Safety summary

What happened

On 12 August 2021, at about 0858 Central Standard Time, an Aero Commander 500-S aircraft, registered VH-LTP, departed Port Lincoln on a private flight to Adelaide, South Australia under instrument flight rules. On board were the pilot and one passenger.

During the descent in instrument meteorological conditions, when passing about 6,000 ft, the aircraft began to encounter turbulence. Following clearance to track direct to the GPS waypoint GULLY, the pilot reported having difficulties entering the area navigation (RNAV) instrument approach data into the aircraft’s touchscreen multi-function display due to turbulence. By the time the pilot had correctly entered it, the aircraft had just passed the waypoint. When the pilot then selected the ‘Direct-To’ option on the display, the autopilot commanded a sharp turn to the right, to commence an orbit to attempt to overfly the waypoint to recapture it.

On observing the aircraft’s track, the controller instructed the pilot to maintain 3,800 ft and to turn onto a heading of 360°, intending to vector the aircraft back towards the waypoint. About 90 seconds later, the controller instructed the pilot to turn onto heading 120°, which the pilot read back. When the controller tried to contact the pilot about 1 minute later, the pilot did not respond and communications were lost.

The aircraft continued on the assigned heading but began descending below its assigned altitude, which was also the minimum sector altitude. For 4 minutes, ATC, with the assistance from the pilot of a nearby aircraft, continued to attempt to contact the pilot of VH-LTP. During this time, the approach controller also issued the pilot three terrain safety alerts.

The pilot contacted the Melbourne Centre controller, who instructed the pilot to transfer back to the Adelaide Approach frequency. Upon regaining communication, the approach controller issued the pilot a terrain safety alert and instructed the pilot to climb immediately to 5,000 ft. The lowest altitude the aircraft descended to was 2,480 ft and the highest point within 5 NM of the aircraft’s track was 1,913 ft.

The aircraft then tracked to Adelaide Airport and landed without further incident.

What the ATSB found

The ATSB found that during the approach, the pilot was experiencing data entry difficulties due to turbulent conditions and inadvertently selected the incorrect radio frequency. Further, several factors including the environmental conditions, data entry difficulties and the timing of the clearance for the GULLY waypoint, likely led to the pilot experiencing a high workload. This in turn, likely affected the pilot’s situational awareness where they did not initially notice the frequency change nor the continued descent and descent below the assigned (minimum sector) altitude.

Safety message

The approach and landing phases are known periods of high workload for pilots. Pilots must continuously monitor aircraft and approach parameters, and the external environment, to ensure they maintain a stable approach profile and make appropriate decisions for a safe landing. The Flight Safety Foundation found that between 1984 and 1994, 50 per cent of controlled flight into terrain accidents were due to inadequate monitoring by the pilot. Distractions and unanticipated events can further increase a pilot’s workload leading to undetected errors and a loss of situational awareness. During high workload phases of flight, pilots should remain focused on monitoring the aircraft instruments and avoid fixating on a problem.

The investigation

The occurrence

On 12 August 2021, at about 0858 Central Standard Time,^[1] an Aero Commander 500-S aircraft, registered VH-LTP, departed Port Lincoln on a private flight to Adelaide, South Australia under instrument flight rules.^[2] On board were the pilot and one passenger.

At about 0930, the pilot made first contact with Adelaide Approach air traffic control (ATC) and was cleared to track direct to Adelaide maintaining 7,000 ft. The controller also advised the pilot to expect vectors for the area navigation Z (RNAV-Z)^[3] approach for runway 23. The aircraft was in instrument meteorological conditions (IMC).^[4]

About 14 minutes later, the controller cleared the aircraft to descend to 5,000 ft and turn left on a heading of 050°. The pilot recalled the aircraft encountering turbulence during the descent, which started when passing about 6,000 ft. At 0947, the controller cleared the pilot to descend to 3,800 ft and to turn right on a heading of 090°. The controller advised the pilot to expect a clearance shortly to navigate to GPS waypoint GULLY, which was the initial approach fix for the RNAV approach. Less than 1 minute later, when the aircraft was 6 NM north-west of GULLY, the controller cleared the pilot for the RNAV approach and instructed the pilot to track direct to GULLY (Figure 1).

The pilot began tracking to GULLY and reported having difficulties entering the RNAV approach details into the aircraft’s touchscreen multi‑function display (MFD) due to the turbulence. By the time the pilot had correctly entered it, the aircraft had just passed the GULLY waypoint and when the pilot then selected the ‘Direct-To’ option on the display, the autopilot commanded a sharp turn to the right, to orbit back around to recapture the waypoint (Figure 1).

Observing the aircraft’s track away from the waypoint, the controller, at 0951, advised the pilot they had missed the approach and instructed them to maintain 3,800 ft. They then instructed the pilot to turn right on a heading of 360° and to expect vectors back to GULLY.  After turning to the assigned heading, the aircraft started a gentle climb. At 0953, the controller instructed the pilot to turn right onto heading 120° and to expect a clearance to navigate to GULLY shortly. The pilot correctly read back the clearance and later recalled having commenced a descent, realising the aircraft was then above the assigned altitude. This was the last radio call the pilot received from the Adelaide Approach controller before the pilot inadvertently switched the radio frequency back to Melbourne Centre.^[5]

About 1 minute later, when the aircraft was 2.5 NM north of GULLY, the controller issued the expected clearance – to resume their own navigation, track direct to GULLY, descend to 3,800 ft and conduct the RNAV approach for runway 23. The controller did not receive a response from the pilot.

The pilot continued on their last assigned heading of 120° and the aircraft began descending below its assigned altitude of 3,800 ft, which was also the minimum sector altitude. For 4 minutes, ATC, with the assistance from the pilot of a nearby aircraft, continued to attempt to contact the pilot of VH-LTP. During this time, the controller also issued the pilot three terrain safety alerts.

At 0957, the pilot of VH-LTP made a radio call to ATC, which was received by the Melbourne Centre, to ask the approach controller whether they should track back to GULLY. When the pilot received no reply, they made another call. The Melbourne Centre controller responded, told the pilot to standby and then instructed them to contact Adelaide Approach on frequency 118.2. The controller then contacted the Adelaide Approach controller to inform them that they had the pilot on the Melbourne Centre frequency, and the pilot was transferring back to the correct frequency. Upon regaining communication with the pilot, the approach controller immediately issued a terrain safety alert and instructed the pilot to climb immediately to 5,000 ft. The aircraft’s altitude at that time was 2,780 ft.

After observing no increase in the aircraft’s altitude for almost 1 minute, the controller contacted the pilot and asked them to confirm climbing to 5,000 ft. The pilot confirmed and the aircraft began to climb. The lowest altitude the aircraft descended to was 2,480 ft, the highest point within 5 NM of the aircraft’s track was 1,913 ft (Figure 2).

Figure 1: VH-LTP flight path (dotted line shows RNAV instrument approach path)

Source: Google Earth overlaid with Airservices data, annotated by the ATSB

Figure 2: VH-LTP’s altitude relative to the terrain (in feet)

Source: Geoscience terrain data overlaid with Airservices data, annotated by the ATSB

The controller vectored the aircraft back to Adelaide Airport for a RNAV-Z approach for runway 05. The aircraft landed at Adelaide Airport at 1028.

Context

Pilot information

The pilot previously held a Commercial Pilot Licence (Aeroplane), but at the time of the incident was exercising the privileges of a Private Pilot Licence, with an instrument rating and a valid Class 2 Aviation Medical Certificate. The pilot had a total flying experience of about 16,800 hours, they had accrued about 9,800 hours on the Aero Commander 500-S and about 4,700 hours total instrument time. The pilot completed their instrument proficiency check on 29 June 2021 and last flew an RNAV approach in instrument meteorological conditions (IMC) on 6 August 2021.

Weather observations

The pilot reported being aware of forecast IMC prior to departing Port Lincoln and planned their flight accordingly. They reported operating in cloud until the aircraft had descended to 1,000 ft on approach for runway 05, and turbulence from 6,000 to 3,000 ft, with the intensity increasing between 4,000 ft and 3,000 ft.

The pilot’s observations were consistent with the forecast weather conditions.

Garmin GTN 650 touchscreen

The Garmin GTN 650 is a navigation aid that combines GPS, communication and navigation functions in an MFD, capable of showing high resolution terrain mapping, graphical flight planning, multiple weather options and traffic display. The touchscreen uses capacitive technology to sense the proximity of skin to the display, responding to light touches, without the need of pressure for detection.

Functions related to this incident include a ‘Direct-To’ key, which when pressed provided a direct course to a selected waypoint. The navigation and communication frequencies could be changed by touching the standby window and the keypad to enter the desired frequency. To flip between active and standby frequencies, the operator needed to touch the active navigation/communication frequency field (Figure 3). Upon touching the frequency field, a text box ‘Hold for Flip-Flop’ is displayed near the knob. If the active navigation/communication frequency field is touched and held it will switch between the navigation and communication frequencies.

Figure 3: Garmin GTN 650 display screen

Source: Garmin, annotated by the ATSB

The pilot reported that when operating in turbulent conditions, they usually put two fingers on the side of the display to stabilise their hand and use their thumb to enter the data. After missing the initial approach fix, and during the second attempt to enter the waypoint into the MFD, the pilot reported that their finger slipped and inadvertently selected the Melbourne Centre frequency, which was the previous frequency selected.

Workload

Speed and timeframe

During the descent, the aircraft’s airspeed was about 170 kt. The pilot reported receiving a late descent to commence the RNAV-Z approach, which resulted in the aircraft being too high and fast to commence the approach. The pilot reported feeling rushed due to the combination of speed and the timing of the descent and approach clearances. The pilot reported their ideal approach speed would be 120-140 kt. However, data from previous flights inbound to Adelaide, that were probably flown by the pilot, show the aircraft conducting approaches at similar speeds.  

Handling speeds for instrument approaches are specified in Airservices Australia’s Aeronautical Information Publication, En Route 1.16.1 These included a range of 120–180 kt indicated airspeed for Category B aircraft during the initial and intermediate approach. VH-LTP was within this range when the pilot was cleared to conduct the RNAV approach.

Monitoring

When assigned heading 120°, the pilot realised the aircraft had climbed since the turning onto the 360° heading, and therefore commenced a descent to the assigned altitude. However, the pilot was unaware the aircraft had subsequently continued to descend while they were out of communication with Adelaide Approach.

The pilot recalled becoming aware that something was wrong when they had not received any further instructions from ATC and recognised the voice of the Melbourne Centre controller. This prompted the pilot to contact the controller, who transferred them back to Adelaide Approach.

Self-assessment

The pilot rated their workload during the incident as 7 out of 10. The pilot reported that typically during an approach, their workload is ‘not too bad’, however, in this incident, the turbulence, difficulties entering in the waypoint, the late clearance for descent and then the aircraft turning away from the approach added to their workload. The pilot thought they were coping well with the situation, but advised that they were busy, with the situation moving quickly and little things were adding up.

Safety analysis

Data entry difficulties

The pilot’s data entry difficulties were consistent with the research conducted on touchscreen interfaces. During turbulent conditions, aircraft touchscreen interfaces have been known to increase data input errors and create slow interaction times. This is due to the display moving or vibrating independently of the pilot's body, which is itself also vibrating. A study conducted by Dodd and others (2014) found that while operating in moderate turbulence, it took pilots over 2 minutes longer to complete a given task when compared to operating in no turbulence. Although necessary when operating in turbulent conditions, stabilising the hand on the edge of the screen can also increase the risk of accidental presses on the edge of the screen (Coutts and others. 2019).

Ineffective monitoring

The pilot was experiencing a high workload, likely due to operating in instrument meteorological conditions, turbulence, encountering data entry difficulties and feeling rushed. This high workload likely resulted in the pilot not initially noticing the frequency change nor the continued descent and descent below the assigned altitude, which was the minimum sector altitude.

Workload is defined as the sum of task demands placed on an individual’s cognitive resources that are used for attention, perception, decision making and action (Skybrary, 2010). Humans are limited in the amount of new information their brain can process at once. Once this limit of cognitive resources has been reached their performance starts to decline with increased error rates and delayed responses, thus resulting in cognitive overload.

During high workload periods, monitoring flight instruments can degrade due to other tasks requiring attention, which can potentially lead to undetected errors (Flight Safety Foundation, 2014). A study on controlled flight into terrain (CFIT) conducted by the International Civil Aviation Organization, Flight Safety Foundation and the United States Federal Aviation Administration (Flight Safety Foundation, n.d.), found that two thirds of all CFIT accidents are a result of altitude error or lack of vertical situational awareness.

Though the pilot reported that they were coping well, pilots are often unaware that their monitoring performance has degraded, subsequently affecting their situational awareness.

Findings

From the evidence available, the following findings are made with respect to the descent below the minimum sector altitude involving an Aero Commander 500-S, registered VH-LTP that occurred near Adelaide Airport, South Australia, on 12 August 2021.

Contributing factors

• Due to turbulence, the pilot had difficulties entering data into the touchscreen multi-function display and the pilot inadvertently selected the incorrect radio frequency.
• The pilot was likely experiencing high workload in instrument meteorological conditions resulting in a loss of situational awareness and ineffective monitoring of the instruments. This led to the aircraft descending below the minimum sector altitude before the pilot regained communication with air traffic control.

Other findings

• The air traffic controller issued another terrain safety alert to the pilot after communications were restored, which likely prevented a controlled flight into terrain accident.

Sources and submissions

Sources of information

The sources of information during the investigation included:

• the pilot
• Airservices Australia
• Bureau of Meteorology
• Geoscience Australia.

References

• Coutts, L.V., Plant, K.L., Smith, M., Bolton, L., Parnell, K.J., Arnold, J., & Stanton, N.A. (2019). Future technology on the flight deck: assessing the use of touchscreens in vibration environments. Ergonomics, 62(2), 286-304. https://doi.org/10.1080/00140139.2018.1552013
• Dodd, S., Lancaster, J., Miranda, A., Grothe, S., DeMers, B., & Rogers, B. (2014). Touch Screens on the Flight Deck: The Impact of Touch Target Size, Spacing, Touch Technology and Turbulence on Pilot Performance. Proceedings of the Human Factors and Ergonomics Society 58^th Annual Meeting, 58(1), 6-10. https://doi.org/10.1177/1541931214581002
• Flight Safety Foundation (n.d.). Controlled Flight into Terrain: Education and Training Aid. Flight Safety Foundation. https://skybrary.aero/sites/default/files/bookshelf/2507.pdf
• Flight Safety Foundation (2014). A Practical Guide for Improving Flight Path Monitoring: Final report of the active pilot monitoring group. Flight Safety Foundation. https://flightsafety.org/files/flightpath/EPMG.pdf
• Skybrary (2010). Workload (OGHFA BN). Skybrary. https://www.skybrary.aero/index.php/Workload_(OGHFA_BN)

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:

• the pilot
• the aircraft operator
• Airservices Australia.

A submission was received from the pilot. 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 2022 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.  Central Standard Time (CST): Coordinated Universal Time (UTC) + 9.5 hours.
2.  Instrument flight rules (IFR): a set of regulations that permit the pilot to operate an aircraft in instrument meteorological conditions (IMC), which have much lower weather minimums than visual flight rules (VFR).
3.  Area navigation (RNAV) approach: An approach flown along a path of GPS waypoints.
4.  Instrument meteorological conditions (IMC): weather conditions that require pilots to fly primarily by reference to instruments, and therefore under instrument flight rules (IFR), rather than by outside visual reference. Typically, this means flying in cloud or limited visibility.
5.  Melbourne Centre: Responsible for enroute services throughout the Melbourne flight information region, which includes the southern half of Australia and the Southern and Indian oceans.

Safety summary

What happened

On 6 July 2021, a Boeing Company 737-376SF (B737), registered VH-XMO and operated by Express Freighters Australia, was operating a freight flight from Perth, Western Australia, to Melbourne, Victoria. While passing 8,000 ft on descent the flight crew received pressurisation system cautions and the cabin depressurised. The flight crew performed the non-normal checklist and continued to Melbourne for an uneventful landing.

What the ATSB found

The ATSB found that an aft cargo bay insulation blanket had not been secured correctly during a previous maintenance activity. During the descent, the unsecured blanket became lodged in the main pressurisation outflow valve, inhibiting its ability to control aircraft pressurisation.

What has been done as a result

The operator conducted an inspection of aft cargo compartment pressurisation components and insulation blankets in their B737 fleet resulting in multiple unserviceabilities being discovered and rectified. Following these inspections, Express Freighters Australia implemented a 4,000 hour or 36 month insulation blanket visual inspection task for the B737-300 and a 4,000 hour or 18 month visual inspection task for the B737-400.

Safety message

Aircraft periodic inspection tasks often require equipment or covering removal to access inspection areas. Maintenance crews are reminded of the importance of ensuring that any items removed for access are thoroughly inspected for serviceability and securely reinstalled. Items must be refitted in accordance with the maintenance manual to prevent unsecured items inhibiting flight critical systems such as the aircraft pressurisation system.

The investigation

The occurrence

On 6 July 2021, a Boeing 737-376 Special Freighter, registered VH-XMO and operated by Express Freighters Australia, was conducting a night freight flight from Perth Airport, Western Australia to Melbourne Airport, Victoria. The operating flight crew comprised of a captain and a first officer. There were no passengers. The aircraft departed Perth at 1206 Coordinated Universal Time (UTC)^[1] (2006 local time) and climbed to a cruising altitude of Flight Level (FL) 370.^[2]

At about 1453, the flight crew commenced their planned descent to Melbourne Airport. While descending through 8,000 ft, the AUTO FAIL and STANDBY annunciator lights illuminated on the pressurisation panel (see the section titled Pressurisation system and Figure 5). The flight crew reported feeling a rush of cold air and a soreness in their ears. The flight crew also observed the cabin differential pressure^[3]gauge reading zero and the main outflow valve position indication gauge reading approximately 50 per cent open.

The flight crew consulted the quick reference handbook (QRH)^[4]and conducted the non-normal checklist^[5]procedure for an ‘AUTO FAIL or Unscheduled Pressurization Change’ (Figure 1). The flight crew conducted the non-normal checklist up to and including step four and noted that the AUTO FAIL light had extinguished, but the cabin pressure was not controllable.

Figure 1: Non-normal checklist

Source: Express Freighter Australia, annotated by the ATSB

It was at about this time that Melbourne Air Traffic Control cleared the flight crew for approach to Melbourne Airport. The flight crew, noting that they were at 4,000 ft on descent and the aircraft was depressurised, elected to suspend any further troubleshooting on the pressurisation system. The aircraft landed without further incident at 1513 (0113 local time).

After landing, the captain contacted engineering control to discuss the occurrence and completed an entry in the aircraft’s technical log. Engineering personnel conducted internal and external inspections of the aircraft including the cargo compartments and the main outflow valve. The engineering inspections revealed multiple aft cargo bay insulation blankets were either missing (Figure 2), installed incorrectly or unsecured and one had been partially ejected from the main outflow valve (Figure 3).

Figure 2: Missing insulation blanket aft cargo bay

Source: Express Freighter Australia, annotated by the ATSB

Figure 3: Main outflow valve after occurrence
 

Source: Express Freighter Australia, annotated by the ATSB

A subsequent fleetwide inspection was carried out with unsecured or missing insulation blankets discovered in the aft cargo bays of four of the operator’s five Boeing 737 Special Freighter aircraft.

Context

Aircraft information

VH-XMO was originally manufactured as a passenger aircraft in the United States in 1987, and later converted for air freight operations. At the time of the occurrence the aircraft had accrued a total of 74,408 flight hours. Express Freighters Australia operated four 737-300 Special Freighters and one 737-400 Freighter.

Pressurisation system

The aircraft’s cabin is pressurised to maintain a safe and comfortable environment for the flight crew. This is achieved by the cabin pressure control system during all phases of operation. Engine bleed air is supplied to the cabin and the control system modulates a forward and a main outflow valve to maintain pressurisation within the aircraft’s cabin (Figure 4). The cabin pressurisation panel (Figure 5) allows the flight crew to set and monitor cabin pressurisation.

