% of the calculated trip fuel.

Final reserve fuel

Final reserve fuel was 45 minutes at holding speed, ISA conditions and 1,500 ft. The POH provided a holding fuel of 23.8 L based on 45 minutes at 45 % engine power. The operations manual provided an expected ‘holding’ fuel flow of 40 L per hour (30 L for 45 mins).

Destination alternate fuel, holding fuel, and additional fuel

Either the ‘destination alternate fuel’ or ‘holding fuel’ was required if the destination forecast was below the alternate minima. The relevant Terminal Area Forecast (TAF) for Archerfield forecast conditions above the alternative minima, so no holding or destination alternate fuel was required for the flight. ‘Additional fuel’ was only required for multi-engine aircraft, or aircraft with pressurisation. Since VH‑EHM was a single-engine non-pressurised aircraft, ‘additional fuel’ was not required. The results are shown below in Table 3.

Table 3: Estimated minimum regulatory fuel required for planned flight to Archerfield

Estimated fuel margin

The results of the above calculations are shown below in Table 4 to determine the fuel margin between the useable fuel on-board and the minimum required fuel.

Table 4: Estimated fuel margin for the planned flight to Archerfield

The results are based on forecast ideal weather conditions. As per the general fuel requirements in the CASR, in determining the quantity of useable fuel required, the pilot in command must also consider the effect of relevant meteorological reports and forecasts.

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 2023 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.