Figure 4: Main outflow valve fuselage location

Source: Express Freighter Australia, annotated by the ATSB

The main outflow valve is operated electrically and has three modes of control: AUTO, STANDBY, and MANUAL.

• AUTO, the flight crew set the planned cruising altitude along with the destination landing altitude. The cabin pressure control system then uses these parameters, along with ambient static pressure, to automatically control cabin pressurisation.
• STANDBY, the flight crew select a cabin altitude and the cabin rate of climb required for the flight.
• MANUAL, the flight crew directly control the opening and closing of the main outflow valve with a switch to control cabin pressurisation.

The AUTO FAIL annunciator light (Figure 5) illuminates on the panel if there is an AUTO pressurisation control failure. If that occurs, pressurisation control automatically transfers to the STANDBY mode and the green STANDBY annunciator illuminates. This occurs under any of the following conditions:

• loss of AUTO AC power
• excessive rate of cabin pressure change (greater than 1,890 sea level ft/min)
• excessive differential pressure
• high cabin altitude (13,875 ft).

Figure 5: Cabin pressurisation panel

Source: Express Freighter Australia, annotated by the ATSB

The main outflow valve incorporates an electrically heated gasket on its mounting surface to the fuselage skin. This gasket provides heat during flight at high altitudes to prevent the main outflow valve freezing in the closed position.

Non-normal checklist

Non-normal checklists are used by the flight crew to manage unexpected events. Illumination of a system annunciator light indicates a non-normal condition and are a cue to carry out the associated checklist.

The flight crew commenced the ‘AUTO FAIL or Unscheduled Pressurization Change’ non-normal checklist (Figure 1) as a response to the illumination of the AUTO FAIL annunciator. The objective of the checklist is to maintain control of the cabin altitude.

The flight crew completed the steps up to and including step four and determined that the cabin altitude was not controllable. The flight crew elected to discontinue the non-normal checklist at this stage due to the aircraft already having depressurised and being at 4,000 ft on the approach to land.

Post occurrence maintenance

After the occurrence, engineers conducted external and internal inspections of XMO. This included inspections of the aft cargo compartment and main outflow valve. A pressurisation control system test, a cabin depressurisation conditional inspection, and an automatic rate check pressurisation run were also carried out. As a result, the following unserviceability’s were recorded in the aircraft’s maintenance log:

• insulation blanket found liberated out of the outflow valve and jamming the valve
• insulation blankets in the aft cargo equipment area appear not to be secured as viewed via the outflow valve opening
• aft equipment bay insulation blankets found to be missing frame cap insulation blankets in several locations
• frame blankets installed over the top of frames in several locations
• one blanket missing from location, located in upper right aft corner of compartment
• one insulation blanket found damaged and partially missing forward of outflow valve
• strip cap frame blankets found adrift
• the main outflow valve heater gasket found unserviceable
• inspections carried out on insulation blankets and attachment stand-offs

Prior to XMO’s return to service, engineers replaced the main outflow valve assembly, the main outflow valve heater gasket, multiple insulation blankets, and multiple insulation blanket stand-offs. Correct orientation of the aft cargo bay insulation blankets was also carried out as required.

Engineering requirements

Over the service life of the Boeing 737 airframe, the Boeing Company have issued two service bulletins directly relating to the insulation blankets in the aft cargo compartment. These service bulletins were applicable to the Express Freighters Australia’s four 737-300 aircraft. They were not applicable to the 737-400 due to the bulletins being incorporated at the time of manufacture.

• 14 July 1988, Boeing service bulletin 737-25-1227 - Cargo compartments – cabin pressurisation outflow valve area insulation blanket replacement:
  • This service bulletin stated that six operators reported being unable to maintain cabin pressure due to a detached insulation blanket inhibiting pressurisation outflow valve operation. The incorporation of this service bulletin reduced the likelihood of loose insulation blankets interfering with the operation of the cabin pressurisation outflow valve by installing insulation retaining studs, replacing the insulation blankets with a heavier cover material, and using wider hook and loop (Velcro) fasteners. Express Freighters Australia confirmed that this service bulletin was applicable to the four Express Freighters Australia’s Boeing 737-300’s and was incorporated on all four aircraft between 1989 and 1991.
• 30 July 2008, Boeing service bulletin 737-25-1572 - Replace fuselage and environmental control system insulation blankets with new insulation blankets:
  • This service bulletin required the replacement of all of the fuselage insulation blankets and the environmental control system duct insulation blankets in the pressurised areas of the aircraft with new, less flammable blankets. On the 24 November 2008 the Civil Aviation Safety Authority (CASA) issued Airworthiness Directive AD/B737/348 for Australian operators to incorporate this Boeing service bulletin. This included the replacement of the blankets installed during service bulletin 737-25-1227 and was applicable to all four Express Freighters Australia’s Boeing 737-300s. This service bulletin was incorporated on XMO on the 01 August 2014.

Scheduled inspections

Aft cargo bay structural and component inspections are an ongoing requirement on the Boeing 737-300/400 aircraft. These scheduled inspections require the disturbance or removal (partial or complete) of insulation blankets. Instructions are included in these inspections to inspect the blankets for serviceability prior to reinstallation.

The following inspections were carried out on XMO prior to the occurrence:

• On 12 November 2016, an Aft Cargo Lower Lobe scheduled inspection was carried out by Singapore Technologies Engineering Aerospace. This inspection is scheduled for every 5 years or 8,000 flying hours.
• On 26 April 2020, a heavy maintenance check was carried out by Singapore Technologies Engineering Aerospace. This included multiple tasks in and around the aft cargo bay and rear pressure bulkhead. This also included an inspection of the aft cargo bay pressurisation components which consist of the main outflow valve and surrounds.

During these inspections there were no subsequent faults or non-routine tasks raised in the maintenance documentation for damaged or missing blankets.

The insulation blankets in the aft cargo bay are not readily accessible due to the interior lining and the cargo bay floor, therefore, removal of the lining and floor is required prior to access and inspection of the area. Following the heavy maintenance check in April 2020, and prior to the occurrence, there were no other maintenance activities documented within the aft cargo bay area that would necessitate the disturbance of the insulation blankets.

Safety analysis

The flight crew reported that passing through 8,000 ft on descent into Melbourne, the AUTO FAIL and STANDBY annunciator lights illuminated, and they felt a rush of cold air and a soreness in their ears. The flight crew also observed the cabin differential pressure gauge reading zero and the main outflow valve position indication gauge reading approximately 50 per cent open.

During descent, the main outflow valve modulates increasing cabin pressure as ambient air pressure increases. It is possible that the failure of the main outflow valve heater gasket, identified by engineering personnel after the occurrence, resulted in the valve freezing in the closed position. As the aircraft descended into warmer air, the valve may have thawed resulting in its sudden opening and observed ingestion of the unsecured insulation blanket.  

However, with the main outflow valve operation inhibited, it was unable to modulate allowing an unrestricted venting of pressurised cabin air. This resulted in the inability to control the aircraft’s cabin pressure reported by the flight crew.

One of four conditions is required to illuminate the AUTO FAIL annunciator on the pressurisation panel. The pilots reported a rush of air and a soreness in their ears, indicating that an ‘excessive rate of cabin pressure change (1,890 sea level ft/min)’ was probably detected by the cabin pressure control system resulting in the illumination of the AUTO FAIL annunciator. The STANDBY annunciator illuminated due to the pressurisation system unable to automatically operate and switching to the STANDBY mode of operation.

The ATSB assessed that the decision by the flight crew to suspend the non-normal checklist at 4,000 ft was appropriate due to the aircraft being already depressurised and on approach to land.

During post occurrence troubleshooting and inspections, the engineers discovered multiple aft cargo bay insulation blankets either installed incorrectly, damaged, unsecured or missing. This included an insulation blanket protruding from the main outflow valve, inhibiting its operation. As part of the post occurrence rectifications, the engineers correctly orientated and installed new blankets where required.

The incorrect installation of some of the blankets along with an inadequate inspection of their condition occurred at some point prior to the occurrence. A heavy maintenance check was carried out on 26 April 2020, 15 months prior to the occurrence. This inspection and maintenance activity included multiple tasks within the aft cargo bay which necessitated the disturbance and partial or complete removal of the insulation blankets. After this date, and prior to the occurrence on 6 July 2021, there was no documented evidence of maintenance carried out in the aft cargo bay which would require access or disturbance of the insulation blankets.

It is unlikely that undocumented maintenance occurred in this area due to the requirement to remove multiple interior linings and the cargo floor to gain access to the insulation blankets. Therefore, it is probable that during the heavy maintenance check that was completed on the 26 April 2020, the insulation blankets were installed incorrectly, and an inadequate area inspection was carried out.

Findings

From the evidence available, the following findings are made with respect to the unscheduled pressurisation change involving a Boeing 737, VH-XMO that occurred on 6 July 2021.

Contributing factors

• An insulation blanket in the aft cargo bay was incorrectly secured leading to it coming adrift and inhibiting the main outflow valve operation. This resulted in uncontrolled pressurisation changes in the aircraft.

Safety actions

Safety action by Express Freighters Australia

Following this occurrence Express Freighters Australia advised that an inspection of aft cargo compartment equipment and insulation blankets were conducted on the four other Boeing company 737 freighter aircraft within the Express Freighters fleet. Multiple unserviceabilities were discovered during this inspection and subsequently rectified.

As a result of the outcome of these inspections, Express Freighters Australia have implemented a 4,000 hour or 36 month insulation blanket visual inspection task for the B737-300 and a 4,000 hour or 18 month inspection task for the B737-400. 

Sources and submissions

Sources of information

The sources of information during the investigation included the:

• the operator (Express Freighters 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:

• the operator (Express Freighters Australia)
• Civil Aviation Safety Authority

No draft report 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 2022 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.  Coordinated Universal Time (UTC): Eastern Standard Time (EST) –10 hours. UTC has been used throughout this report as the flight crossed several Australian time zones.
2.  Flight level: at altitudes above 10,000 ft in Australia, an aircraft’s height above mean sea level is referred to as a flight level (FL). FL 370 equates to about 37,000 ft.
3.  Cabin differential pressure: the difference in pressure between inside the aircraft cabin and the local external atmosphere.
4.  Quick Reference Handbook (QRH): an approved document for the flight crew that contains procedures for non-normal and emergency conditions for use in flight.
5.  Non-normal checklist: A specific procedure to be carried out in response to a system annunciator light.

Safety summary

What happened

On 12 July 2021, a Bombardier DHC-8-402 (Q400), VH-QOY was being operated by QantasLink from Sydney to Albury, New South Wales and return. After take-off, the flight crew inadvertently omitted to retract the landing gear and did not identify this omission when completing the after-take-off checklist.

The cabin crew alerted the flight crew that the landing gear was still extended, and the flight crew retracted the landing gear when the aircraft was at about 15,900 ft. The aircraft had exceeded the maximum altitude with landing gear extended (15,000 ft).

What the ATSB found

The ATSB found that both pilots were heavily focused on aircraft performance after take-off, so the positive rate and subsequent gear-up calls were not made. Neither pilot identified these omissions.

When completing the after-take-off checklist, the pilot monitoring provided the ‘landing gear’ challenge and the pilot flying incorrectly called ‘up, no lights’ in response. Both pilots observed that the 3 green landing gear lights were illuminated but neither recognised that this was problematic for this stage of flight. It is likely that both pilots had a strong expectancy that the landing gear had been retracted after take-off.

What has been done as a result

QantasLink advised that both flight crew underwent additional training focused on threat and error management techniques. The occurrence was also included in a safety article, which discussed omissions, threat and error management and situational awareness.

QantasLink advised it had also initiated a program of focused risk monitoring for its operational ramp-up out of the COVID-19 pandemic. Metrics included human factors and performance, crewmember wellbeing, flight data and a return-to-work training program.  

Safety message

This occurrence demonstrates how diverted attention or focus may result in errors of omission, especially where a task may be reliant on standard verbal cues. Highly-repetitive, routine tasks may result in pilots developing strong expectations that a task has been completed, even if it has not been, and make it difficult for pilots to identify an omitted action. Accordingly, it is essential that when flight crews are completing checklists, they focus on confirming that the relevant conditions have been met.

The investigation

The occurrence

On the afternoon of 12 July 2021, a Bombardier DHC-8-402 (Q400), registered VH-QOY, was being operated by QantasLink on a scheduled passenger flight from Sydney to Albury, New South Wales, and return. The aircraft was crewed by 2 pilots and 2 cabin crew with 22 passengers on board.

The flight crew agreed that a take-off from intersection G on runway 34 left would be a more efficient option due to the planned light weight of the aircraft. The operator’s performance software calculations confirmed this assessment, and determined the use of flap 10 would be required if using that intersection. The flight crew stated that, although a flap 10 take-off was a normal operation, a flap 5 departure was the routine departure configuration.

The flight crew agreed that the first officer (FO) would be pilot flying (PF) and the captain would be pilot monitoring (PM) for the first sector.^[1] The FO decided that they would hand fly the aircraft up to 5,000 ft in order to refresh and maintain their handling skills.

During the take-off briefing, the flight crew noted that flap 10 had a lower maximum flap extended speed of 181 kt (compared to 200 kt for flap 5). The crew discussed that, as they expected an increase in aircraft performance due to the light weight of the aircraft and normal take-off power, there was an increased potential for exceeding the flap speed limit. When reviewing the maintenance log, the captain also noted that there was a deferred requirement for maintenance action related to propeller balance.

At about 1527 Eastern Standard Time,^[2] the aircraft took off. The captain recalled being very focused on the correct pitch attitude for take-off and monitoring the airspeed in relation to the flap speed limit. The captain stated that, although the FO adopted the correct pitch attitude on this occasion, they had previously observed some FOs pitch higher than stated in the operator’s procedures.

The captain explained that, by the time their scan was complete, the aircraft was at approximately 400 ft and fast approaching the first standard instrument departure turn at 600 ft. Their focus then switched towards this turn and anticipation of what the FO may require at that time. Although the captain confirmed that the aircraft had achieved a positive rate of climb, they inadvertently did not make the required ‘positive rate’ call to confirm to the FO that the aircraft was safely climbing.

The FO reported that they were also very focused on airspeed and maintaining runway centreline. As a result, the FO did not identify that the positive rate call had not been made. The FO also did not make the ‘gear up’ call, the next standard call in the flight crew’s after take-off procedural flow.

At this time both of the flight crew were still very focused on the aircraft’s performance, and neither pilot identified that the landing gear had not been retracted.

After passing 600 ft, the FO turned the aircraft onto the assigned heading (230°) and soon after the crew commenced the acceleration drills. At about 1,100 ft, the FO asked for the flaps to be raised, which the captain actioned. The crew selected an airspeed of 210 kt for the climb.

The FO recalled that, while still hand flying the aircraft, they called for the ‘after take-off’ checklist. The first item on the checklist was landing gear. Both pilots were required to look at the item to confirm the correct state (landing gear handle up and no lights illuminated). The captain recalled that they stated ‘landing gear’ and the FO responded ‘up, no lights’. Neither pilot detected that the landing gear was still down.

The captain later reported that, when completing the checklist, they saw the 3 green lights for the landing gear but did not recognise that the green lights indicated an abnormal situation for that stage of flight. At the time they thought it was a safe indication, and they noted that during the landing phase 3 green lights was the correct indication for the equivalent checklist item. The FO later reported that they could not specifically recall what they saw on the landing gear panel at that time.

The FO recalled that they engaged the autopilot when the after-take-off checklist was completed. Recorded flight data showed that the autopilot was engaged at 1530, when the aircraft was approaching 3,400 ft.

Following the after-take-off checklist, both pilots noted that the aircraft was noisier than normal with a vibration also noticeable. They also recalled that there did not seem to be a problem with the aircraft’s performance during the climb. The captain stated that the noise and vibration were uniform and not problematic but were somewhat distracting. The FO recalled thinking the flap and/or gear may still be extended. The FO recalled looking at the landing gear panel at this time and noticing the 3 green lights, but they did not recognise that this was an abnormal indication at that stage of flight. They recalled consciously looking for red lights that may indicate a problem and not seeing any.

The captain advised the FO that the noise and vibration was probably related to the propeller balance maintenance log entry. The captain suggested reducing climb speed, as this would normally reduce such noise and vibration. The FO agreed with this action, although they recalled thinking that the noise was not what they would have expected from a propeller balance issue. At 1531, as the aircraft was climbing through 6,000 ft, the crew selected a speed of 185 kt. This action reduced the noise and vibration, which appeared to reinforce the crew’s assessment of the source of the problem.

Later during the climb, the FO provided the after-take-off public address announcement to the passengers.^[3] The cabin crew then contacted the flight crew, asking if it was normal for the landing gear to still be extended at that time. The flight crew immediately looked at the landing gear panel and identified that the handle was down with 3 green lights illuminated, indicating that the landing gear was still extended.

The flight crew confirmed that the aircraft ‘s speed was below the maximum landing gear operating speed (200 kt) and then, at 1536, they retracted the landing gear. They believed no landing gear speed limits had been exceeded, but when they retracted the landing gear the aircraft was at about 15,900 ft. This was above the 15,000 ft altitude limit for flight with gear extended.

Upon raising the landing gear, the flight crew noted that the unusual noise and vibration stopped. The captain later reported that, in hindsight, the unusual but uniform noise and vibration were not consistent with what would normally be expected with a propellor balance problem, but they had not considered other explanations for the noise and vibration at the time it was noticed.

The aircraft continued climbing to the assigned level (FL 240). After reaching their cruise altitude, the flight crew contacted the operator’s maintenance watch department to advice that they had exceeded the maximum landing gear extended altitude. The crew were subsequently directed to conduct a precautionary return to Sydney. The aircraft landed back in Sydney at 1630 without further incident.

Context

Personnel information

Captain

The captain had 15,870 hours flight time, of which 12,280 hours was on DHC-8 aircraft. Their flying career included flying various single and multi-engine aircraft in general aviation prior to joining QantasLink. They had been a captain on the Q400 since 2013. Their last recurrent proficiency (simulator) check was conducted on 2–3 July 2021.

The captain reported having 6 hours sleep the previous night and 13 hours sleep in the previous 48 hours. They indicated that they did not feel tired or fatigued when they signed on for duty. The flight from Sydney to Albury was their first flight that day.

First officer

The first officer (FO) had about 1,470 hours flight time, with about 1,215 hours on the Q400. They had commenced their flying career with QantasLink. Their last recurrent proficiency (simulator) check was conducted on 29–30 June 2021.

FO reported having 8 hours sleep the previous night and 16 hours sleep in the previous 48 hours. However, the FO noted that their sleep the night before the flight was broken and they felt a little tired during the commute to work. This improved after drinking coffee and becoming more engaged with required tasks during the flight planning stage. The FO reported not feeling tired during the flight. The flight from Sydney to Albury was their first flight that day.

Recent flight experience

The operator stated that its recurrent proficiency checks from early 2020 to August 2021 included ‘a focus on the methodical actioning of checks and checklists to capture Flight Crew that may have not flown as regularly due to the impact of COVID19’.

The operator also advised that its flight crew had experienced a reduction in flight hours since March 2020 due to the impact of COVID on the aviation industry. Prior to March 2020, captains averaged 40–50 hours flight hours per month and FOs averaged 50–60 hours per month.

With regard to the crew of VH–QOY, the captain conducted 41 hours per month since March 2020 and the FO conducted 36 hours per month. Prior to the occurrence flight on 12 July 2021, the captain conducted 28 hours line flying (29 sectors) in the previous 30 days and 127 hours in the previous 90 days (excluding the proficiency check). Their last flight was on 7 July (5 days before). The FO conducted 20 hours line flying (22 sectors) in the previous 30 days and 47 hours in the previous 90 days (excluding the proficiency check). Their last flight was on 1 July (11 days before). Both pilots met the minimum regulatory requirements for currency.^[4]

The operator noted that it had allocated safety pilots to join a flight crew in situations where a pilot had not flown for a significant period of time. A safety pilot had not been allocated, nor was it required, for the occurrence flight.

Aircraft information

General

The Bombardier DHC-8-402 (Q400) was a was a twin turbo propeller aircraft capable of carrying up to 80 passengers. VH-QOY was manufactured in 2010. The Q400 was normally operated with 2 flight crew and 2 cabin crew.

Landing gear system

The aircraft was equipped with a retractable tricycle landing gear. Operation of the landing gear was via a handle on the main instrument panel on the flight deck, in front of the FO’s seat. The handle illuminated when the landing gear was transiting between positions, and 3 green lights on the landing gear advisory panel indicated the gear was down and locked (Figure 1). When the gear was fully retracted (in the ‘up’ position), there were no lights illuminated.

Figure 1: Landing gear handle and advisory panel

Source: QantasLink, annotated by the ATSB

Relevant aircraft limits included:

• maximum landing gear operating speed – 200 kt (the landing gear cannot be raised or extended when operating above this speed)
• maximum landing gear extended speed – 215 kt (the aircraft cannot be flown above this speed with the landing gear extended)
• maximum altitude with landing gear or flaps extended – 15,000 ft.

A review of the recorded data for the flight confirmed that neither of the landing gear speed limits were exceeded. However, the landing gear remained extended above 15,000 ft and was retracted at about 15,900 ft.

The operator advised that, although the maximum landing gear altitude limit was exceeded, the aircraft maintenance manual did not require an inspection of the landing gear. The manufacturer confirmed that an inspection of the landing gear was not required. It stated that the 15,000 ft altitude limit was a practical value, not a limiting value. It was based on the normal envelope of landing gear operation. It advised that the aircraft had been demonstrated above 15,000 ft with the gear extended and confirmed the gear would operate normally above 15,000 ft. As such, no inspection was required.

Flight crew procedures

Landing gear retraction after take-off

The landing gear is normally retracted shortly after take-off. As per the operator’s procedures, this was normally triggered by the pilot monitoring (PM) who, after take-off, observed the radio altimeter and vertical speed indicator to confirm that the aircraft had a positive rate of climb. They then called ‘positive rate’, which triggered the pilot flying (PF) to confirm this and then call ‘gear up’. The PM then selected the landing gear handle up and confirmed the advisory lights were extinguished (Figure 2).

There was no hardware system to alert the flight crew that the gear has not been retracted after take-off, nor was such a system required to be installed.

Figure 2: Take-off procedure extract

Source: QantasLink, annotated by the ATSB

After-take-off checklist

As per the operator’s procedures, the PF called for the after-take-off checks (part of the normal checklist) at their discretion. This was normally done once the aircraft had been safely established in the climb. The checklist was preceded by checks that were completed from memory by the PM.

The checklist was a challenge and response type, with the first item being the landing gear. The PM ‘challenges’ the PF by calling ‘landing gear’ and both pilots were then required to observe the landing gear advisory panel to ensure that the handle was up and all lights were extinguished. Once verified, the PF ‘responds’ by calling ‘up, no lights’ (Figure 3).

Figure 3: After-take-off checklist extract

Source: QantasLink  

The Q400 flight crew operating manual stated that:

The purpose of the [normal] checklist is to confirm critical items have been actioned and/or confirmed appropriate for the phase of flight. Though the response is by one crew member it is the responsibility of both crew members to clearly communicate if an item has not been actioned or is incorrect.

Related occurrences

A review of the ATSB aviation occurrence database identified 7 other occurrences during January 2011 to December 2021 on commercial air transport flights involving Australian domestic airline operators where a flight crew inadvertently did not raise the landing gear after take-off. Details of these occurrences included:

• Soon after take-off, a DHC-8 struck a bird, which resulted in the flight crew inadvertently not announcing the ‘gear up’ call and retracting the landing gear at the normal time. The after-take-off checklist was completed but the crew did not identify that the landing gear was still down. The crew subsequently detected the problem and realised the aircraft had exceeded the maximum landing gear extended speed.
• Soon after take-off, an auxiliary power unit caution light on a DHC-8 illuminated, which distracted the flight crew’s attention and they did not raise the landing gear at the normal time. The after-take-off checklist was completed but the crew did not identify that the landing gear was still down. On climbing through the transition altitude (10,000 ft), they noted the aircraft had poor climb performance and identified the problem. The landing gear was retracted when the aircraft was above the maximum landing gear operating speed.
• Soon after take-off, the flight crew of a DHC-8 were distracted by another aircraft in the circuit, which turned unexpectedly towards them. The crew omitted the ‘gear up’ call and did not raise the landing gear at the normal time. The problem was subsequently identified, and no speed limits were exceeded.
• Shortly after take-off, the PF of a Saab 340 made the operator’s standard ‘positive rate, gear up’ call. The PM did not recall hearing this call, which meant that the landing gear was inadvertently not retracted and the PM did not make the standard call in response (‘selected’). The crew detected their error when conducting the climb checklist (after the aircraft passed through 3,800 ft). They instinctively retracted the gear; however, at that time the aircraft was above the maximum landing gear retraction speed. Factors that influenced this omission and its non-detection included both crew focusing on departure procedures and the local weather, and the crew likely expecting that the landing gear was retracted as normal. In addition, the PM was experiencing a level of fatigue (AO-2014-189).
• The flight crew of an A320 incorrectly calculated the aircraft’s take-off speeds, which resulted in the airspeed exceeding the flap limit speed soon after take-off and the crew did not retract the landing gear after achieving a positive rate of climb. While troubleshooting a buffeting sound, the PF found that the landing gear was still extended and called ‘gear up’, and the landing gear was retracted when the aircraft was above the maximum landing gear retraction speed (AO-2018-067).
• The flight crew of an A320 were following another A320 on departure and the crew elected to conduct a TOGA take-off to avoid wake turbulence. After take-off, the crew omitted the ‘gear up’ call and did not raise the landing gear. The problem was identified after the flaps were retracted, and the landing gear was retracted when the aircraft was above the maximum landing gear retraction speed.
• Soon after take-off, the PM of a DHC-8 delayed the ‘positive rate’ due to distractions associated with radio noise. Following the call, the PF made the ‘gear up’ call, but the PM retracted the flaps instead, and the crew did not identify the error at that time. After reaching 7,000 ft, the crew commenced the after-take-off checklist and identified that the landing gear was still down. The landing gear was retracted when the aircraft was above the maximum landing gear operating speed.

In addition to the 8 occurrences where the landing gear was not retracted soon after take-off (including the 12 July 2021 occurrence), there were also 6 other occurrences where a flight crew inadvertently raised the flaps instead of the landing gear after take-off. These occurrences generally did not result in the landing gear remaining extended for a significant period of time.

Overall, for these 2 types of incorrect configuration after take-off, 11 occurrences took place between January 2011 and March 2020 and 3 took place between April 2020 and December 2021. Between April 2020 and December 2021, the number of domestic airline departures in Australia was significantly reduced (by 55%) compared to previous years. Although the rate of the incorrect landing gear configuration after take-off occurrences was 1.9 per million departures from January 2011 to March 2020 and 6.1 per million departures during April 2020 to December 2021, this difference was not statistically significant.^[5]    

The ATSB also reviewed the Aviation Safety Reporting System (ASRS) database for occurrences in the United States between January 2011 and October 2021 where an airline flight crew did not retract the landing gear after take-off. There were 8 reports. Some key details included:

• In most cases, the flight crew reported some form of distraction or additional workload after take-off.
• In most cases, the flight crew also reported noise, vibration or buffeting and in some cases decreased climb performance and/or increased fuel burn associated with the landing gear being down.
• In 2 cases, the flight crews reported detecting the problem when the after-take-off checklist was completed, whereas in 5 cases the crew reported the after-take-off checklist was completed but they did not detect the problem.
• Several crews reported being unable to identify the reason for the problem even after extensive troubleshooting, including 3 cases where the landing gear being down was only detected when the crew went to select the landing gear down for landing. One pilot noted that during troubleshooting everything ‘appeared normal and all symbols were green’. Another pilot noted that this type of occurrence (not raising the landing gear) was so rare that their crew, which was very experienced, did not even consider it when troubleshooting.

Safety analysis

Landing gear not retracted after take-off

After taking off from Sydney, the flight crew did not raise the landing gear. Ultimately the problem was not identified until the aircraft had reached 15,000 ft. No landing gear speed limits were exceeded. Although the 15,000 ft maximum altitude for operating with the landing gear extended was exceeded, this had no subsequent effect on the serviceability of the aircraft. Nevertheless, the occurrence was of some concern as the flight crew did not identify the incorrect configuration for an extended period of time.

Conducting a take-off is a specialised task that is acquired through comprehensive training and significant experience; it involves conducting routine, frequently-practiced tasks in a largely automatic manner with occasional conscious checks on performance. Errors, known as slips and lapses, will occasionally occur when conducting such skill-based tasks (Reason 1990). Omitting a step or an action is one of the most common forms of error (Reason 2002), and they are often associated with interruptions, distractions or attention being diverted to other tasks. Accordingly, occurrences where flight crew forget to raise the landing gear after take-off almost always involve some form of distraction or diverted attention.

In this case, both pilots were heavily focused on the aircraft’s speed soon during the initial climb. Additionally, the captain was focused on the aircraft’s pitch attitude, having previously observed other pilots pitch higher than the operator’s procedures stated in similar situations, and the workload of the first officer (FO) was increased while hand flying the aircraft.

The confirmation of positive rate and subsequent call was a frequently-practiced action for the crew and therefore one normally conducted automatically, with little conscious oversight. Their diverted focus of attention was probably sufficient to result in the omission of the positive-rate call and neither pilot identifying that it had not been made.

Not calling ‘positive rate’ removed the standard verbal cue for the FO to call ‘gear up’, increasing the likelihood that the gear-up call would not be made. At this time, both pilots were still focused on aircraft performance but also shifting their focus to the increasing workload of the standard instrument departure.  

Misidentification of landing gear status

Errors of omission are often difficult to detect by the people who make them (Sarter and Harrison 2000), and the absence of something (such as an action) is harder to detect than the presence of something (Wickens and others 2013). Trying to recall from memory whether actions have already been completed is also vulnerable to source memory confusion, such that the current situation is confused with the memory of many previous occasions when the action was successfully done (Dismukes and others 2007). Accordingly checklists, like the challenge and response after-take-off checklist, perform a vital role in ensuring omissions in a flight crew’s procedural flow are captured (Barshi and others 2016).

When actioning the after-take-off checklist on this occasion, the ‘landing gear’ item was called by the captain (as pilot monitoring) but neither pilot identified the problem. The captain observed that the 3 green landing gear lights were illuminated, but did not identify it as being problematic for that stage of flight. Although the FO could not recall what they saw at that stage, they also recalled seeing the 3 green lights later in the flight and not thinking it was problematic.

Expectations strongly influence where a person will search for information and what they will search for (Wickens and McCarley 2008), and they also influence the perception of information (Wickens and others 2013). Pilots frequently conduct the task of raising the landing gear and, in almost all cases, it is done successfully. Consequently, the pilots on this occasion had a strong expectancy that the landing gear had been retracted, and they probably conducted the after-take-off checklist with a high degree of automaticity, rather than consciously looking for what was required (that is, no green lights). Although the 3 green lights provided a clearly visible indication that the landing gear was still down, green lights are often associated with something being in a safe state; therefore this cue can be interpreted incorrectly in this phase of flight if the flight crew’s attention is not focussed on exactly what they are looking for.

This type of occurrence is very rare but, when it occurs, a flight crew’s further troubleshooting of symptoms associated with the landing gear still being down is often not effective. In this case, both pilots reported observing noise and vibration from the aircraft. After some discussion, the crew associated the noise and vibration with the propeller balance maintenance log entry. In an effort to reduce the noise and vibration, the crew reduced the climb speed. This reduced the abnormal indications and seemingly confirmed that the propeller balance was the source of the problem, consistent with the effects of confirmation bias.

Awareness displayed by cabin crew

The cabin crew displayed a high level of vigilance regarding the aircraft state. Their willingness to bring this to the attention of the flight crew allowed the flight crew to identify the problem and retract the landing gear as soon as possible and highlights the strength of timely communications between crew members.

Reduced flying activity

Skill decay or skill degradation refers to the loss of trained or acquired skills or knowledge following periods of non-use (Arthur and others 1998). Skill decay increases as the retention interval (or time since learning) increases, and it also increases depending on the quantity and quality of the initial and recurrent training and the amount of on-the-job exposure (Arthur and others 1998, Sanli and others 2018, Vlasblom and others 2020).

Procedural skills involving the retrieval and application of step-by-step actions are more sensitive to skill decay than many other types of activities (Goodwin 2006, Stothard and Nicholson 2001, Wisher and others 1999). As recently noted by the European Union Aviation Safety Agency (EASA 2021):

Procedural tasks that require specific procedural or declarative knowledge (e.g. checklists that require more items than prescribed on paper) may be more susceptible to skill decay than higher order cognitive tasks (e.g. decision making) or perceptual/psychomotor tasks. Cognitive shortcuts for procedures decay rapidly, requiring a significant increase in cognitive resources, in particular for procedures that are normally routine. By their prescriptive nature, procedures are easily subject to slips and lapses. Procedures must be viewed as highly sensitive to proficiency decay.

In this case, the FO had undertaken less than the operator’s normal amount of flying since March 2020. In particular, the FO had conducted less than one third of their normal amount of flying in the previous 90 days and had not conducted any flights for 11 days. However, the operator was aware of the potential issues associated with reduced flight recency and had introduced measures to mitigate the risk. Both flight crew had recently undertaken a proficiency check. Overall, there was insufficient evidence to conclude that the FO’s reduced flight recency contributed to the procedural errors made by the flight crew on this occasion.

Findings

From the evidence available, the following findings are made with respect to the incorrect configuration involving Bombardier DHC-8-402, registered VH-QOY, near Sydney Airport, New South Wales on 12 July 2021.

Contributing factors

• After take-off, the flight crew’s attention was heavily focused on maintaining the aircraft's speed and pitch, resulting in the omission of the ‘positive rate’ call. This removed a trigger for the ‘gear up ‘call, which neither pilot identified, and subsequently the landing gear was not retracted after take-off.
• Although the landing gear handle was in the down position and 3 green lights were illuminated, the pilot flying incorrectly called 'up, no lights' when conducting the after-take-off checklist. The pilot monitoring did not identify the error. It is likely that both pilots had a strong expectancy that the landing gear had been retracted after take-off when completing the checklist.

Other findings

• The cabin crew observed that the landing gear remained extended longer than normal following take-off. They advised the flight crew, resulting in the landing gear being retracted.

Safety actions

Safety action by QantasLink

QantasLink advised that both flight crew underwent additional simulator and human factors training, which was focused on threat and error management techniques.

QantasLink published an article describing this occurrence, which discussed omissions, threat and error management and situational awareness.

QantasLink advised it had also initiated a program of focused risk monitoring for its operational ramp-up out of the COVID-19 pandemic. Metrics included human factors and performance, crewmember wellbeing, flight data and a return-to-work training program.

Sources and submissions

Sources of information

The sources of information during the investigation included:

• the flight crew of VH-QOY
• QantasLink
• Bombardier
• Airservices Australia.

References

Arthur W, Bennett W, Stanush PL and McNelly TL (1998) ‘Factors that influence skill decay and retention: A quantitative review and analysis’, Human Performance, 11:57-101.

Barshi I, Mauro R, Degani A and Loukopoulou L (2016) Designing flightdeck procedures, NASA Technical Memorandum NASA/TM—2016–219421.

Dismukes RK, Berman BA and Loukopoulos LD (2007) The limits of expertise: Rethinking pilot error and the causes of airline accidents, Ashgate Aldershot UK.

European Union Agency Safety Agency (2021), Safety issue report – Skills and knowledge degradation due to lack of recent practice, downloaded from www.easa.europa.eu.

Goodwin GA (2006) The training, retention, and assessment of digital skills: A review and integration of the literature, Research Report 1864, U.S. Army Research Institute for the Behavioral and Social Sciences.

Reason J (2002) ‘Combating omission errors through task analysis and good reminders’, BMJ Quality & Safety, 11:40–44.

Sanli EA and Carnahan H (2018) ‘Long-term retention of skills in multi-day training contexts: A review of the literature’, Industrial Journal of Ergonomics, 66:10–17.

Sarter NB and Alexander HM (2000) 'Error types and related error detection mechanisms in the aviation domain: An analysis of aviation safety reporting system incident reports', The International Journal of Aviation Psychology, 10:189–206.

Stothard C and Nicholson R (2001), Skill acquisition and retention in training: DSTO support to the army ammunition study, Defence Science and Technology Organisation, report DSTO-CR-0218.

Vlasblom JID, Pennings HJM, Van der Pal J and Oprins EAPB (2020) ‘Competence retention in safety-critical professions: A systematic literature review’, Educational Research Review, 30: 10.1016.

Wickens CD, Hollands JG, Banbury S and Parasuraman R (2013), Engineering psychology and human performance, 4th edition, Pearson Boston, MA.

Wickens, CD and McCarley, JS (2008), Applied attention theory, CRC Press, Boca Raton, FL.

Wisher RA, Sabol MA and Ellis JA (1999) Staying sharp: Retention of military knowledge and skills, US Army Research Institute, Special Report 39.

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:

• the flight crew of VH-QOY
• QantasLink (operator)
• the Civil Aviation Safety Authority
• Transportation Safety Board of Canada

Submissions were received from the operator and the captain. 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 2022 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this publication is licensed under a Creative Commons Attribution 3.0 Australia licence. Creative Commons Attribution 3.0 Australia Licence is a standard form licence agreement that allows you to copy, distribute, transmit and adapt this publication provided that you attribute the work. The ATSB’s preference is that you attribute this publication (and any material sourced from it) using the following wording: Source: Australian Transport Safety Bureau Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

__________

1. Pilot flying (PF) and pilot monitoring (PM) are procedurally assigned roles with specifically assigned duties at specific stages of a flight. The PF does most of the flying, except in defined circumstances, such as planning for descent, approach and landing. The PM carries out support duties and monitors the PF’s actions and aircraft flight path.
2. Eastern Standard Time (EST) was Coordinated Universal Time (UTC) + 10 hours.
3. This announcement was required to be conducted after passing the transition altitude (10,000 ft) or when the aircraft was established in the cruise. The provision of the announcement cancelled the sterile flight deck with the cabin crew.
4. Civil Aviation Safety Regulation Part 61 required that, within the previous 90 days of a flight, a pilot had conducted at least 3 take-offs and landings or completed a proficiency check. Additional requirements existed for the conduct of instrument approaches.
5. Calculated using Fisher’s exact test (p = 0.09). [Note: the number of departures for December 2021 was not available and was estimated using previous months. Figures will be adjusted if required in the final report.]

During a maintenance inspection of a Pacific Aerospace CT4/B aircraft, registered VH-YCR, a significant crack was identified within the rear spar of the aircraft’s left wing. The Civil Aviation Safety Authority (CASA) requested technical assistance from the ATSB in the metallurgical examination of the spar. To facilitate this assistance, the ATSB initiated an external investigation under the provisions of the Transport Safety Investigation Act 2003.

The ATSB has concluded the metallurgical examination of the rear spar and provided the results of that work to CASA on 4 May 2022.

CASA is responsible for, and will administer, the release of any information from the ATSB’s examination. Any enquires relating to the matter should be directed to CASA at www.casa.gov.au.

Source: CASA

_____________

The information contained in this update is released in accordance with section 25 of the Transport Safety Investigation Act 2003 and is derived from the initial investigation of the occurrence. Readers are cautioned that new evidence will become available as the investigation progresses that will enhance the ATSB's understanding of the accident as outlined in this update. As such, no analysis or findings are included.

Safety summary

What happened

On the afternoon of 6 July 2021, a Saab 340B aircraft, registered VH-ZLJ, departed Perth Airport for a scheduled passenger flight to Albany, Western Australia, with two flight crew, one cabin crew, and 16 passengers on board.

At about 1623, while climbing through an altitude of 6,000 ft, and before entering icing conditions, the crew activated the engine anti-ice and wing de-ice systems. Shortly afterwards, caution lights illuminated indicating a fault in the de-ice system. After levelling off at 7,000 ft, the crew actioned the relevant abnormal checklist, but the caution lights remained on. In response, the flight crew requested a descent to 5,000 ft to exit icing conditions and decided to return to Perth.

During the descent, the captain handed over control of the aircraft as pilot flying to the first officer. Over the next 2.5 minutes, air traffic control (ATC) communicated with the flight crew on multiple occasions, issuing a series of vectors and requesting flight information. As the aircraft was levelling off at 5,000 ft under autopilot control, ATC instructed the crew to make a right turn. About 20 seconds after beginning the turn, the aircraft’s aerodynamic stall warning stick shakers activated. The first officer initiated the stall recovery procedure before the captain took control as pilot flying to complete the recovery. The aircraft returned to Perth, landing at 1642.

What the ATSB found

The ATSB found that the aircraft’s right wing inboard de-ice boot probably delaminated shortly before encountering icing conditions, triggering the de-ice system fault that led to the flight crew’s decision to return to Perth.

During the return, the pilot flying became task saturated due to high workload and did not notice the aircraft’s reducing airspeed, which was also missed by the pilot monitoring due to a focus on other tasks until the stick shaker activated. The crew managed the recovery from the potential stall condition effectively and the aircraft returned safely to Perth.

What has been done as a result

The aircraft operator has amended flight crew training simulator sessions and related training material to include flight at minimum manoeuvring speeds – minimum airspeeds that provide a margin above a stall during aircraft manoeuvring.

Safety message

This stick shaker activation incident highlights that during periods of high workload, where there is an increased chance of making mistakes or errors, flight crews should prioritise monitoring critical flight parameters. Effective communication can help flight crew recognise a situation when their workload is becoming overwhelming, and consequently better manage the situation – for instance, giving themselves more time to complete the required tasks by discontinuing an approach, or deferring air traffic control requests appropriately.

The investigation

The occurrence

At 1619 Western Standard Time^[1] on 6 July 2021, a Saab 340B aircraft, registered VH-ZLJ (Figure 1) and operated by Regional Express, departed Perth Airport for a scheduled passenger flight to Albany, Western Australia, carrying 16 passengers. The crew comprised the captain (pilot flying),^[2] the first officer (pilot monitoring) and one cabin crewmember.

Figure 1: VH-ZLJ

Source: Supplied

At about 1623, the aircraft was climbing through 6,000 ft above mean sea level (AMSL) with the autopilot engaged. Consistent with operator procedures, the flight crew activated the engine anti‑ice and wing de-ice systems as the aircraft was entering icing conditions. Shortly afterwards, an ice protection master caution light, and TIMER light illuminated, indicating a fault in the de-ice system (see the section titled Airframe de-ice system). Soon after, the flight crew requested, and received, a clearance from air traffic control (ATC) to level off at 7,000 ft to avoid icing conditions. They then began actioning the TIMER light on abnormal checklist.

At 1629 the flight crew requested, and received, a diversion from ATC to avoid storm cells. Despite the crew actioning the checklist, the TIMER light remained on during the wing inboard de‑ice boot cycle, so at 1630 they commenced a descent to 5,000 ft to exit icing conditions. Shortly afterwards, the crew notified ATC that they would return to Perth.

During the subsequent descent, the captain handed over control of the aircraft as pilot flying to the first officer. Over the next 2.5 minutes, ATC communicated with the flight crew on eight separate occasions, issuing a series of radar vectors and flight information requests. At about 1632, the aircraft was levelling off at 5,000 ft with the autopilot engaged when ATC instructed the crew to make a right turn. About 20 seconds after beginning the turn, the aircraft’s aerodynamic stall warning stick shakers activated (see the section titled Stall warning system). The first officer initiated the stall recovery procedure before the captain took control as pilot flying to complete the recovery.

The aircraft returned to Perth without further incident, landing at 1642.

Context

Flight crew

The captain held an Air Transport Pilot Licence (Aeroplane), and had 4,695 hours of flying experience, of which over 4,474 hours were on the Saab 340B. The captain made the following comments and observations about the incident.

• The decision to return to Perth was based on the de‑ice system fault indication, the low freezing level along the route and no engineering support at the destination (Albany).
• As pilot monitoring during the descent from 7,000 ft to 5,000 ft, the captain notified the operator and communicated with the cabin crew and passengers regarding preparation for the return to Perth.
• During the descent their workload was relatively high, but the captain recalled feeling comfortable handling it at the time. However, in hindsight the captain felt that they did not monitor the first officer appropriately or focus enough attention on the aircraft’s airspeed and engine power levels.

The first officer held a Commercial Pilot Licence (Aeroplane), and had 605 hours of flying experience, of which over 386 hours were on the Saab 340B. The first officer made the following comments and observations about the incident.

• After taking over as pilot flying, the first officer became overly focussed on complying with ATC clearances since:
  • the captain was occupied and not available to perform standard clearance cross-checks
  • they were aware of previous instances of flight crew’s deviating from clearances because only one pilot had acknowledged them.
• The focus on ATC clearances, in combination with managing related communications, caused the first officer to feel ‘task saturated’^[3] and that the workload was ‘very high’.
• The first officer believed that the task saturation resulted in their instrument scan breaking down and not paying sufficient attention to the airspeed and engine power settings.
• The flight crew could have better managed the workload by giving themselves more time to complete tasks.

The ATSB found no evidence to indicate either flight crew were experiencing a level of fatigue known to affect performance.

Meteorological information

The relevant graphical area forecast indicated a freezing level of about 8,000 ft at Perth, reducing to about 6,000 ft at Albany. Moderate showers with broken^[4] cloud from 2,500 ft to above 10,000 ft, severe icing above 6,000 ft, and moderate turbulence below 7,000 ft were also forecast.

The flight crew reported rain and storm cells in the Perth Airport area and encountered occasional cloud at 7,000 ft, where the outside air temperature was 4° C (icing conditions).

Airframe de-ice system

The aircraft’s airframe de-icing system consisted of inflatable boots, located on the leading edges of the vertical and horizontal stabiliser, and the inboard and outboard section of the wings. The boots were rapidly inflated, using engine bleed air, to crack any accumulated ice and then deflated. A timer control unit regulated the boot inflation cycles in a sequence – stabiliser, outboard wing, and then inboard wing. Sensors monitored the boot cycles, and a TIMER light would illuminate together with an ice protection master caution light if a fault was detected with the de‑ice system.

Post-flight examination of the de-ice system revealed a delamination in the right wing inboard de‑ice boot on the underside of the wing (Figure 2). The aircraft operator stated that this was probably caused by an internal stitching failure, resulting in the loss of system pressure. The boot was replaced and the de-ice system was subsequently tested serviceable. The ATSB reviewed the aircraft’s maintenance history related to the de-ice system and found it compliant with the required maintenance.

Figure 2: Right wing inboard boot delamination

Source: Operator, annotated by ATSB

Stall warning system

The stall warning and identification system fitted to the Saab 340B included:

• two independent stall warning computers
• two angle of attack (AOA) sensors – one mounted on each side of the fuselage
• stick shaker device on each control column that provided a physical warning of an impending aerodynamic stall in the form of vibrations and aural clacker sound when activated
• stick pusher device that applied forward force to the control column to reduce aircraft AOA when a stall condition was identified.

The AOA activation level for the aircraft’s stall warning system was dependent on flap position, engine anti-ice operation, and airframe de-ice operation. Flight in icing conditions required the operation of engine anti-ice and boot de-ice systems. Accumulation of ice and/or operation of the wing boot de-ice system alters the stall characteristics of the wings. This was compensated for by the stall warning system activating the stick shaker at a lower AOA (earlier) when the aircraft was configured for flight in icing conditions. The stick pusher AOA activation was unchanged. There was no indication in the cockpit of the AOA, but if an AOA sensor activation level was exceeded, the stick shaker and aural clackers activated, and the autopilot (if engaged) disengaged. If the AOA sensor values increased further, the stick pusher was activated.

Post-flight testing revealed that the left AOA sensor triggered the stall warning systems about 1 or 2° earlier than the specified parameters. The left AOA sensor was replaced, and the stall warning system was successfully tested. The ATSB identified that the aircraft’s maintenance history related to the stall warning system was compliant with the required maintenance.

Recorded data

Data from the aircraft’s flight data recorder was downloaded, but the aircraft manufacturer identified that some data parameters (AOA and elevator angle) had been corrupted.

Figure 3 shows verified flight data for certain recorded parameters at the time of the incident. The data shows that during level flight at 7,000 ft, engine power was at about 60% torque. During the descent to 5,000 ft, engine power was reduced by the flight crew to about 15% torque. After levelling off at 5,000 ft, engine power remained at that level, however the aircraft’s pitch angle increased, while the airspeed decreased. The pitch increase and airspeed reduction continued after the turn was commenced. About 20 seconds later, the autopilot disconnected, and the flight crew initiated the stall recovery.^[5] Altitude lost during recovery was about 450 ft.

The investigation could not determine which AOA sensor triggered the stick shaker, but the flight data was consistent with the aircraft approaching a stalled condition, so the early triggering of the left AOA sensor did not affect the outcome.

Figure 3: VH-ZLJ flight data

Source: ATSB

Similar occurrences

In 2013, an ATSB research report^[6] identified that 245 stall warnings in high capacity aircraft had been reported between 2008 and 2012 in Australia. Almost all of those were low risk events of momentary duration and were responded to promptly and effectively by the flight crew to maintain control of the aircraft. However, there were also several higher risk incidents where stick shaker activation occurred on approach to land when aircraft were in a low speed, high AOA configuration. In these cases, the risk of a stall developing was increased by a lack of awareness of decreasing airspeed and increasing AOA prior to the stall warning, probably due to increased flight crew workload during this phase of flight. None of the reported occurrences resulted in an actual stall.

Safety analysis

Shortly after the aircraft departed Perth, the flight crew received cockpit indications of a de-ice system fault. The fault was probably triggered by a delaminated de-ice boot on the underside of the right wing that failed during its inflation cycle shortly before encountering icing conditions. With a low freezing level and forecast icing conditions along the planned route, the de-ice fault indication led to the flight crew’s decision to return to Perth.

While manoeuvring towards Perth, engine power was reduced to descend the aircraft from 7,000 ft to 5,000 ft. However, when the aircraft levelled off at 5,000 ft, engine power was not increased. Consequently, as the airspeed reduced due to the low engine power, the autopilot maintained the selected altitude (5,000 ft) by pitching the nose up, increasing the AOA and reducing the airspeed further. This condition went unnoticed until one of the AOA sensors reached the level required for stick shaker activation.

The first officer, the pilot flying in the time leading up to the stick shaker activation, probably became task saturated while managing flying tasks (changing flight state to descend, level flight and turn) as well as communicating with, and following ATC instructions. This task saturation reduced the attention that the first officer paid to managing the airspeed and engine power. At that time, the captain (pilot monitoring) was also not monitoring these key parameters due to a focus on other communication tasks related to the return to Perth. The crew’s reduced awareness of airspeed resulted in the potential stall going unnoticed until the stick shaker activated.

In addition, during that rapid sequence of events, neither pilot recognised the first officer’s high workload and task saturation. Consequently, no attempt was made to alleviate the situation, for example by discontinuing the approach or deferring ATC requests for information until the aircraft was straight and level. Such actions would have provided additional time and opportunity for the crew to refocus on flight instruments and key parameters.

Findings

From the evidence available, the following findings are made with respect to the stick shaker activation involving Saab 340B, VH-ZLJ, 30 km south-west of Perth Airport, Western Australia on 6 July 2021.

Contributing factors

• The aircraft’s right wing inboard de-ice boot probably delaminated shortly before encountering icing conditions, triggering the de-ice system fault that led to the flight crew’s decision to return to Perth.
• The pilot flying became task saturated due to high workload and did not notice the reducing airspeed, which was also missed by the pilot monitoring due to a focus on other tasks until the stick shaker activated.

Safety actions

Safety action by Regional Express

Regional Express has amended flight crew training simulator sessions and related training material to include flight at minimum manoeuvring speeds – minimum airspeeds that provide a margin above a stall during aircraft manoeuvring.

Note: Saab advised the ATSB that it intended to follow-up with Regional Express to obtain further details on the out of limits AOA vane and the delaminated de-ice boot.

Sources and submissions

Sources of information

The sources of information during the investigation included:

• the flight crew
• Regional Express
• Saab
• Airservices Australia
• Bureau of Meteorology.

References

ATSB aviation research investigation report AR-2012-172, Stall warning in high capacity aircraft: The Australian context, 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:

• Regional Express, including flight crew
• Saab
• Civil Aviation Safety Authority
• Airservices Australia.

Submissions were received from Regional Express, and the flight crew. 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 2022 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. Western Standard Time (WST): Coordinated Universal Time (UTC) + 8 hours.
2. Pilot Flying (PF) and Pilot Monitoring (PM): procedurally assigned roles with specifically assigned duties at specific stages of a flight. The PF does most of the flying, except in defined circumstances; such as planning for descent, approach and landing. The PM carries out support duties and monitors the PF’s actions and the aircraft’s flight path.
3. Having too many tasks to complete without enough time, tools, or resources to do them. This can lead to an inability to focus on what really matters. Task saturation can be insidious, and people can become too busy to recognise that they are overloaded.
4. Cloud cover: in aviation, cloud cover is reported using words that denote the extent of the cover – ‘broken’ indicates that more than half to almost all the sky is covered.
5. The flight data did not record stick shaker or aural warning activation. However, the airspeed increase, pitch angle reduction, and engine torque data was consistent with flight crew identifying the stall warning and initiating recovery procedures.
6. ATSB (Australian Transport Safety Bureau) (2013), AR-2012-172, Stall warnings in high capacity aircraft: The Australian context, Available from the ATSB website.

Safety summary

What happened

On 21 June 2021, a Boeing Company 787-9, registered VH-ZNH and operated by Qantas Airways, was prepared for a scheduled passenger flight from Sydney, New South Wales, to Perth, Western Australia. During initial climb, the flight crew selected the landing gear lever to UP. Shortly after, they received a warning, indicating that neither main landing gear had retracted to the ‘up and locked’ position. Despite consulting the aircraft’s electronic checklist, the flight crew were unable to resolve the retraction issue. The landing gear lever was then selected to DOWN, with positive gear extension indications, and the aircraft returned to Sydney for an uneventful landing.

What the ATSB found

The ATSB found that two of the five downlock pins, one in each main landing gear, had not been removed following towing of the aircraft to the domestic terminal aircraft bay. In addition, these gear pins were not identified during subsequent external inspections, prior to the departure. When the flight crew selected the landing gear to retract, the nose gear successfully retracted and locked in the up position. However, the two installed pins prevented any movement of the main landing gear. There was no damage to the main landing gear.

What has been done as a result

Following the occurrence, the operator distributed memos to engineering, flight and ramp crew, highlighting the quantity and location of the gear pins on the Boeing 787, and the importance of following the documented ramp, pre-flight and dispatch procedures.

The memo to engineering also emphasised the importance of checking the pin location, rather than relying on streamers for identification. In addition, training packages for engineering, ramp and flight crew were updated with additional detail. Further, the operator advised they were working to relocate the gear pin stowage to the flight deck, in line with other aircraft types, to enable ease of access to visually verify pin removal and stowage.

Safety message

‘Remove before flight’ streamers are a reminder to remove covers, or lockout devices, prior to flight. Failure to remove these devices and covers can prevent the functionality of certain aircraft systems. The streamers are subject to varying environmental conditions that can reduce their visibility.

Expectation can also affect identification of these warning devices. Put simply, the likelihood of detecting ‘remove before flight’ streamers is significantly reduced if they are not expected to be there. The same principle can also prevent the discovery of damaged and/or missing components.

The investigation

The occurrence

Pre-flight ground operations

On the morning of 21 June 2021, a Boeing Company 787-9 (787), registered VH-ZNH (ZNH) and operated by Qantas Airways, was being prepared for a scheduled passenger flight from Sydney, New South Wales, to Perth, Western Australia. At about 0745 Eastern Standard Time,^[1] ZNH was towed, by Qantas Engineering (engineering), from the aircraft parking location to domestic terminal bay 11. The tow crew on the ground consisted of the person in charge (PIC), two wing walkers and the tow‑motor driver.^[2] On board the aircraft was a 787 licenced aircraft maintenance engineer (LAME), in the role of flight deck operator (FDO)^[3] and an aircraft maintenance engineer (AME), who was undergoing informal general familiarisation training.

Upon arrival at bay 11, chocks were placed at the aircraft wheels and the tow‑motor was unhitched, to allow it to be used for another tow. The PIC removed the nose gear downlock pin (gear pin), then walked to the rear of the aircraft and removed one gear pin from the right main gear. At the same time, one of the wing walkers removed a gear pin from the left main gear.

The PIC and wing walker had not towed a 787 before and, as such, contacted the LAME via the aircraft intercom system, to enquire where the pins were stowed on this aircraft type. The AME, under instruction from the LAME, advised the PIC that the pins were stored in the electrical equipment centre (EEC) located just aft of the nose gear. The AME then relayed that the PIC could leave the removed pins on the nose gear and the LAME would stow them when the aerobridge^[4] arrived and they could exit the aircraft.

Despite the offer to leave the pins on the nose gear, the PIC identified a short ladder nearby and opened the EEC. The ladder was of insufficient height to see into the pin stowage location, so the PIC felt around and physically identified pin stowage holes. After stowing the three gear pins, the PIC closed the EEC, returned the ladder and departed, with the wing walker, to conduct another tow.

Unavailability of an aerobridge operator resulted in the LAME and AME being unable to leave the aircraft for about 20 minutes. When the aerobridge was manoeuvred into place, at about 0810, the LAME and AME exited the aircraft via the associated stairs. The LAME noted that the nose gear pin had been removed. The LAME reported that they then looked down toward the main gear and did not identify any streamers associated with the gear pins (see the section titled Landing gear downlock pins).

As there were no gear pins on the nose gear, the LAME directed the AME to confirm the gear pins had been stowed in the EEC, utilising the same nearby ladder previously used by the PIC. The ladder height was again insufficient to allow the AME to see into the pin stowage area, but they were able to physically feel the presence of the pins. After closing the EEC panel and returning the ladder, the LAME and AME waited at the aircraft until a car arrived at about 0819, to take them to the engineering office. The LAME signed for the removal and stowage of the landing gear pins on the electronic and paper technical logs (tech log). The LAME then returned the paper tech log to the aircraft at about 0842.

The flight crew, who arrived at the aircraft at about 0900, consisted of the captain, the first officer and a second captain, who was filling the role of relief pilot.^[5] The flight crew reviewed the tech log and noted the endorsement that the gear pins had been removed and stowed. As the third crew member, the relief pilot conducted the external inspection, between 0938 and 0945.

The relief pilot reported conducting the external inspection as per the flight crew operations manual (FCOM), with no anomalies identified, before returning to the flight deck to assist with the remainder of the pre‑flight preparations. The pre-flight, including landing gear system checks, were completed, with nil anomalies identified.

Aircraft dispatch for the 787 was conducted by two Swissport^[6] ground crew, a supervisor and a crew member undergoing training, who arrived at ZNH about 1020. Once all ground services had been completed, the supervisor conducted the final external inspection prior to pushback. The supervisor reported checking all doors and panels were closed and secured, and no streamers were identified. The aircraft was pushed back at about 1025.

Occurrence flight

ZNH was cleared to depart via runway 16R,^[7] initially heading to the south of the airport before making a westerly turn onto the planned route. The captain was the pilot flying (PF) and the first officer was the pilot monitoring (PM),^[8] with the second captain (relief pilot) seated in the flight deck for take-off.

At about 1032, shortly after take-off and with the aircraft established in a positive climb, the PM positioned the landing gear selector to UP. Ten seconds after selecting the landing gear up, the crew received a GEAR DISAGREE caution on the Engine Indication and Crew Alerting System (EICAS) and an associated aural alert. The flight crew observed indications that the nose landing gear had retracted however, both main landing gear continued to display a grey crosshatch symbol, indicating that they remained ‘in transit’ (see the section titled Landing gear operation).

The flight crew configured the aircraft for safe operation with the landing gear extended in accordance with the operator’s procedures. The PF then requested the PM commence the GEAR DISAGREE EICAS procedure through the aircraft’s electronic checklist. The flight crew advised air traffic control (ATC) of a landing gear problem and then levelled the aircraft at an altitude of about 9,000 ft, away from the airport and over water, to troubleshoot the issue.

The PF, who reported experiencing a degree of startle upon receiving the initial GEAR DISAGREE EICAS message, asked the PM to read out the electronic checklist procedure again, to confirm it was fully understood.

On completion of the GEAR DISAGREE checklist, the main landing gear continued to indicate ‘in transit’. The flight crew discussed cycling the landing gear, selecting DOWN and then UP, to troubleshoot the issue however, due their proximity to the airport, they decided to return to Sydney.

The flight crew reported that guidance for the continued management of the abnormal gear indication from the electronic checklist and Quick Reference Handbook was limited to noting airspeed and fuel duration considerations, and therefore decided to extend the gear via the normal gear selection. They further agreed to select landing gear down earlier in the approach than they normally would, to allow sufficient time to assess and resolve any abnormalities, noting the substantial fuel endurance available, if required. The landing gear lever was then selected to ‘DOWN’ and the crew received a positive ‘green’ indication on the EICAS that confirmed all of the landing gear was ‘down and locked’.

The flight crew reported that ATC asked if they wanted emergency services to be on standby for the landing however, they advised that this was not required. This decision was based on the aircraft’s landing gear now indicating normal extension. At 1106, ZNH touched down on runway 16R, for an uneventful landing. The aircraft was taxied to bay 11, and engineering subsequently identified that two landing gear pins were still installed, one in each main gear.

Context

Landing gear operation

The gear retraction sequence commences when the landing gear lever is placed in the UP position. The EICAS landing gear position indication changes from a green DOWN indication to a white crosshatch in-transit indication. When the landing gear has retracted, and is being held in place by uplocks, the landing gear hydraulic system is automatically depressurised. At this point, no landing gear indications are displayed on the EICAS.

The normal transit time for gear retraction is approximately 10 seconds. If any landing gear is not up and locked after about 40 seconds, the EICAS caution message GEAR DISAGREE is displayed. The EICAS gear position indication then displays an expanded non‑normal format. The flight crew can then see which gear is UP, in‑transit, or DOWN (if the gear never unlocked from the down position). In this instance, the flight crew were presented with an UP indication for the nose gear and in-transit indications for each of the main gear (Figure 1).

Figure 1: EICAS expanded message indication, showing nose gear up and locked, and main gear 'in-transit' (crosshatched)

Source: Supplied, annotated by ATSB

The EICAS checklist ‘objective’ for GEAR DISAGREE included for the flight crew ‘to extend the gear using an alternate gear extension’. The checklist also had a note to not exceed the ‘gear extended speed limit’ and that flight with the gear extended would increase fuel consumption. The flight crew reported that, as the nose gear retracted without issue, they did not suspect normal gear operation was affected. Therefore, they elected to attempt normal gear extension, before considering alternate procedures. The normal gear extension was completed without issue.

Landing gear downlock pins

The landing gear downlock pins (gear pins) were installed to prevent inadvertent gear retraction during maintenance or towing operations. The 787 has five gear pins, one for the nose gear and two for each main landing gear (Figure 2).

Figure 2: Typical left main gear showing pins and streamers installed.

Source: Supplied, annotated by ATSB

The gear pin is a quick release style and has a ‘REMOVE BEFORE FLIGHT’ streamer attached via a split ring. The aircraft maintenance manual (AMM) contained procedures for installation and removal of the gear pins. The main gear pins are located above head height and the AMM procedures recommend a 1.83m (6 ft) ladder, be utilised for installation and removal.

If the aircraft was required to be operated with the landing gear locked in the extended position, for example for maintenance purposes, the gear pins were to be secured as per the AMM. In that instance, a bolt and nut were to be installed on the end of the pin, with a large washer located between the bolt and the pin quick-release, to ensure retention of the pin in the landing gear.

The gear pin stowage box was located in the EEC, just aft of the nose gear, and was attached to the aircraft structure (Figure 3). The hinged door, which had receptacles for the four main gear pins, tilted outwards from the top and included a lanyard to limit swing. The box portion contained two receptacles, one for the nose gear pin and one for the steering bypass pin.^[9] Qantas engineering personnel reported that a ‘tall ladder’ was required to be able to see into the stowage box.

Figure 3: Gear pin stowage box showing hinged door with four main gear pin receptacles

Source: Supplied, annotated by ATSB

Airbus A330s and Boeing 737s were the aircraft types that Sydney Qantas engineering were most familiar with. These aircraft types have a total of 3 gear pins, one in the nose and one in each of the main gear. In addition, the gear pins on these aircraft types are stowed in the flight deck, readily accessible to flight and ground crew.

Gear pin streamers

Boeing advised that gear pins are classified as ground support equipment (GSE) and therefore not included in the type design of the aircraft. As a result, there is no minimum specifications, nor do they direct a maintenance or cleaning program.

Ground and flight crew described the streamers, of various aircraft types, as being different lengths and various states of cleanliness, which could reduce visibility in certain environmental and light conditions. Due to their installed location, the landing gear pin streamers were subject to contamination from oil, grease and grime. In addition, the streamers were known to wrap around the gear on occasion, particularly in wet and/or windy conditions.

An image of one of the missed main landing gear pins, upon return to Sydney, showed it to be dull and frayed however, it’s condition prior to the flight could not be determined (Figure 4). Examination of ZNH’s gear pin streamers noted degraded condition, particularly with respect to the normal intensity of the high-visibility colour.

Figure 4: Gear pin identified on landing (left) and VH-ZNH gear pins (right).

Source: Supplied, annotated by ATSB

CCTV footage showed that, when ZNH was towed into bay 11, prior to the flight, one streamer on each main landing gear was visible (Figure 5). It was these streamers that were identified and removed by the tow crew. The footage showed the wing walker and the PIC, climb the rear tyre of each main gear^[10] and remove a pin (with attached streamer) from the side braces. Streamers associated with the drag braces were not readily visible in the CCTV footage.

The footage also showed:

• evidence of recent rain when the aircraft was being towed to bay 11, but good light conditions
• some light rain at about the time the 787 LAME and AME, departed the flight deck, checked the EEC for the presence of pins and waited for a lift to the office
• the tow crew and AME used a 0.9 m (3 ft) 3-step ladder to access the EEC
• partly cloudy / sunny conditions when the relief pilot and dispatch crew conducted their respective external inspections, and during pushback.

Figure 5: VH-ZNH with visible main gear streamers

Source: Sydney Airport, annotated by ATSB

Procedures

Qantas engineering were responsible for aircraft towing operations, in accordance with the Towing section of the Qantas Engineering Procedures Manual. Section 7 Post aircraft tow, step 4 stated that it was either the tow crew PIC or the FDO’s responsibility to ‘remove and stow aircraft main and nose landing gear downlock pins’.

Swissport had been contracted by Qantas to carry out receipt and dispatch procedures for the 787, since its introduction in 2017. All other Qantas aircraft types had receipt and dispatch activities conducted by Qantas engineering at Sydney Airport. Qantas Ramp Operations Manual procedures were to be followed by Swissport staff, for receipt and dispatch of the 787. The aircraft dispatch procedure stated:

- Ensure the Steering By-pass pin is fitted

- Any landing gear downlock pins are also removed and stowed in the correct place

- Advise Engineering if Pitot covers are still present. Ensure they are removed prior to departure.

The section manage pins and covers identified:

There can be a total of five (5) downlock pins that can be fitted to the aircraft.

- One (1) nose gear downlock pin, and

- Up to Four (4) main gear downlock pins.

The ‘external inspection sequence’ included, ‘observe whether the main gear downlock pins have been removed’.

The flight crew were required to conduct their external inspection in accordance with the Qantas flight crew operations manual (FCOM) 787 Amplified procedures – exterior inspection. The section regarding the left and right main landing gear areas inspection included ‘gear pins – as needed’ (Figure 6)

Figure 6: FCOM external inspection of main gear areas

Source: Supplied

Awareness of five gear pins

Qantas ground crew

As a result of the COVID-19 pandemic travel restrictions, voluntary redundancies were offered to affected personnel. A subsequent restructure of Qantas engineering took place in February 2021, which resulted in certain engineers being transitioned into new roles, on new aircraft types.

Only one of the Qantas ground crew that spoke with the ATSB,^[11] the 787 LAME, advised they were aware that the 787 had five gear pins. In addition, it was the first time towing a 787 for all the ground crew except the 787 LAME. One of the wing walkers was a LAME on the 737, 747 and Airbus 330 (A330), while the PIC and the AME being trained by the 787 LAME, held category A licences^[12] on the A330.

Fight crew

All flight crew recalled undergoing computer-based training on external inspection procedures during their initial ground school (in 2018 and 2019), and then conducting an external inspection under supervision from a training captain, during initial flight‑line operations. Neither captain could specifically recall the number, nor exact location, of the main gear pins, from their training.

The first officer was aware of the five pins, due to their flying experience on the 767,^[13] which had a similar design of landing gear to the 787. The previous flying experience of the two captains was on aircraft types that had three gear pins. In addition, the second captain (relief pilot) and first officer advised they had never seen pins installed, as they were typically removed before the flight crew conduct their external inspection. Further, the flight crew reported their process was to look for the streamers during their external inspection, as opposed to sighting the actual pin locations.

Swissport dispatch crew

The Swissport crew consisted of a trainer and trainee. The trainer had experience dispatching 787 aircraft for several airlines, they advised that they were aware of the five gear pins on the 787 and had occasionally seen gear pins and streamers installed, for example, when aircraft were under tow.

The Swissport trainee, while experienced in dispatching other aircraft types at another Australian airport, was undergoing Sydney familiarisation training. The trainee advised that ZNH was only the second 787 they had dispatched, the first being a few hours earlier that day. The trainee recalled being aware of the five pins, from their training however, had never seen the pins installed. Further, as the trainer had conducted the external inspections that day, the trainee had not specifically looked out for them.

Both Swissport crew reported their dispatch procedures were to check all ground equipment was clear and inspect the aircraft for security of doors and panels, and presence of any streamers. The Swissport crew would install the steering bypass pin when the tow‑motor was connected, and then remove it when the tow‑motor was disconnected, following pushback. The steering bypass pin was located near the nose gear pin. The trainer advised that, had the nose gear pin still been installed, it would have been an alert to check all the other gear pins had been removed. However, as the nose gear pin was not installed, the trainer concluded that all pins had been removed, as they had not encountered a situation where only some of the pins had been removed.

Procedure if gear pin streamers identified

Both the flight crew and Swissport dispatchers reported that, if they identified any streamers during their external inspection, they were to contact Qantas engineering, who would then remove the associated pin or cover. Further, all reported being aware of the possibility that streamers can become caught up in, or stuck to, the landing gear.

Similar occurrences

Boeing advised they had received reports from other operators of inadvertent departures with gear pins installed. As the gear pins are classified as GSE, this type of event was not required to be reported to Boeing and subsequently that were not able to give an accurate estimate of how often situations like this may have occurred. Further, Boeing advised that, from the reports they did have, they were not aware of any outcomes more serious than a ‘return‑to‑base for a safe landing’.

The investigation identified two similar events involving Australian-operated 787s, in 2014 and 2021, noting this type of occurrence was not required to be reported to the ATSB. In 2014, the aircraft had been dispatched, and gear pin streamers were noted by the crew of another taxiing aircraft. The aircraft returned to bay and two gear pins were located, one in each main gear. In this occurrence, the gear pin streamers were noted to be ‘dirty’, short in length and ‘not as visible as new’ streamers.

On 19 June 2021, at about 2205 local time, two gear pin streamers were identified by the second officer during the flight crew external inspection. This was despite the tech log being endorsed that the pins had been removed and stowed. The second officer’s report stated that, when inspecting the right main gear from the front, nothing abnormal was noted. The second officer then moved to the rear of the main gear and used their torch to check the gear pin location holes. At this point they observed a gear pin streamer wrapped up and stuck to the main gear leg.

The second officer reported that, due to the wet and windy conditions, the streamer had been caught up on the main gear and, in combination with low light, it was difficult to see. Before returning to the flight deck to notify engineering, the second officer identified a gear pin streamer on the left main gear. The pins were subsequently removed by engineering and the aircraft departed as scheduled.

Safety analysis

In this occurrence, multiple factors led to an aircraft departing configured such that the flight crew were unable to retract the main landing gear. Specifically:

• The tow crew used the visible streamers to identify what they incorrectly believed were the only three gear pins installed the aircraft.
• While the LAME was aware the 787 had five gear pins, they did not confirm all of them had been removed and stored before signing the tech log. Instead, the AME physically checked for the presence of pins without knowing the number to expect.
• The ladder used to access the gear pin stowage location was not of sufficient height to allow visual confirmation of pin stowage. Sighting of ‘empty’ pin stowage receptacles would have provided a clear indication that not all pins had been removed from the landing gear.
• The flight and dispatch crews conducted their external inspection with no expectation of finding streamers indicating gear pins were still installed. This was likely due to the tech log being endorsed and a belief that the gear pins had been removed by engineering personnel.
• There was probably reduced visibility of the streamers, due to their degraded condition and the likelihood they were stuck on the gear, from a combination of grime and the recent wet and windy conditions.

Research has demonstrated that people are more likely to detect targets (such as gear pin streamers) when they are expected and less likely to detect targets that are not expected (Wickens and McCarley 2008). In addition, bias can occur when prior knowledge, combined with an expected outcome, influences decision making.

The tow crew were expecting to see the gear pin streamers, as they had just completed moving the aircraft. The tow crew then identified, and removed, three main landing gear pins, which was consistent with the aircraft types they had experience on. In contrast, as the gear pins were typically removed prior to the flight and dispatch crew external inspections, they were not expecting to see any gear pin streamers.

Findings

From the evidence available, the following finding is made with respect to the landing gear indication and return involving Boeing 787, VH-ZNH, near Sydney Airport, New South Wales, on 21 July 2021.

Contributing factor

• Two of the five landing gear pins were not removed as per the operator’s procedures, nor identified by engineering, flight crew or dispatch during pre-departure checks. This resulted in the aircraft departing without the functionality to retract the main landing gear.

Safety actions

Safety action by Qantas Airways Ltd

Following the occurrence, Qantas considered the viability of a maintenance program for ‘remove before flight’ streamers. However, it was determined that enhanced training and procedures would have greater benefit in reducing the risk of a similar occurrence. In response to an internal investigation, Qantas has advised the ATSB of the following actions:

• Release of memos to engineering, flight and ramp crew, highlighting quantity (five) and location (images) of gear pins on the 787, and the importance of following the documented procedures. The memo to engineering also emphasised the importance of checking the pin location, rather than relying on streamers. The memo to flight crew and ramp emphasised that engineering were to be contacted if any gear pins were identified, before continuing with the external inspection.
• Updated training packages for engineering, ramp and flight crew.
• Relocation of the pin stowage from the electrical equipment centre (EEC) to the flight deck, in line with other aircraft types, to enable ease of access to verify pin stowage.

Sources and submissions

Sources of information

The sources of information during the investigation included:

• Qantas engineering personnel involved in the tow
• Qantas flight crew
• Swissport dispatch crew
• Qantas Airways
• Boeing

References

Wickens, C.D. and McCarley, J.S (2008). Applied attention theory. Boca Raton, FL: CRC Press.

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:

• Qantas, including the involved engineering and flight crew
• Swissport, including dispatch crew
• Boeing and the United States National Transportation Safety Board.

Submissions were received from:

• Boeing
• Qantas

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 2021 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.  Eastern Standard Time (EST): Coordinated Universal Time (UTC) + 10 hours.
2. The PIC coordinates the tow with the ground crew via voice and hand signals, and with the flight deck via the aircraft intercom system. The wing walkers walk at each wing tip, monitor for clearance and can alert the PIC of impending danger, via hand signals or air horn.
3. The Flight Deck Operator (FDO), is a LAME who is licenced on the aircraft type and is required to ensure correct aircraft configuration throughout the tow duration.
4. Aerobridge – a moveable, elevated platform corridor connecting the aircraft to the terminal building.
5. The relief pilot, in this instance, was on board to allow the first officer a rest period during the flight. This was to ensure the first officer remained within flight and duty requirements for their return flight to Sydney later that day. The two captains were ending their duty in Perth.
6. Swissport provided ground services and cargo handling to Qantas, and other airlines, within Australia and globally.
7. Runway numbering: the number represents the magnetic heading closest to the runway (runway 16 at Sydney Airport is oriented 168° magnetic) and R indicates the right most of two parallel runways.
8. Pilot Flying (PF) and Pilot Monitoring (PM): procedurally assigned roles with specifically assigned duties at specific stages of a flight. The PF does most of the flying, except in defined circumstances; such as planning for descent, approach and landing. The PM carries out support duties and monitors the PF’s actions and the aircraft’s flight path.
9. When the steering bypass pin is inserted, the steering hydraulics on the landing gear are bypassed. This allows the aircraft to be moved by tow motor, without having to deactivate the entire aircraft hydraulics.
10. It was reported that ladders are typically not utilised for installation or removal of gear pins.
11. The ATSB did not interview the second wing walker or tow motor driver as they were not directly involved with the gear pins removal and/or stowage.
12. The category A licence gives the holder limited licensing privileges that are matched to the knowledge, competencies and assessments specified in the Civil Aviation Safety Regulations (CASR) Part 66 Manual of Standards for the category A licence.
13. The 767 is no longer operated by Qantas.

Executive summary

What happened

On 23 June 2021, a Cessna A150M Aerobat, registered VH-CYO, departed from the Sunshine Coast Airport, Queensland, with an instructor and student pilot on board. The purpose of the aerobatic training flight was to introduce and practice spin entry and recovery techniques.

The aircraft climbed to about 6,000 ft above mean sea level and arrived at the area intended to conduct aerobatics about 20 minutes after departure. Radar data showed that the aircraft then entered into a left spin that continued for about 55 seconds until the aircraft impacted terrain. The instructor and student were fatally injured, and the aircraft was destroyed.

What the ATSB found

Forward movement of the aircraft and the low angle of entry indicated that the aircraft was most likely in the process of recovering from the spin when it impacted with trees.

Examination of the aircraft did not identify any mechanical defect. However, the aircraft was significantly disrupted and therefore functionality of the flight controls was unable to be fully assessed. Pre- and post-accident medical information did not identify any underlying conditions in either pilot that may have contributed to the accident.

The aerobatics instructor was experienced in conducting spins, primarily in the Pitts Special aircraft type. However, it was likely that they had no experience in spinning a Cessna A150 Aerobat or any similar variant. The instructor’s theoretical spin training provided to the aerobatic student pilot (and another student at the same time) did not include instruction on the recovery technique as prescribed in the Aerobat pilot’s operating handbook (POH). Further, the ATSB established that it was likely the instructor intended to practice 2 spin recovery techniques. One of those techniques, broadly known as the Mueller/Beggs recovery method, has been shown to not recover a Cessna A150 Aerobat established in a spin to the left. The other method known as PARE, aligned closely with the aircraft’s POH and, if utilised, it would recover the aircraft from a spin.

The ATSB was unable to ascertain which of the recovery technique(s) was being utilised at the various stages of the spin recovery preceding the accident. For this reason, the ATSB was unable conclude if the use of an inappropriate recovery technique contributed to the accident.

What has been done as a result

The ATSB has issued a Safety Advisory Notice SAN (AO-2021-025-SAN-001) for aerobatic pilots and aerobatic instructors who conduct spins utilising the Mueller/Beggs spin recovery method, to raise awareness of its limitations.

Safety message

Although the reason for the accident could not be fully established, the investigation identified that one of the spin recovery methods that was to be practiced on the day of the accident would most likely not recover the Cessna A150M Aerobat from a spin.

This investigation presents a timely reminder that pilots should review the pilot’s operating handbook of the aircraft type that they intend to operate. Prior to intentionally spinning an aircraft, pilots should obtain instruction and/or advice in spins and recovery techniques from an instructor who is fully qualified and current in spinning that model. Further, aerobatic pilots and instructors should be aware and also teach the Meuller/Beggs method of spin recovery advantages, but most importantly its limitations in that it will not recover all aircraft types from a spin.

The occurrence

Aerobatics instructional flights

Two private pilots (students), who were members of the Sunshine Coast Aero Club, contracted an aerobatics instructor to provide aerobatic flight training in the aero club’s Cessna A150M Aerobat (Aerobat), registered VH-CYO. That training included theoretical and practical training aspects.

As the instructor did not work at the aero club, the aero club’s chief flying instructor (CFI) conducted a check flight with the instructor in the Aerobat to assess the instructor’s ability. The CFI was not rated in aerobatics, and the check flight was limited to an assessment of the instructor’s general handling and area knowledge. The CFI stated that the instructor performed the flight to a high standard and concluded that the instructor had the requisite skill and knowledge to conduct the flight training in the aero club’s Aerobat.

The students hired the aero club’s Aerobat for the practical flight training. The training was split into 2 days, commencing on the 16 June 2021. On that day 4 flights were undertaken, with 2 one-hour flights per student. The students undertook theoretical and practical instruction on:

• stall recovery techniques
• stall turns
• loops
• barrel rolls
• aileron rolls.

It was reported that, during the practical flight phase on that day, the instructor demonstrated each of the manoeuvres before handing control to the student.

Pre-flight briefing on the day of the accident

On the morning of 23 June 2021, the 2 students and the instructor continued the aerobatics training from Sunshine Coast Airport, Queensland, commencing with pre-flight theoretical instruction on spin training. The briefing contained information about:

• what is a spin^[1]
• inverted and upright spins
• what is a spiral dive^[2]
• difference between a spin and a spiral dive
• the Mueller/Beggs emergency spin recovery method
• the PARE method for spin recovery.

One of the students indicated that, during the pre-flight briefing, they were not instructed on what recovery method was recommended in the Aerobat Pilot’s Operating Handbook (POH), or that it closely aligned with the PARE method. Further, they were instructed on the advantages of the Mueller/Beggs method, but not on its limitations; namely, if the Mueller/Beggs method was utilised on an Aerobat, the aircraft would not recover from a spin to the left (see Aerodynamic spins).

Both students were instructed to write down the 2 spin recovery methods on a piece of paper for reference in flight when the practical component of the spin recovery was to be undertaken. One of the students indicated that they believed they were going to utilise both methods of spin recovery during their flight instruction. The first method written down on both students’ spin recovery notes was the Mueller/Beggs method.

Accident flight

At 1103 Eastern Standard Time,^[3] VH-CYO took off from the Sunshine Coast Airport, with the instructor and one of the aerobatic student pilots on board. The flight was being conducted under visual flight rules (VFR), and visual meteorological conditions existed during the flight. The accident flight was the first of 4 one-hour flights intended for that day (2 per student).

The aircraft departed to the south-west and climbed to about 6,000 ft above mean sea level (AMSL). Radar data showed that the aircraft arrived at the area intended to conduct aerobatics about 20 minutes after departure (Figure 1).

Figure 1: VH-CYO flight track radar data showing take-off point and accident site

Source: Google Earth, annotated by the ATSB

Figure 2 shows recorded radar data for the last 3 minutes of the flight. It indicated that, within the last 90 seconds, the aircraft conducted a 180° left turn, decelerated while maintaining altitude, and then descended rapidly, with the point of decent beginning at 5,800 ft above ground level (AGL). That manoeuvring was indicative of the planned entry into a spin.

At about 1122, 55 seconds after the initiation of the spin, the aircraft impacted terrain. The aircraft was destroyed, and the 2 occupants were fatally injured.

Figure 2: VH-CYO last 3 minutes of recorded flight data viewed from the left and above

Source: Google Earth, annotated by the ATSB

Figure 2 shows recorded radar data for the last 3 minutes of the flight. It indicated that, within the last 90 seconds, the aircraft conducted a 180° left turn, decelerated while maintaining altitude, and then descended rapidly, with the point of decent beginning at 5,800 ft above ground level (AGL). That manoeuvring was indicative of the planned entry into a spin.

At about 1122, 55 seconds after the initiation of the spin, the aircraft impacted terrain. The aircraft was destroyed, and the 2 occupants were fatally injured.

1.  Spin: a sustained spiral descent of a fixed-wing aircraft, with the wing’s angle of attack beyond the stall angle.
2.  Spiral dive: a steep descending turn with the aircraft in an excessively nose-down attitude and with the airspeed increasing rapidly.
3.  Eastern Standard Time (EST) was Coordinated Universal Time (UTC) + 10 hours.

Context

Pilot information

Instructor

Experience and qualifications

The instructor held a valid commercial pilot licence (aeroplane) that was issued on 10 August 2017. The licence included the following ratings and endorsements:

• single engine aeroplane class rating
• manual propeller pitch control, design feature endorsement
• aeroplane formation, spinning and aerobatics (low level) flight activity endorsement
• flight instructor rating with grade 2, spinning, formation (aeroplane) and aerobatics training endorsements
• aeroplane formation, spinning and aerobatics flight activity endorsements.

The instructor had their own aviation company that predominantly conducted aerobatic joy flights and instructional flights in the company’s 2 Pitts Special aircraft. The ATSB had access to the pilot’s logbook information up to 16 February 2020, at which point the instructor had accumulated 1,112.3 flight hours. The information in those logbooks indicated that the instructor had about 100 hours of flight experience in a Cessna 152 (a similar, non-aerobatic variant of the Aerobat), but none of that recorded experience was aerobatic in nature.

In the week prior to the accident, the instructor provided information to the Sunshine Coast Aero Club that they had about 100 hours experience in the Cessna 152. However, there was no mention of experience in the Cessna A150 Aerobat. The provided information was consistent with the information in the instructor’s logbook.

The ATSB was informed that, in recent times, the instructor had been utilising cloud-based pilot logbook software to record flight experience. The ATSB was unable to gain access to the cloud-based system.

The instructor’s initial and ongoing aerobatics training was conducted in the Pitts Special aircraft. Apart from the instruction flights in VH-CYO during the week prior to the accident, the ATSB was unable to identify any previous aerobatic experience in the Cessna A150 Aerobat or any other similar Cessna variants.

The person who conducted the aerobatics training to give the instructor a rating for aerobatics, and a rating to instruct in aerobatics, stated that they informed the instructor of the limitations in the Mueller/Beggs method during their initial aerobatics training. All of the practical flying training aspects were conducted in a Pitts Special.

Medical information and recent history

The instructor held a current class 1 medical certificate with no restrictions, and the accompanying medical records did not indicate any underlying medical issues at the time of the accident.

The instructor was reported to be fit and well on the day of the accident. There were no issues identified in the post-accident medical and toxicological results (including carbon monoxide) that may have affected the instructor’s operation of the aircraft.

Aerobatic student

Experience and qualifications

The aerobatic student pilot held a valid private pilot licence (aeroplane) that was issued on 13 June 2010. They also held a single-engine aeroplane class rating, and manual propeller pitch control and retractable undercarriage design feature endorsements. The student had a total of 248.5 flight hours experience.

The student had conducted an introductory aerobatic flight in an American Champion Aircraft Corp 8KCAB with an instructor in December 2014. That flight did not include spins.

Medical information and recent history

The aerobatic student pilot held a current class 2 medical certificate with no restrictions, and the accompanying medical records did not indicate any underlying medical issues at the time of the accident. The student was reported to be well rested and in good spirits on the morning of the accident. There were no issues identified in the post-accident medical and toxicological results (including carbon monoxide) that may have affected the student’s operation of the aircraft.

Operator information

The Sunshine Coast Aero Club was located at the Sunshine Coast Airport. At the time of the accident, the aero club had about 100 members and 3 aircraft: 2 Recreational Aviation Australia (RAAUS) registered Sling 2 aircraft and a Cessna A150 Aerobat (Aerobat), registered VH-CYO.

The Aerobat was recently purchased with the intent to conduct aerobatic instructional flights for its members. At the time of the accident, the aero club did not have a Civil Aviation Safety Regulation (CASR) Part 141 certificate to conduct flight training in a VH registered aircraft, nor was it required for an instructor to conduct spin or aerobatics flight training. There were no aerobatics-trained instructors at the aero club.

The aero club sought the assistance of a contracted aerobatics instructor to conduct the aerobatic flight training, utilising the instructor’s own flight training approval. The first aerobatic flight training conducted by the aero club utilising VH-CYO was the week prior to the accident with the flight instructor who was on board the accident flight.   

Aircraft information

General information

The Cessna 150 is a high-wing, 2-seat, single piston engine aeroplane designed for flight training. The Aerobat was a slightly modified model that was designed to conduct basic aerobatic training. The type of manoeuvres approved in the Aerobat Pilot’s Operating Handbook (POH) included spins.

VH-CYO

VH-CYO (Figure 3) was manufactured in 1976 and first registered in Australia in 1995. It had been owned and operated by the Sunshine Coast Aero Club since March 2021. However, it had not been utilised for aerobatics until the week prior to the accident.

Figure 3: VH-CYO Cessna A150M Aerobat

Source: Simon Coates

Maintenance information

The aircraft had a current certificate of airworthiness, certificate of registration, and maintenance release with no outstanding maintenance, or defects listed.

Subsequent to the accident, it was reported to the ATSB that the right-side radio push-to-talk switch had a defect which prevented radio calls from that position. Therefore, all calls had to be made from the headset and microphone plugged into the left-side jack point. It was also reported that the defect did not affect the intercom between pilots.

Rudder stop modification

The Cessna 150 and 152 series aircraft had a mandatory rudder stop modification identified as Single Engine Bulletin (SEB) 01-1. The Service bulletin was also mandated by Federal Aviation Administration Airworthiness Directive (AD) 2009-10-09 and therefore automatically mandated in Australia. The purpose as stated in the bulletin was as follows:

To provide an enhanced rudder stop, bumper, doubler and attachment hardware designed to assist in preventing the possibility of the rudder overriding the stop bolt during full left and/or right operation of the rudder.   

 VH-CYO had the rudder stop modification incorporated at the time of the accident.

Weight and balance

The aircraft’s published maximum take-off weight (MTOW) according to the Pilot Operating Handbook was 727.3 kg (1,600 lb). The aircraft’s weight for the accident flight was estimated to be about 14.3 kg over the MTOW on departure and about 7.1 kg overweight at the time of the accident.

Taking into consideration the aircraft’s calculated weights at take-off, and at the time of the accident, a centre of gravity (CG) calculation could not be carried out, as the aircraft’s weight was outside that of the published CG calculation limits (Figure 4). As the aircraft was over the MTOW, the data was extrapolated outside of the chart limits to get an estimate of the CG location at the time of the accident. The extrapolated value placed the CG aft of mid-range, but well forward of the aft limit.

Figure 4: Estimated weight and balance

Source: Cessna, annotated by the ATSB

Recorded information

The aircraft flight path was derived from primary^[4] and secondary^[5] surveillance radar data recorded by Airservices Australia. The data included the aircraft’s position with a time stamp and altitude above mean sea level (AMSL) at 5-second intervals.

Figure 5 shows the last 90 seconds of flight with the spin entry beginning about 55 seconds before impact with terrain. The decent rate varied between data points, with an average descent rate of about 5,000 ft/min. The radar returns stopped at about 1,200 ft above mean sea level (AMSL), which was 800 ft above ground level (AGL) at the accident site. That was most likely due to the aircraft descending below radar coverage.

The last 2 recorded data points without pins were considered to be predictive and not an accurate representation of the aircraft position.

Figure 5: Radar data with timestamp, airspeed, altitude and vertical decent rate labelled

Source: Google Earth, annotated by the ATSB

Site and wreckage examination

The accident site was located in a dense stand of trees that stood about 15–20 m high and straddled a creek line in a band about 50 m wide, with open areas of farmland on either side (Figure 6).

Figure 6: Area of accident site

Source: Google Earth, annotated by the ATSB

The wreckage trail extended about 50 m from the initial tree impact point, until the final piece of wreckage, oriented in an east-west direction. There were several notable tree impact points, including trees that had been broken in half or completely felled by the impact forces. Calculations of the tree impact damage heights indicated the final flight path angle was a descent of 12.8° (Figure 7).

Figure 7: Final flight path angle of entry

Source: ATSB

The main wreckage came to rest at the base of a tree that was struck at a height of about 10 m. The aircraft structure was significantly disrupted as a result of impacting several trees (Figure 8).

Figure 8: Aircraft main wreckage at the base of a large tree that was struck

Source: ATSB

The ATSB conducted an examination of the aircraft wreckage. The examination identified that:

• the disruption to the aircraft and foliage, coupled with the length of the wreckage trail, indicated that the aircraft had significant forward speed at impact
• the flaps were in the retracted position
• the aircraft had no evident pre-impact defects with the flight controls or aircraft structure
• the aircraft was intact prior to impact with terrain
• the engine had no obvious defects upon external examination, was free to rotate and had compression on all 4 cylinders
• the throttle setting was captured at an idle position (full out and bent to one side) during the accident sequence
• the propeller rotational damage signatures were minimal, indicating a low power setting
• both seats were in the full aft position.

Survivability aspects

General information

During the accident sequence, the cockpit area was completely disrupted due to significant impacts with trees, leading to the occupants’ liveable space being compromised. For this reason, it was considered unlikely that the accident was a survivable event.  

Flight notes and tracking overdue arrivals

The CASR Part 91 Manual of Standards (MOS) stated that for some types of visual flight rules (VFR) flights a pilot was required to submit a flight plan, nominate a SARTIME for arrival, or leave a flight note with a responsible person. These included air transport flights, a flight over water, a flight in a designated remote area, or a flight at night proceeding beyond 120 NM from the departure aerodrome. In other cases, a pilot could elect to submit a flight plan, nominate a SARTIME or leave a flight note.

If a flight note was left with a responsible person, then that person had to be over 18 years old, have access to at least 2 operative telephones, and satisfy the pilot that they know how to contact the Joint Rescue Coordination Centre (JRCC) and will do so immediately in the event that the pilot’s flight was overdue.

In summary, a flight note was not formally required for flights similar to that conducted in VH-CYO on the day of the accident. However, flight notes or another method of identifying if an aircraft is overdue is highly recommended.

The ATSB was informed that the Sunshine Coast Aero Club had a method of tracking estimated arrival times. That method involved instructors informing the aero club administration of estimated arrival times and aircraft movements. The ATSB noted that the instructor of VH-CYO did not inform the aero club administration about the estimated time of the aircraft’s return. It was considered likely that the instructor, being a contractor who had not worked with the aero club before, was not informed or aware that it was the aero club procedure to do so.

The aircraft accident occurred at 1122. It was scheduled to return to refuel at about 1200, and it was reported missing at about 1515 by the aero club chief pilot when the second student raised concerns about the aircraft not returning from its flight.

The post-mortem reports for the pilots indicated that the occupants’ chances of survival would not have improved if the location of the wreckage was identified sooner.

Emergency and personal locator beacons

The aircraft was not fitted with a fixed emergency locator transmitter (ELT), nor was it required to be under the current regulations. 

ATSB research into the effectiveness of ELT’s in aviation accidents (AR-2012-128) stated that:

Data from the ATSB database show that ELTs function as intended in about 40 to 60 per cent of accidents in which their activation was expected. Records of the Australian Maritime Safety Authority’s SAR incidents shows that search and rescue personnel were alerted to aviation emergencies in a variety ways including radio calls and phone calls, and that ELT activation accounted for the first notification in only about 15 per cent of incidents. However, these ELT activations have been directly responsible for saving an average of four lives per year.

A personal locator beacon (PLB) was identified on the accident site in an area that was away from and not likely to be located by the occupants of the aircraft (if they had survived the impact). The beacon was in date and passed a self-function test to indicate that it was serviceable.

The ATSB research report also mentioned PLBs with the following suggestion:

…carrying a personal locator beacon (PLB) in place of or as well as a fixed ELT will most likely only be beneficial to safety if it is carried on the person, rather than being fixed or stowed elsewhere in the aircraft.

Aerodynamic spins

General description

An aerodynamic spin is a sustained spiral descent in which an aircraft’s wings are in a stalled condition, with one wing producing more lift than the other. This difference in lift sustains the rotation and keeps the aircraft in the spin. The nose angle can also vary considerably. In a fully developed, upright, left spin, an aircraft will simultaneously roll to the left while yawing to the left, making a vertical corkscrew path through the air. A spinning aircraft will descend more slowly than one in a vertical or spiral dive and it will also have a lower airspeed, which may oscillate.

Intentional spins are normally entered from a stall in straight and level flight, with the reduction in power, the application of full back elevator and full rudder in the intended direction of rotation at the moment of stall.

When entering a spin, an aircraft’s motion through the air is irregular at first. This is known as the incipient phase of the spin. Though the nature of the incipient spin is heavily dependent on the aircraft type and the manner of entry, recovery may be more rapid and require less control input in this stage compared with recovery from a developed spin. After a number of rotations and depending on the aircraft type, loading, and control inputs, an aircraft in an incipient spin may then settle into a regular rotating descent, known as a developed spin. A spin may steepen (nose-down) or flatten (nose more horizontal) as it continues, potentially requiring different recovery techniques. Figure 9 shows the various stages from spin entry until recovery.

Figure 9: Various stages of a spin and recovery

Source: New Zealand Civil Aviation Authority, Spin avoidance and recovery

Recommended practices in preparation for spin

The CASA Flight Instructor Manual: Aeroplane stipulated:

The aeroplane must be clear of inhabited areas and normally in an area designated for the practice of such exercises. In addition, it should be at a height sufficient to ensure recovery by 3,000FT above ground level. The pre-spinning check will vary from aeroplane to aeroplane but will normally be similar to that used as a pre-stalling check in that particular aeroplane. In most aeroplanes flaps and undercarriage must be retracted during both the spin and spiral. … In all cases a 360° turn to ensure that all is clear around and below should be carried out immediately prior to commencing each exercise.

The information stipulated in the flight instructor’s manual was commonly referred to by the acronym HASELL, which is:

• Height – sufficient to recover by 3,000 ft AGL
• Airframe – wheels up / flaps up / CG ok / trim set
• Security – seat belt tight / no loose items in aircraft or pockets
• Engine – temperature and pressure / carburettor heat / mixture / fuel quantity and selection
• Location – aerobatic area / no built-up area or public gathering within 600 m / forced landing fields available
• Lookout – 360° turn or wingover.

Spin recovery techniques

Aircraft manufacturer spin recovery information

There was information provided by the aircraft manufacturer on spin recovery in 2 sections of the Cessna A150M Aerobat POH. This information was the same in each section, and the spin recovery technique stated that:

Should an inadvertent spin occur, the following procedure should be used:

1). Retard throttle to idle position.

2). Place ailerons in neutral position.

3). Apply and hold full rudder opposite to the direction of rotation.

4). Just after the rudder reaches the stop, move the control wheel briskly forward far enough to break the stall. Full down elevator may be required at aft centre of gravity loadings to assure optimum recovery.

5). Hold these control inputs until rotation stops.

6). As the rotation stops, neutralise rudder and make a smooth recovery from the resulting dive.

It also stated that:

Variations in basic airplane rigging or in weight and balance due to installed equipment or cockpit occupancy can cause differences in behaviour, particularly in extended spins. These differences are normal and will result in variations in the spin characteristics and in recovery lengths for spins of more than 3 turns. However, the above recovery procedure should always be used and will result in the most expeditious spin recovery.

Cessna also provided further information in a document tilted Spin Characteristics of Cessna Models 150, A150, 152, A152, 172, R172 and 177. Apart from reiterating the recovery procedure provided in the POH, it also stated information including:

Basic Guidelines for Intentional Spins

1). Know your aircraft thoroughly.

2). Prior to doing spins in any model aircraft, obtain thorough instruction in spins from an instructor fully qualified and current in spinning that model.

PARE spin recovery method

The PARE spin recovery method is generic and typical of most light single engine aircraft types. PARE is an acronym that stands for:

• Power, idle
• Ailerons, neutral (and flaps up)
• Rudder, full opposite to the spin direction and held in that position
• Elevator, forward

Hold these inputs until rotation stops, then:

• Rudder, neutral
• Elevator, easy pull to straight and level or climbing attitude.

A comparison between the Cessna A150 Aerobat recovery method and the PARE method indicated there was little difference between the 2 methods, with the exception that the Cessna method emphasised the use of the term ‘briskly’ in regards to the forward movement of the elevators, and that full forward elevator may be required.

Mueller/Beggs (emergency) spin recovery technique

The Mueller/Beggs recovery technique, sometimes referred to as the emergency spin recovery technique, was documented in an aerobatic article written by an aerobatic pilot, Eric Mueller, in the 1980s. The article stated that it was a technique designed to recover a Pitts Special aerobatic aircraft from an upright or inverted spin, even if the pilot was disorientated. The technique is as follows:

1. Power off

2. Remove your hands from the stick

3. Apply full opposite rudder

4. Neutralise the rudder and recover to level flight.

Another aerobatic pilot, Gene Beggs, popularised the recovery technique in a series of articles. Beggs’ reference manual titled Spins in the Pitts Special stated:

With this method you can quickly and easily recover from any spin in the Pitts Special. It is easy to remember and execute even if you are frightened or confused; furthermore, it is not necessary to know whether the spin is upright or inverted, the recovery is the same in either case.

In the Beggs course notes for advance spin recovery, the frequently asked questions section stated:

The question I hear most is “Will the emergency spin recovery work on all aircraft?” No, not exactly! Although I have found it works beautifully in the vast majority of cases, there are rare exceptions. You may occasionally encounter a spin mode in some aircraft in which you must physically apply nose-down elevator. This is extremely rare, and I assure you it will never happen in a Pitts Special.

A newsletter titled Spinoffs written by Beggs in 1985 indicated that the author was informed by another pilot that a Cessna A150 Aerobat would not recover using the emergency recovery method (Meuller/Beggs method). Beggs decided to conduct some spin testing in a standard Cessna 150. The following is a summary of that testing:

• The aircraft would recover using the emergency recovery technique (hands off) in fully developed spins to the right.
• The aircraft would not recover using the emergency recovery technique in fully-developed spins to the left, no matter how many turns the aircraft was allowed to do.
• If the elevator was pushed forward briskly during the emergency recovery technique to the left, the aircraft would always recover promptly in one additional turn with pitch attitude almost perfectly vertically down.
• In spins both to the right and left, the use of opposite aileron (out spin aileron) would produce a recovery from the spin. This was opposite to the results obtained in all other aircraft types (that had been previously spun by Beggs).
• In the Cessna 150, the use of in-spin aileron always increased the rate of rotation and steepened the pitch attitude. This was also opposite to the results obtained in all other aircraft types.

Beggs stated having conducted thousands of emergency spin recoveries in numerous aircraft types. The Cessna 150 was one of the very few aircraft that required the application of full forward elevator to recover.

Regulatory requirements and guidance

The CASR Part 61 MOS, Volume 2, Section 6, Unit FAE-8 – Spinning, described the skills and knowledge required to execute and recover from an upright spin.

Unit FAE-8, element 4, titled underpinning knowledge, stated that the following items were required to be imparted to students:

…

(o) standard spin entry and recovery techniques for the aircraft being flown;

(p) number of turns normally required for spin recovery in the aeroplane type;

…

(r) Mueller-Beggs spin recovery action and limitations on its application

(s) ‘g’ and any other limitations applicable to spinning for the aeroplane type.

The Civil Aviation Aeronautical Publication (CAAP) 155-1(0) Aerobatics was issued in 2007. In relation to spin recovery, it stated:

Modern aerobatic aircraft designs normally have predictable spin characteristics and respond to the standard spin recovery technique. However, older aircraft and non-certificated or amateur built aircraft may have special characteristics which require particular recovery procedures. Therefore, pilots need to be familiar with, and practised in, the spin recovery procedure specified for the particular aircraft type.

It also stated:

Spin recovery procedures will vary between aircraft types and situations. The aircraft flight manual should be the final authority for spin recovery procedure…

The CAAP also discussed the Mueller/Beggs spin recovery method. The ATSB requested CASA’s interpretation on the Mueller/Beggs spin recovery limitations referenced in the MOS. It stated:

Civil Aviation Advisory Publication (CAAP) 155-1(0) – Aerobatics, published January 2007 provides guidance to pilots on aerobatics operations. Section 7 - Risk management and TEM includes subsection 7.24 Mueller-Beggs Spin Recovery, which describes the Mueller-Beggs recovery technique and associated limitations. There is also reference to the techniques in the underpinning knowledge sections of the Units of competency in Appendix A of the CAAP from which the MOS references were drawn.

As stated in 7.24.1, the main limitation, as is that it is known, is the technique is not effective in a number of aircraft types. 7.24.4 advises pilots to determine the extent to which the technique has been tested and found to be reliable in a particular aircraft type. It also states pilots wishing to test the procedure should also be familiar with the normal spin recovery procedure specified for the type. While 7.24.5 states the technique is not recommended, it may prove to be useful in the event a pilot becomes disoriented.

Based on the above information CASA’s opinion of the limitations of the Mueller-Beggs technique are:

1. Application of the technique may not be effective for the aircraft in which the training is conducted,

2.Use of the technique will likely delay the recovery from the spin and consequently increase the height lost, perhaps to a point recovery cannot be achieved,

3. The technique might be in conflict with the aircraft manufacturer’s recommended technique.

Reference to the technique is included in the CAAP to make pilots aware of its existence as an alternative recovery technique. In the event they become disoriented from high rates of rotation, which can be encountered in an upright or inverted spin, the technique might effect a recovery if other recovery techniques applied are unsuccessful.

Associated with a range of regulatory changes in December 2021, the CAAP was removed from the CASA website in January 2022.^[6]

In April 2020, CASA issued Advisory Circular AC 61-16 v1.0 (Spin avoidance and stall recovery training). In addition to a variety of other guidance, it stated:

Before selecting an aircraft for stalling or spinning training, consult with the manufacturer and other users to establish what manoeuvres are safe to conduct, including steep turns, stalls, stalls with a wing drop and spinning.

It also stated that, prior to spinning any aircraft, pilots should:

- Comply with aircraft flight manual weight and balance and manoeuvre limitations, placards and, if provided, procedures and advice for each intended manoeuvre…

- Obtain thorough instruction in spins from an instructor fully qualified and current in spinning that model…

- Enter each spin at a high altitude. Plan recoveries to be completed well above the minimum legal altitude…

- Conduct all spin entries and recoveries in accordance with the procedures recommended by the manufacturer...

In the guidance for instructors, it stated:

- Ensure the aircraft is operated in accordance with the aircraft flight manual limitations and entry and recovery procedures for manoeuvres including stalling and spinning…

- Recognise and avoid the potential for negative training with a clear understanding of what the desired training outcome is for the lesson. The latent effects of negative training can stay with a pilot throughout their career…

The effects of centre of gravity on spins

The CASA Flight Instructor Manual: Aeroplane included a section for spins and how it is affected by the CG. It stated:

The effect of the position of the Centre of Gravity (CG) must be pointed out to the student if movement of this position within the limits laid down has a great effect on the spinning characteristics of the aeroplane. Normally a forward CG results in a steeper spin with a high rate of descent. A forward CG makes recovery much easier and may even prevent a spin altogether, resulting in a spiral dive. An aft CG tends to flatten the attitude resulting in a lower rate of descent. The recovery action to be taken when an aeroplane is spinning in a flat attitude is the same as the normal recovery technique with respect to the actual control movements. However, in the flat spin case it is essential to ensure that full control movement is applied in the recovery action and that this is maintained if necessary, for a much longer period than normal. In some aeroplanes it takes many turns to recover from a flat spin.

Related occurrences

Cessna A150 Aerobat (VH-CYO), Cairns, Australia, December 1995

The ATSB received a report from a previous pilot of VH-CYO about an incident that occurred in the aircraft involving a flat spin. The incident occurred near Cairns Airport in December 1995. A summary of that event was as follows:

• On the day of the incident, the aircraft (VH-CYO) was being operated as an aerobatic aircraft with a student and instructor on board. The purpose of the day’s instructional flights was stalls, spins and spin recoveries.  
• On the day of the training sequence, air traffic control (ATC) clearance was obtained to operate between 5,000 ft and 3,000 ft AMSL.   
• The spin training exercise commenced at 5,000 ft and consisted of showing recovery, student follow through, and finally student completing the entry and recovery. 
• As a final exercise, the student was instructed to commence the spin at about 5,000 ft, to allow the ‘spin’ to fully develop and recover from the spin when instructed.   
• The instruction to recover was given at about 4,300 ft and the student was observed to apply the correct Cessna A150 POH recovery technique. However, the aircraft failed to recover from the spin. The instructor took control of the aircraft and applied the POH spin recovery method, but the aircraft failed to recover from the spin. 
• As the aircraft descended towards the cleared level of 3,000 ft, the instructor believed the aircraft would not respond to the POH recovery method and may had entered a flat spin. The instructor attempted to force the nose down by commencing a backwards and forwards full deflection of the elevator motion. That action did not assist, so the instructor coordinated full throttle acceleration to elevator deflection with the thought that it might assist in getting a nose-down attitude. 
• The instructor regained some control of the aircraft as it passed 1,000 ft, with the aircraft exiting the spin and entering a spiral dive. At approximately 700 ft, recovery from the resulting dive was completed, ATC was advised that the aircraft had flown below the minimum specified altitude, and a clearance was obtained to return to Cairns Airport. 

A subsequent engineering inspection, which included a check of the aircraft rigging, did not identify any defects.   

The instructor of the 1995 flight advised the ATSB that, after some consideration, they believed that the issue was most likely one of a rear centre of gravity in the loading of the aircraft. The instructor stated that they were 182 cm and about 85 kg, with the student being at least 188 cm and about 90–95 kg.  Both seat positions were adjusted to the rear stop and the fuel load was from memory sufficient for about 3.0 hours total. 

The instructor of the 1995 flight stated that they had spoken to 2 other pilots, who had detailed that, while conducting spinning together in another C150, they had experienced difficulty in exiting a planned spin, and their experience seemed to have been very similar to what the instructor encountered. 

Cessna 152 accident, Concord, United States, 29 January 2018

A Cessna 152 aircraft, registered N93316, lost control and impacted terrain, fatally injuring the pilot. A subsequent inspection of the aircraft identified that one of the rudder cables had failed and the other had frayed to a point where about 50% of the strands had fractured.^[7]

4.  Primary radar returns are produced by radar transmissions that are passively reflected from an aircraft and received by the radar antenna. The received signal is relatively weak and provides only position information, not the aircraft’s altitude.
5.  Secondary radar returns are dependent on a transponder in the aircraft replying to an interrogation from a ground station. An aircraft with its transponder operating is more easily and reliably detected by radar and, depending on the mode selected by the pilot, the aircraft’s pressure altitude is also displayed to the air traffic controller.
6.  CASA advised that the CAAP was intended to be replaced by AC 61-18 Aerobatics.
7.  National Transportation Safety Board investigation WPR18FA075

Safety analysis

Introduction

Radar data indicated that, while being used to conduct spin training, the Cessna A150 Aerobat (VH-CYO) entered a spin at 5,800 ft above ground level and the spin was not fully recovered before the aircraft impacted terrain. Site and wreckage examination indicated that the aircraft had significant forward velocity, a low angle of entry, and the throttle was captured in the idle position. Those items of evidence indicated that the aircraft was most likely in the initial stages of recovery from the spin when the aircraft impacted terrain.

In previous training with the student on board, the instructor had demonstrated each manoeuvre before handing control to the student. The accident occurred during the first manoeuvre of the training session, and the ATSB was unable to ascertain which of the 2 pilots (instructor or student) was controlling the aircraft at various stages of the spin and for the initiation of the recovery.

This analysis discusses several possible reasons for the aircraft not being fully recovered from a spin before impacting terrain. These include:

• mechanical failure
• flight control obstruction
• aft centre of gravity and flat spin
• pilot incapacitation
• interference with the controls
• incorrect recovery technique.

Potential scenarios to explain absence of recovery from spin

Mechanical failure

A failure of the aircraft structure, the flight control system, or a rudder locking past the rudder stops have contributed to aircraft accidents in the past. However, examination of the aircraft structure and flight controls of the aircraft did not reveal any pre-impact defects. The aircraft also had a modification incorporated to prevent the rudder-stop locking issue that had contributed to some previous Cessna 150 accidents.

Further, there was evidence that the aircraft was in the early stages of recovery from the spin, which indicated that whatever had delayed the recovery had been overcome prior to impacting terrain.

Overall, it was considered unlikely that some type of mechanical failure of the flight controls contributed to the accident. However, due to the disruption and displacement of the wreckage, the ATSB was unable to completely rule out the possibility of a mechanical issue.

Flight control obstruction

Aircraft accidents have previously occurred where foreign object obstruction has led to flight controls becoming locked, preventing the pilots from controlling their aircraft. If an object had locked the controls of VH-CYO, the initial stages of the recovery evident before impact with terrain would indicate that the controls became unlocked, or more controllable, during the final stages of the descent.

The examination of the aircraft did not reveal any issues in relation to flight control locking due to foreign object fouling. However, due to the disruption and displacement of the wreckage, the ATSB was unable to rule out the possibility of a flight control obstruction, but it was considered to be unlikely.  

Aft centre of gravity and flat spin

The further aft the aircraft’s centre of gravity is, the more difficult it may be to lower the nose in order to recover from a spin. In this case, the aircraft was slightly over the maximum allowable take-off weight (MTOW) for the entire flight. Regarding the aircraft loading and centre of gravity (CG), and after interpolating the data (as the aircraft was outside its weight limit), it was considered to be within the desired CG range, trending towards aft of nominal.

It is possible that the spin entry or recovery actions created a flat spin, where the nose was comparatively high compared to a normal spin (with the nose slightly down). This can be exacerbated by an aft CG and can make the aircraft slower to respond to recovery techniques. Previous incidents in the same aircraft type have shown that flat spins can be very difficult to recover, even when the appropriate recovery technique is applied for an extended period.

In summary, it is possible that the aircraft entered a flat spin that was unable to be fully recovered, and that the aft CG may have exacerbated the difficulty in recovering from the spin. However, there was insufficient evidence to conclude that this occurred.

Pilot incapacitation

Both of the pilots were reported to be well at the time of the accident, and the pre and post-accident medical information did not identify any conditions or issues with either pilot that may have contributed to the accident. Also, as previously noted, the aircraft was most likely in the initial stages of recovery from the spin when the aircraft impacted terrain. Accordingly, the ATSB considered it unlikely that pilot incapacitation contributed to the accident.

Interference with the controls

The Cessna A150 Aerobat is a dual control aircraft. If an inexperienced pilot were to ‘freeze’ at the controls or make other inappropriate flight control inputs, it may be difficult for the instructor to regain control of the aircraft.

As previously noted, it was not possible to ascertain which of the 2 pilots (instructor or student) was controlling the aircraft at various stages of the spin and for the initiation of the recovery. In addition, during the previous week, the student conducted steep turns, stall recovery, loops, and barrel and aileron rolls. The student had also done a small amount of aerobatics several years before the accident and had a reasonable amount of flight experience. Overall, none of the available evidence indicated that the student was susceptible to freezing at the controls or making other inappropriate flight control inputs.

Incorrect recovery technique

The instructor owned, was trained on, and had significant aerobatic experience in the Pitts Special aerobatic aircraft. However, the ATSB could not identify any aerobatic experience for the instructor in the Cessna A150 Aerobat or similar variants, apart from the previous week’s instructional activities with the same students. That training did not include spin entry and recovery techniques.

The aircraft manufacturer’s guidance document on spin characteristics stipulated that, if the instructor was unfamiliar with the aircraft type’s spin characteristics, then they should obtain thorough instruction in spins from an instructor qualified and current in spinning that particular model aircraft. The Civil Aviation Safety Regulation (CASR) Part 61 Manual of Standards (MOS) stated that underpinning knowledge for spin training included the standard spin entry and recovery techniques for the aircraft being flown. Other CASA guidance highlighted the importance of being familiar with the spin recovery method specific to the aircraft type. However, the ATSB could not identify if the instructor had sought additional information about the Aerobat’s spin characteristics. It is possible that, due to the instructor’s general familiarity and experience on a similar, non-aerobatic variant (the Cessna 152), they did not consider that recovery techniques successfully utilised on other aircraft types would not work equally as effectively on the Cessna A150 Aerobat.

The theoretical spin recovery training conducted by the instructor on the morning of the accident included 2 recovery methods; namely the PARE and Mueller/Beggs methods. The PARE method was closely aligned with (but not exactly the same as) the method described in the Cessna A150 Aerobat Pilot’s Operating Handbook (POH) and the Cessna guidance document. The Mueller/Beggs method has proven to be a very effective method of spin recovery in most aircraft types, though there are a few aircraft types that will not recover using this method. The Cessna A150 Aerobat and similar variants are aircraft types that most likely will not recover from a spin to the left, as was the case in this accident.

The Part 61 MOS stated that the instructor should teach the students the method of recovery in the aircraft type that they will be operating in. It also stated that the limitations of the Mueller/ Beggs method should also be discussed. However, according to the second student who received the theoretical instruction, the instructor did not highlight to the student’s what recovery method was recommended in the POH, or that the Aerobat would not recover utilising the Mueller/Beggs technique.

Further, the instructor informed the students to write down both methods of recovery (Mueller/Beggs and PARE) on a piece of paper for reference during the flight. The second student was of the firm belief that they would be conducting both methods of spin recovery in the Aerobat, with the first method written down being the Mueller/Beggs method.

The ATSB considered it likely that the instructor was not aware or did not recall that the Aerobat would not recover utilising the Mueller/Beggs method in a spin to the left. Further, the evidence indicates that the instructor intended to utilise both methods of recovery in 2 separate spin sequences on the accident flight.

If the Mueller/Beggs method was being used for the first exercise, it would provide a viable explanation of the accident sequence. However, based on the available evidence, the ATSB was unable to establish if the Mueller/Beggs method was being utilised at the time of the accident, or if it contributed to the delayed recovery time.

Survival aspects

Management of overdue aircraft

Due to the nature of the impact, the accident was not survivable. However, the investigation noted that there were potential areas for improvement that could be relevant in other situations.

The Sunshine Coast Aero Club had a common practice for flight instructors to log an estimated arrival/return time with the aero club’s administrator prior to departing for a flight. However, that procedure was not utilised on the day of the accident. The ATSB considered it likely that the contracted flight instructor was not informed of the procedure and therefore did not inform the administrator of their estimated time for return. As a consequence, the aero club did not discover that the aircraft was overdue for some time, and subsequently reported the aircraft missing about 3 hours after it was due to return.

Although there is no regulatory requirement to do so for many types of flights under the visual flight rules, a standardised method to identify if an aircraft is missing would decrease the amount of time for the Joint Rescue Coordination Centre to be notified and for a subsequent search for the aircraft to commence.

Emergency locator transmitter and portable locator beacons

The aircraft was not fitted with a fixed emergency locator transmitter (ELT). Although a fixed ELT is not a regulatory requirement, they are an effective safety feature that have been shown to significantly reduce the amount of time between an aircraft accident and identifying the location of the aircraft by search and rescue.

A portable locator beacon (PLB) was identified in an area away from the aircraft occupants that would not likely have been located without an extensive search. Carrying a PLB would be much more beneficial to safety if it is carried on the person, rather than being fixed or stowed elsewhere in the aircraft.

Findings

From the evidence available, the following findings are made with respect to the collision with terrain accident involving Cessna A150 Aerobat registered VH-CYO on 23 June 2021.

Contributing factors

• While the student and instructor were conducting aerobatic spin training starting at 5,800 ft above ground level, for reasons that could not be established, the aircraft did not fully recover from a spin before impact with terrain.

Other factors that increased risk

• It is likely that the aerobatics instructor had no flight experience conducting spinning and/or spin instruction in the Cessna A150 Aerobat or similar variants. It was also considered probable that they did not seek advice from an experienced aerobatic instructor on the A150 type about the aircraft’s spin characteristics.
• The spin training theory provided by the instructor to the 2 aerobatics students was generic in nature and did not highlight the limitations of the Mueller/Beggs spin recovery technique or provide guidance on the method recommended in the Cessna A150 Pilot’s Operation Handbook, as stipulated in the Civil Aviation Safety Authority aerobatic instruction procedures and guidance material.
• It was likely that the aerobatics instructor intended to practice the Mueller/Beggs method of spin recovery during the accident flight in the Cessna A150 and was likely unaware that the aircraft type was one of the few types that would not recover from a spin to the left utilising that technique.
• On the day of the accident the aircraft operator was not utilising a flight following procedure to identify if an aircraft was overdue, nor were they required to under the current regulations. Therefore, the overdue aircraft was not identified and reported as missing for 3 hours after it was due to return.

Other findings

• The aircraft structure and flight controls were examined, and no pre-impact defects were identified.
• It could not be determined which pilot was controlling the aircraft during the various stages of the accident flight and spin recovery.
• The aircraft was not fitted with a fixed emergency locator transmitter, nor was one required by the regulations. Fixed emergency locator transmitters have been shown to be an effective safety feature to reduce the amount of time taken to identify an aircraft’s location, even if the occupants are incapacitated.

Safety actions

Safety Advisory Notice

Safety advisory notice to aerobatic pilots and instructors

The ATSB strongly encourages all aerobatic pilots and aerobatics flight instructors to be aware:

• the Mueller/Beggs method of spin recovery does not recover all aircraft types from a spin
• the Mueller/Beggs spin recovery method limitations should be emphasised during spin theory training
• the Mueller/Beggs method of spin recovery will not recover a Cessna A150 Aerobat or similar variants from a spin in some circumstances
• they should review the pilot’s operating handbook of the aircraft type that they intend to operate for the recommended spin recovery technique
• prior to doing spins in any model aircraft, they should obtain instruction and/or advice in spins from an instructor who is fully qualified and current in spinning that model.

Glossary

Sources and submissions

Sources of information

The sources of information during the investigation included the:

• Sunshine Coast Aero Club
• student pilot who underwent theoretical and practical training with the instructor
• the instructor’s aerobatics instructor
• Civil Aviation Safety Authority
• Queensland Police Service
• aerobatic subject matter experts
• Airservices Australia.

References

ATSB Research Investigation AR-2012-128, The effectiveness of emergency locator transmitters in aviation accidents.

New Zealand Civil Aviation Authority (2014) Spin avoidance and recovery.

Experimental Aircraft Association Inc. (1985) ‘Spinoffs-Gene Beggs’, International Aerobatic Club Sport Aerobatics Magazine.

Beggs G (2001) Aerobatics with Beggs: Spins in the Pitts Special (A guide and reference manual for aerobatic instructors and students).

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:

• the Sunshine Coast Aero Club
• the student pilot
• the instructor’s aerobatic instructor
• the Civil Aviation Safety Authority (CASA)
• the aircraft manufacturer.

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

Purpose of safety investigations & publishing information

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 2022 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 morning of 22 June 2021, the pilot of a Piper PA-32-300 Cherokee Six, registered VH‑CWK, prepared for a private flight at Moorabbin Airport. The pilot was the only occupant and intended to conduct several circuits during the flight to maintain experience in the aircraft type.

Between 1129 and 1200 Eastern Standard Time, the pilot conducted an engine run up and completed 3 circuits using runway 35 right without any incident or abnormal engine indications.

At 1200, as the aircraft climbed through 505 ft above ground level to complete a fourth circuit fuel flow to the engine significantly reduced and within 12 seconds, the engine lost power.

Soon after the engine power loss, the aircraft commenced descending. The pilot identified one of few clear areas available for a forced landing and manoeuvred the aircraft towards that area. As the aircraft approached the selected landing area, the pilot recognised that insufficient height remained to clear trees along the southern edge of the selected landing area. The aircraft impacted the trees before colliding with rising ground about 35 m beyond the trees. The aircraft was destroyed, and the pilot sustained serious injuries.

What the ATSB found

The ATSB found that as the aircraft climbed upwind from the departure runway, fuel flow reduced and the engine lost power. Despite extensive examination, the reason for the fuel flow reduction could not be determined.

It was also identified that the incorrect engine variant for the aircraft serial number was inadvertently fitted to the aircraft. This did not affect the operation of the aircraft or contribute to the power loss.

Safety message

This accident highlights the challenges pilots face when confronted with a loss of engine power at low level and with few suitable forced landing areas within the glide capability of the aircraft.

Pilots can best mitigate the effects of a power loss by forward planning, which reduces your mental workload under stress, and always maintaining control of your aircraft. Maintain glide speed and use no more than a moderate bank angle to avoid entering a stall and/or spin. During the forced landing, aim to arrive at the ground with wings level and the aircraft level with the ground, which improves your prospects of survivability. The ATSB research publication Managing partial power loss after take-off in single-engine aircraft provides further guidance, which is also applicable to a complete power loss.

The investigation also found that while in this case the fitment of an incorrect engine variant did not contribute to the accident, it underlines the importance of the correct interpretation of the manufacturer’s type certificate documentation. If the guidance is unclear, a maintainer should contact the Civil Aviation Safety Authority or the original equipment manufacturer for clarity on permitted action.

The investigation

The occurrence

On the morning of 22 June 2021, the pilot of a Piper PA-32-300 Cherokee Six, registered VH‑CWK, prepared for a private flight at Moorabbin Airport. The pilot was the only occupant and intended to conduct several circuits to maintain experience in the aircraft type.

Between 1129 and 1200 Eastern Standard Time (EST),^[1] the pilot conducted an engine run up and completed 3 circuits using runway 35 right without any incident or abnormal engine indications.

At 1200, the aircraft climbed through 505 ft above ground level (AGL) to complete a fourth circuit. At that time, as was the pilot’s usual practice, engine RPM was reduced from about 2,700 to the desired climb power of 2,500 RPM and the throttle was left at a full open setting. At the same time as the RPM reduced, fuel flow to the engine significantly reduced and within 12 seconds, the engine lost power. Soon after the power loss, the aircraft commenced descending and the pilot made a MAYDAY^[2] broadcast stating ‘…I’ve got a power loss’.

As the descent continued, the pilot targeted a glide speed of 90 kt and searched for a landing site within a 30° arc either side of the aircraft’s nose (Figure 1). The pilot identified one of few clear areas available for a forced landing within the gliding capability of the aircraft and manoeuvred the aircraft for a forced landing in that area (see the section titled Aircraft details).

As the aircraft approached the selected landing area, the pilot recognised that the aircraft was close to stalling and that insufficient height remained to clear trees along the southern edge of the selected landing area. The left wing subsequently impacted the trees about 14 m AGL and separated from the aircraft. The aircraft then rolled left and pitched nose down before impacting rising ground about 35 m beyond the trees. The aircraft was destroyed and the pilot sustained serious injuries, including to their head, which significantly impaired the pilot’s recollection of the flight.

Figure 1: Overview of the flightpath after the power loss

Source: Google Earth and OzRunways, annotated by ATSB

Note: Area ‘A’ was an excavated quarry. Area ‘B’ had been developed after the satellite imagery was captured and before the accident.

Context

Aircraft details

The Piper PA-32-300 Cherokee Six is a low‑wing, piston-engine aircraft with a two-blade variable‑pitch propeller, fixed tricycle landing gear and could be fitted with up to 7 seats. VH-CWK, serial number 32-7840052 (Figure 2), was manufactured in the United States in 1977 and first registered in Australia in 1978.

Prior to the accident flight, the aircraft had not flown for 4 weeks. On the morning of the accident, the aircraft was fitted with two overhauled magnetos, a test run of the engine was carried out and the aircraft was released for service.

Figure 2: VH-CWK

Source: David Carter

The ATSB estimated the weight of the aircraft at the time of the accident as 1,281 kg (1,542 kg maximum allowable). The information manual for the aircraft provided best range glide speeds of 87 kt for an aircraft weight of 1,542 kg and 78 kt for 1,315 kg. The information manual indicated that the maximum glide range from 505 ft AGL was 0.66 NM (no wind).^[3]

The aircraft was fitted with a main and a wingtip fuel tank in each wing. The combined capacity of the 4 tanks was 318 litres (315 litres usable). The aircraft was reported to have departed with full fuel in the right wingtip and two main tanks. The left wingtip tank had been used for the engine run earlier in the day and was reported to be close to full. The pilot recalled intending to use the right main fuel tank for the entire flight. This tank contained sufficient fuel for the planned flight.

Engine

The PA-32-300 was equipped with a fuel-injected Lycoming IO-540 series engine of 300 horsepower. The aircraft type certificate data sheet (TCDS) stated that the aircraft type could be fitted with either an IO-540-K1A5 or an IO-540-K1G5 engine variant.

During its operating life, VH-CWK had been fitted with both K1G5 (factory installed) and K1A5 variants. Including K1A5 variants fitted by several maintenance organisations. In July 2020, VH‑CWK was fitted with an IO-540-K1A5 that had been rebuilt following a propeller strike in another aircraft. Prior to the accident flight, this engine had completed about 55 hours since being fitted to VH-CWK, and a total of 1,413.3 hours in service.

The difference between the K1A5 and K1G5 engine variants was the fuel pump type. The K1A5 variant was fitted with a vane type fuel pump, the K1G5 was fitted with a diaphragm type. The K1G5 variant was also 1 pound (0.454 kg) lighter than the K1A5.^[4]Both variants operated in the same fuel pressure range and no fuel system or other airframe differences applied to the fitment of either variant.

The TCDS did not specify a specific serial number range for the fitment of a K1A5 engine variant, but did so for the K1G5 variant as follows:

PA-32-300

Engine

- Lycoming IO-540-K1A5, Bendix injector type RSA-10ED1

- Lycoming IO-540-K1G5 (See NOTE 12)

NOTE 12:

- Lycoming engine Model IO-540-K1G5 with Hartzell propeller HC-C2YK-1(F), Blade Model 8475D-4, S/N 32-7640066 (only) and S/N 32-7640072 through 32-7940290.

When reviewing the TCDS, the maintainer responsible for installation of the engine in July 2020 interpreted the allowable engine variants as follows:

…the K1A5 was an allowed engine as there was no notes or restrictions…while noting if using a K1G5 that could only be installed in certain serial range.

As a consequence of that interpretation, a K1A5 variant was fitted to the aircraft. Past fitment of this variant by several maintenance organisations indicated a similar interpretation of the TCDS.

However, the manufacturer stated the following as the correct interpretation of the TCDS:

The type design specifies the engine model IO-540-K1G5 for Piper Model PA-32-300 serial numbers 32-7640066, 32-7640072 and up (ending 32-7940290). All other PA-32-300 aircraft serial numbers had engine model IO-540-K1A5 defined.

This interpretation is summarised in Table 1.

Table 1: PA-32-300 engine variant serial number applicability

The Civil Aviation Safety Authority approved aircraft flight manual, the information manual and the illustrated parts catalogue also referred to a K1G5 variant as the appropriate engine for installation in VH-CWK.

Meteorological information

A meteorological report for Moorabbin Airport, recorded at 1200, included a northerly wind of 9 kt, visibility greater than 10 km, no cloud, and a temperature of 14 °C.

Wreckage information

The aircraft impacted the tree about 14 m (Figure 4) above its base with the wing flaps retracted.

The left wing contacted the tree first and separated from the aircraft. Ground impact scars were located on an upslope 35 m from the tree base, 10.2 m below the height of the tree impact. After ground impact, the right wing also separated from the fuselage. The engine and fuselage were the last items in the wreckage trail, coming to rest 55 m from the tree.

Figure 3: Accident site viewed in the direction of flight

Source: ATSB

Note: The red objects were tarpaulins provided by first responders to the accident.

Fuel system

First responders to the accident drained a significant amount of fuel from the right main fuel tank.

While examination of the fuel system was limited by accident damage, the on-site examination identified that the:

• left main fuel tank, along with both tip tanks, were breached during the accident and empty
• left main and both tip tank caps were fitted and secure
• right main tank was not breached
• right main tank was selector lever was in the right tank position and the selector valve inlet port was connected to the supply line from the right main tank.
• right main tank cap had been removed by first responders

In addition:

• all fuel tank vents were clear of obstructions
• the electric fuel pump switch was selected on and functioned correctly
• no pre-existing defects were found with the supply lines and hoses
• the in-cabin fuel sump drain lever was secured behind a fastened panel

The fuel selector valve and fuel filter were retained for further examination. Fuel collected from the right main tank and fuel selector was tested and found to be of the correct type and free of contaminants.

Engine controls and propeller

The engine struck the ground on its left forward cylinder area, separated from its mount and rotated 90° around the aircraft centreline. The throttle and mixture cables fractured and separated during the impact. The throttle lever was found in the full position and the propeller control was in 

the maximum RPM position. The mixture lever was positioned at about half of its range of travel. However, it could not be determined if the levers had moved from pre-impact positions as a result of impact forces or during efforts to extract the pilot.

The bending damage to the propeller blades indicated that little or no power was being delivered to the propeller at the time of impact.

The engine and associated accessories were retained for further examination.

Recorded information

The aircraft was fitted with a J.P. Instruments EDM-830 Engine Data Management system. Data recovered from this unit captured data points every 6 seconds and included the power loss (Figure 5).

The first abnormal indication was a reduction in fuel flow from 113 litres per hour (l/h) to 47 l/h at 1200:49 (A). At the same time, RPM reduced from about 2,700 to about 2,500 (B) and exhaust gas temperature (EGT) increased across all cylinders (C). The next data point at 1200:55 showed the fuel flow reduced further to 22 l/h (D) and all EGT values, along with RPM, reduced significantly as power was lost (E). Following the power loss, fuel flow varied between 20 l/h and 16 l/h in conjunction with changes in RPM from the windmilling propeller.

No abnormal cylinder head temperature (CHT), manifold pressure (MAP) or oil pressure (Oil Px) indications were recorded before or after the power loss.

Figure 4: Graphical representation of data recovered from VH-CWK

Source: ATSB

Airservices provided recorded surveillance data of the flight. The aircraft was recorded commencing descent at 1200:53 at a groundspeed of 80 kt (headwind component was at least 9 kt). During the descent the groundspeed reduced further to between 68‑60 kt.

Engine, fuel selector and filter examination

In December 2021,^[5] the engine was disassembled and examined at a CASA-approved 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.

The ATSB examined the control positions required to replicate the recorded RPM and fuel flow indications. It was determined that to reproduce a fuel flow of 20 l/h with the throttle fully open, the mixture control needed to be moved close to the cut-off position. The propeller control movement needed to reduce RPM from 2,700 to 2,500 was significantly less. However, it was noted that control rigging variations could produce slightly different control movement ranges.

The ignition system components and fuel system components, including both fuel pumps, were tested. No defect was identified with these components that could have contributed to the power loss.

The fuel selector was examined at the ATSB’s technical facilities in Canberra, and it was found to function as designed.

Safety analysis

Engine power loss and collision with terrain

During the fourth circuit in a session of circuits, as the aircraft climbed through about 505 ft above ground level, the engine lost power. Following the power loss, the pilot established a glide at a speed below that targeted, but close to the correct speed for the aircraft weight. This led to near optimum glide performance. However, few suitable landing areas were available within the glide capability of the aircraft. The pilot selected an open area, but the aircraft did not clear trees along the southern boundary of the area. The aircraft impacted these trees separating the left wing before impacting the ground.

The on-site examination of the wreckage and later engine examinations and analysis of the recorded data found no indications of an induction issue, including icing, an ignition system or fuel system problem.

The recorded data showed that a significant reduction in fuel flow led to the power loss. Notably, after the power loss, the fuel flow did not reduce to zero, but varied between 20 and 16 l/h (in conjunction with RPM changes) indicating that fuel starvation did not lead to the power loss. Further, wreckage examination, flight records and witness reports indicated that there was sufficient fuel on board, and in any tank, to power the engine.

The ATSB considered the possibility that the pilot inadvertently reduced the mixture control instead of, or at the same time, as manipulating the propeller control. However, the pilot’s head injuries greatly limited their recollection of actions taken before and after the power loss. The recorded data showed that RPM reduced at the time, and by the amount, that the pilot intended to reduce RPM. Therefore, it is almost certain that the pilot manipulated the propeller control as intended.

At the same time, the fuel flow reduced significantly from 113 l/h to 22 l/h. The mixture lever movement required to produce this fuel flow reduction was significantly larger than that associated with the intended RPM change and required the lever to be positioned near the cut-off stop. After the accident, the mixture lever was found to be at a mid-travel position, but as the impact had caused significant damage to the engine controls, including breaking the mixture control actuating arm, the pre-impact position could not be determined. Overall, there was insufficient evidence to determine if the mixture control was inadvertently moved.

While damage limited the examination of the fuel system, no defect was identified that could have led to the fuel flow reduction. The engine was dismantled, examined and all fuel system components were tested. These components functioned correctly when examined. In summary, a reason for the fuel flow reduction could not be determined.

Incorrect engine variant

The type certificate data sheet did not specify a specific serial number range for the fitment of a K1A5 engine variant but did so for the K1G5 variant. This led the maintainer to interpret the aircraft type certificate data sheet as permitting the fitment of an K1A5 to VH-CWK instead of the correct K1G5. Past fitment of the K1A5 variant by other maintenance organisations indicated they had similarly assessed the TCDS as permitting this combination of engine variant and aircraft serial number.

However, the aircraft flight manual and the illustrated parts catalogue stated the only applicable engine variant for VH-CWK to be a K1G5, which was also confirmed by the aircraft manufacturer.

The differences between the fitted variant and the correct variant were minor and did not affect the operation of the aircraft, nor did they contribute to the power loss and collision with terrain.

Findings

From the evidence available, the following findings are made with respect to the collision with terrain involving a Piper PA-32-300 aircraft, VH-CWK, near Moorabbin Airport, Victoria on 22 June 2021.

Contributing factors

• As the aircraft climbed through 505 ft above ground level, fuel flow reduced and the engine lost power. During the subsequent forced landing the aircraft collided with terrain. The reason for the fuel flow reduction could not be determined.

Additional finding

• The incorrect engine variant for the aircraft serial number was inadvertently fitted to the aircraft. This did not affect the operation of the aircraft or contribute to the power loss.

Sources and submissions

Sources of information

The sources of information during the investigation included:

• the pilot
• the aircraft owner and maintainer
• Piper Aircraft (aircraft manufacturer)
• Lycoming Engines (engine manufacturer)
• engine data monitor
• Airservices Australia
• OzRunways
• Bureau of Meteorology

References

Australian Transport Safety Bureau, Avoidable Accidents No. 3 - Managing partial power loss after takeoff in single-engine aircraft

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:

• the pilot
• the aircraft owner and maintainer
• Piper Aircraft
• Lycoming Engines

Submissions were received from:

• the pilot
• the aircraft owner and maintainer
• Piper Aircraft

The submissions were reviewed and, where considered appropriate, the text of the report was be 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 2022 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. Eastern Standard Time (EST): Universal Coordinated Time (UTC) + 10 hours.
2. MAYDAY: an internationally recognised radio call announcing a distress condition where an aircraft or its occupants are being threatened by serious and/or imminent danger and the flight crew require immediate assistance.
3. The glide range capability of an aircraft reduces with an increase in headwind.
4. VH-CWK was reweighed after the fitment of the K1A5 variant of engine.
5. Examination of the engine was delayed due to Coronavirus related travel and working restrictions. During the period between the accident and examination, the engine was inhibited and stored in a secure location.